177 results on '"Artificial networks"'
Search Results
2. Presenting the Model of the Effect of the Degree of Group Similarity on the Quality of Accounting Information and the Quality of Financial Reporting on the Valuation of the Initial Public Offering of Shares Using the Algorithm of Artificial Neural Networks.
- Author
-
Hooshyar, Mohammadreza, Darabi, Roya, and Fallah, Mirfeiz
- Abstract
Objective. The pricing and valuation of initial public offerings (IPOs) are the subject of extensive literature in finance, primarily focusing on well-documented pricing anomalies such as IPO underpricing and significantly abnormal post-issuance returns. The use of peer company information to value IPOs has received attention due to their signaling role. Among the things that can be compared in the field of peer companies is the quality of their accounting information and financial reporting. Due to the lack of accounting information related to IPOs, practitioners rely significantly on comparable Peer Accounting Information when attempting IPO Valuation. Firms desiring to obtain capital through an IPO depend on more accounting information than their own when transitioning from a relatively opaque to a relatively transparent information environment, such as the accounting information of their priced peers. According to these cases, this research aims to present the model of the effect of the degree of group similarity, the quality of accounting information and the quality of financial reporting on the valuation of the initial public offering of shares using the algorithm of artificial neural networks. Method: This research is practical in terms of purpose and data collection method; it is ex-post facto research in accounting proof research. The statistical population of this research includes 92 companies listed on the Tehran Stock Exchange between 2012 and 2020, which uses the multiple regression method to test the hypotheses. The degree of similarity of the quality of accounting information of the supplier company with the peer group has been calculated and tested separately using three indicators of abnormal accruals, profit predictability and profit stability. Also, considering that artificial network patterns can be used to predict the valuation of an initial public offering of stocks and probably have different powers, this research compares the power of three algorithms (firefly algorithm, machine regression algorithm, decision, and tree algorithm). Results: The comparable companies' approach to pricing IPOs depends largely on the availability of accounting information from peer companies already priced in the market. However, to be most effective, peer accounting information should be useful in making decisions about how to use the accounting information of peer companies. The results showed that the valuation of shares in an initial public offering is possible based on the degree of similarity of the companies in the group and the quality of the accounting information. Stock valuation in initial public offerings based on profit and sales approaches differs in terms of the similarity of group companies and the quality of the company's accounting information. Group companies and the quality of accounting information aren't associated with incorrect pricing (overvaluation and undervaluation) in initial public offerings. The results of neural networks also indicated that the firefly algorithm has a higher power to predict the initial public offering valuation than the machine regression algorithm; the firefly algorithm had a higher power to predict the initial public offering valuation than the tree algorithm. The decision tree algorithm is more likely to predict the valuation of the initial public offering of stocks than the support vector machine regression algorithm. Conclusion: in the comparison between valuation theory and practice (use of information of peer companies), the second view is more effective, and the accuracy of information of peer companies is a key component of pre-publication estimates of the accuracy of information of IPO companies, and the information set of an IPO, it is not limited to its own accounting information, but also includes the information of its peers. Given that companies often decide to offer their initial shares when they think they can maximize their equity earnings (such as when they feel their stock is overvalued or during periods of booming stock markets and heightened investor sentiment), the results of this research can be used by potential investors to reduce the effect of underwriters' incorrect valuation and to monitor them, especially in the emotional stock market; Therefore, it is suggested that buyers of shares of companies that are listed on the stock market for the first time should consider the accounting information and quality of financial reporting of similar and peer companies, especially the quality of accruals and the predictability and stability of their profits; because it contains information about the future price estimation of initial supply companies. Underwriters and initial price estimators are also suggested to consider the quality of accounting and financial reporting of their peers in estimating the initial price of initial public offering companies. In the end, due to the higher power of the Firefly algorithm to predict the valuation of the initial public offering of stocks compared to the vector machine regression algorithm and the decision tree, it is suggested to use these algorithms for valuation due to the lower error. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
3. Comparing discriminating abilities of evaluation metrics in link prediction
- Author
-
Xinshan Jiao, Shuyan Wan, Qian Liu, Yilin Bi, Yan-Li Lee, En Xu, Dong Hao, and Tao Zhou
- Subjects
link prediction ,evaluation metrics ,discriminating abilities ,artificial networks ,Science ,Physics ,QC1-999 - Abstract
Link prediction aims to predict the potential existence of links between two unconnected nodes within a network based on the known topological characteristics. Evaluation metrics are used to assess the effectiveness of algorithms in link prediction. The discriminating ability of these evaluation metrics is vitally important for accurately evaluating link prediction algorithms. In this study, we propose an artificial network model, based on which one can adjust a single parameter to monotonically and continuously turn the prediction accuracy of the specifically designed link prediction algorithm. Building upon this foundation, we show a framework to depict the effectiveness of evaluating metrics by focusing on their discriminating ability. Specifically, a quantitative comparison in the abilities of correctly discerning varying prediction accuracies was conducted encompassing nine evaluation metrics: Precision, Recall, F1-Measure, Matthews correlation coefficient, balanced precision, the area under the receiver operating characteristic curve (AUC), the area under the precision-recall curve (AUPR), normalized discounted cumulative gain (NDCG), and the area under the magnified receiver operating characteristic. The results indicate that the discriminating abilities of the three metrics, AUC, AUPR, and NDCG, are significantly higher than those of other metrics.
- Published
- 2024
- Full Text
- View/download PDF
4. Remarks on the Mathematical Modeling of Gene and Neuronal Networks by Ordinary Differential Equations
- Author
-
Diana Ogorelova and Felix Sadyrbaev
- Subjects
neuronal networks ,dynamical systems ,artificial networks ,critical points ,attractors ,Mathematics ,QA1-939 - Abstract
In the theory of gene networks, the mathematical apparatus that uses dynamical systems is fruitfully used. The same is true for the theory of neural networks. In both cases, the purpose of the simulation is to study the properties of phase space, as well as the types and the properties of attractors. The paper compares both models, notes their similarities and considers a number of illustrative examples. A local analysis is carried out in the vicinity of critical points and the necessary formulas are derived.
- Published
- 2024
- Full Text
- View/download PDF
5. Dynamical Models of Interrelation in a Class of Artificial Networks
- Author
-
Sadyrbaev, Felix, Atslega, Svetlana, Brokan, Eduard, Pinelas, Sandra, editor, Graef, John R., editor, Hilger, Stefan, editor, Kloeden, Peter, editor, and Schinas, Christos, editor
- Published
- 2020
- Full Text
- View/download PDF
6. Machinability study and ANN-MOALO-based multi-response optimization during Eco-Friendly machining of EN-GJL-250 cast iron.
- Author
-
Laouissi, Aissa, Nouioua, Mourad, Yallese, Mohamed Athmane, Abderazek, Hammoudi, Maouche, Hichem, and Bouhalais, Mohamed Lamine
- Subjects
- *
IRON founding , *MACHINING , *SURFACE roughness , *ANT lions , *JOB performance , *PRODUCTION control , *CAST-iron , *SAND casting - Abstract
The current work examines the performance of Minimum Quantity Lubrication when turning of EN-GJL-250 cast iron compared to dry and wet cooling methods. The Taguchi design L36 has been chosen for the planification of experimentation. Then, ANOVA has been established after data acquisition in order to define the effect of cutting conditions such as the used inserts, cutting depth, feed rate and cutting speed on the studied factors. Furthermore, the surface roughness has been deeply studied using 3D roughness topography to evaluate the MQL effect. Finally, the approach ANN-MOALO was found to be helpful for future industrial applications for predicting part quality and power consumption with accurate results and optimizing cutting parameters that helps to achieve the best production control. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
7. Bio-instantiated recurrent neural networks: Integrating neurobiology-based network topology in artificial networks.
- Author
-
Goulas, Alexandros, Damicelli, Fabrizio, and Hilgetag, Claus C.
- Subjects
- *
ARTIFICIAL neural networks , *RECURRENT neural networks , *NEURAL circuitry , *TOPOLOGY , *BIOLOGICAL networks - Abstract
Biological neuronal networks (BNNs) are a source of inspiration and analogy making for researchers that focus on artificial neuronal networks (ANNs). Moreover, neuroscientists increasingly use ANNs as a model for the brain. Despite certain similarities between these two types of networks, important differences can be discerned. First, biological neural networks are sculpted by evolution and the constraints that it entails, whereas artificial neural networks are engineered to solve particular tasks. Second, the network topology of these systems, apart from some analogies that can be drawn, exhibits pronounced differences. Here, we examine strategies to construct recurrent neural networks (RNNs) that instantiate the network topology of brains of different species. We refer to such RNNs as bio-instantiated. We investigate the performance of bio-instantiated RNNs in terms of: (i) the prediction performance itself, that is, the capacity of the network to minimize the cost function at hand in test data, and (ii) speed of training, that is, how fast during training the network reaches its optimal performance. We examine bio-instantiated RNNs in working memory tasks where task-relevant information must be tracked as a sequence of events unfolds in time. We highlight the strategies that can be used to construct RNNs with the network topology found in BNNs, without sacrificing performance. Despite that we observe no enhancement of performance when compared to randomly wired RNNs, our approach demonstrates how empirical neural network data can be used for constructing RNNs, thus, facilitating further experimentation with biologically realistic network topologies, in contexts where such aspect is desired. • Constructing artificial neural networks with network topology of animal brains. • Network topology as a structural prior effecting performance of neural systems. • Framework for building neurobiologically realistic brain models. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
8. Lengthening of average path length in social networks due to the effect of community structure
- Author
-
Amrit Lal Sangal, Harsh Kumar Verma, and Himansu Sekhar Pattanayak
- Subjects
Random graph ,General Computer Science ,Social network ,business.industry ,Computer science ,Artificial networks ,Community structure ,020206 networking & telecommunications ,02 engineering and technology ,Graph model ,Average path length ,Fire spread ,Shortest path problem ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,business ,Algorithm - Abstract
Community structure is a common phenomenon observed in various social networks. In this work, a novel community detection algorithm is proposed by estimating the effect of community structure on the average path length of a network. The Erdos-Renyi graph model is used as a reference to compute the change in the average path length of a network due to community formation. By experimenting with artificial networks, it is found that community structure in a social network contributes towards the lengthening of the average shortest path length. A random graph is found to have a lesser average shortest path length than a social network with community structure. As the size of individual communities increases, there is a decrease in the difference of average shortest path lengths, compared with a random graph containing an equal number of nodes and edges. This relationship is used to predict the average community size and their numbers in a network. The findings mentioned above are applied to the proposed algorithm. The proposed community detection algorithm is an enhancement over Fire Spread community detection algorithm (Pattanayak, 2019), in which the value of R for the R-radius neighborhood subgraph is automatically calculated.
- Published
- 2022
9. Noise reduction of signals received from wearable sensors along with integrating their information with machine learning.
- Author
-
Hadjian Jaber, Mohammad Sedigh and Kazemi, Akram
- Subjects
- *
NOISE control , *COMPUTERS in medicine , *WEARABLE technology , *ACQUISITION of data , *SIGNALS & signaling - Abstract
Sensors have been widely used in data collection systems, especially for medical applications. However, these types of devices suffer from different types of noise, which reduces the quality of the data and thus the reliability. Low data quality is a major barrier to computer-based diagnostics. Therefore, noise tolerance is essential in data collection and analysis based on wearable sensors. In this regard, in this study, two important issues in this area, namely the elimination of noise from received signals and the integration of a set of signals in order to improve decision making have been examined. The simulation results show that the proposed frameworks have a good performance in improving the accuracy and increasing the processing speed. [ABSTRACT FROM AUTHOR]
- Published
- 2020
10. El efecto de la estructura de las redes sociales sobre la confianza. Un análisis de simulaciones computarizadas y evaluación de la tesis de Coleman.
- Author
-
Ignacio García-Valdecasas, José
- Subjects
CONTROL theory (Engineering) ,SOCIAL impact ,COMPUTER simulation ,SOCIAL structure ,GLOBALIZATION ,SOCIAL networks - Abstract
Copyright of Papers: Revista de Sociologia is the property of Universitat Autonoma de Barcelona and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2020
- Full Text
- View/download PDF
11. Cost-Aware Deployment of Check-In Nodes in Complex Networks
- Author
-
Jing Yuan, Sanyang Liu, Qian Li, and Yiguang Bai
- Subjects
Class (computer programming) ,Optimization problem ,Check-in ,Computer science ,Distributed computing ,Artificial networks ,Scale (descriptive set theory) ,Complex network ,Computer Science Applications ,Human-Computer Interaction ,Control and Systems Engineering ,Software deployment ,Metric (mathematics) ,Electrical and Electronic Engineering ,Software - Abstract
It is challenging to deploy check-in nodes optimally in a complex network so as to perform specific check-in like services, e.g., fuel supplements etc., but crucial in many real-world applications. In this article, we propose and study the new optimization problem of placing check-in nodes with the minimum cost, i.e., the problem of finding the minimum-cost check-in nodes (MCCN), which is of great interests for real-world application situations, but even more difficult. We motivate the new algorithms through three typical worse cases by the often-used greedy-type algorithms, i.e., One-to-Many, Many-to-One, and Duplicate-Overrides. With this respect, the proposed novel optimization algorithms utilize a novel metric of contribution density for selecting check-in nodes iteratively, which successfully avoid the occurrence of the two worse cases of One-to-Many and Many-to-One. We also introduce an extra backward extraction step in one of new algorithms, which overcomes the crucial worse case of ``duplicate overrides'' and largely improves the algorithmic performance in solving the introduced optimization problem of MCCN. Meanwhile, we extend the new contribution density metric to a more general class of functions and study their effectiveness to eliminate the two worse cases of One-to-Many and Many-to-One; also, a detailed analysis on complexity and performance of the proposed algorithms is presented to show their numerical efficiency and accuracy. Extensive experiments over two classical artificial networks, i.e., BA network and ER network, and ten real-world networks, under two typical cost setups, show the proposed algorithms significantly outperform the four state-of-the-art algorithms. We also demonstrate that our proposed algorithms are much reliable and robust with different experiment settings of deployment cost and network-type and scale.
- Published
- 2022
12. Identifying ecological strategic points based on multi-functional ecological networks: A case study of Changzhi City, China.
- Author
-
Yuan, Yuan, Wang, Juan, Qiao, Na, Huang, Yuhan, and Bai, Zhongke
- Subjects
- *
ECOLOGICAL disturbances , *CORRIDORS (Ecology) , *REGIONAL development , *BIOLOGICAL networks , *PROTECTED areas , *HYDROELECTRIC power plants - Abstract
Building ecological networks with close connections and less disturbance is important for balancing regional development and ecological protection. The construction of traditional ecological networks focuses on the potential ecological corridors extracted based on the minimum cumulative resistance model (MCR), and less on other types of corridors. The extraction of ecological strategic points mainly prioritizes their characteristics in the weak areas of ecological corridors and is concentrated on a single ecological network. Taking the case of Changzhi, China, this study comprehensively considers biological, hydrological, and human processes to construct natural ecological networks (i.e., biological migration networks and hydro-ecological networks) and artificial networks by linking multi-source data and using spatial analysis tools, including the MARXAN, MCR, InVEST 3.6.0, and ArcGIS 10.2. A coupling analysis of natural ecological networks and artificial networks is used to extract the intersection areas between the elements of the natural ecological network as ecological key points, as well as the conflict areas between natural ecological and artificial networks as ecological disturbance points, revealing the distribution characteristics of ecological strategic points to propose corresponding recommendations. The results reveal that (1) the source area of multi-functional ecological networks is hydro-ecological sources > biological conservation sources > socioeconomic sources. The corridor length is artificial corridors > hydro-ecological corridors > biological migration corridors. (2) There were 567 ecological key points with an area of 1716.61 km2 to be protected, to promote the spatial connection of ecological elements. Additionally, 6.81% of the socioeconomic source area should be converted to natural ecological sources. The areas spanning 2.7 km2, 3.9 km2, and 1.76 km2 corresponding to the first-, second-, and third-level natural corridors in the socioeconomic sources should belong to natural corridors. (3) The proportions of the disturbed biological conservation source area and corridor length were 4.24% and 0.58%, respectively. The area of the disturbed hydro-ecological sources and length of the disturbed corridors accounted for 0.93% and 0.12%, respectively. There were 2266 ecological disturbance points with an area of 180.91 km2 that needed to be controlled by eliminating, weakening, and early warning. These results provide a reference for future ecological space delineations, and can be used to coordinate ecological protection and regional development. • Multi-functional ecological networks were constructed by considering natural ecological and artificial processes. • Ecological strategic points were extracted through a coupling analysis of natural ecological and artificial networks. • The differentiated recommendations of ecological strategic points were proposed. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
13. A data-driven dynamic method of downhole rock characterisation for the vibro-impact drilling system.
- Author
-
Afebu, Kenneth Omokhagbo, Liu, Yang, and Papatheou, Evangelos
- Subjects
- *
MACHINE learning , *CONVOLUTIONAL neural networks , *PARAMETRIC modeling , *SIGNAL processing , *PROSPECTING - Abstract
For the real-time characterisation of an inhomogeneous impact inhibiting constraint such as downhole rock layers, an unconventional method using machine learning (ML) and drill-bit vibrations is investigated. An impact oscillator with one-sided elastic constraint is employed in modelling the bit-rock impact actions. Measurable drill-bit dynamics, such as acceleration, were acquired and processed into features and 2D-images that were later used in developing ML models capable of predicting the stiffness of impacted rock constraint. Explored ML networks include Multilayer Perceptron (MLP), Convolutional Neural Network and Long Short-Term Memory Network. Both simulation and experimental studies have been presented to validate the proposed method while using coefficient of determination (R 2) and normalised mean absolute error (NMAE) as the performance metrics of the ML models. Results showed that the feature-based models had better performances for both simulation and experiment compared to the raw signal and 2D-image based models. Aside being simple and computationally less expensive, the feature-based MLP models outperformed other models having R 2 values > 0.7 and NMAE values < 0.2 for both simulation and experiment, thus presenting them as the preferred ML model for dynamic downhole rock characterisation. In general, this study presents a new modality to achieving logging-while-drilling during deep-hole drilling operations such as carried out in hydrocarbon, mineral and geothermal exploration. [ABSTRACT FROM AUTHOR]
- Published
- 2025
- Full Text
- View/download PDF
14. Kohonen Artificial Networks for the Verification of the Diameters of Water-pipes
- Author
-
Jacek Dawidowicz
- Subjects
Self-organizing map ,Computer science ,Artificial networks ,Water pipe ,General Environmental Science ,Marine engineering - Abstract
The design of the water distribution system is inherently linked to the execution of calculations, which aim, among other things, to determine the flow rate through individual pipes and the selection of diameters at the appropriate speed. Each step in the calculations is followed by an evaluation of the results and, if necessary, a correction of the data and further calculations. It is up to the designer to analyse the accuracy of the calculation results and is time-consuming for large systems. In this article, a diagnostic method for the results of hydraulic calculations, based on Kohonen Network, which classifies nominal diameters [DN] on the basis of data, in the form of flows, has been proposed. After calculating the new variant of the water distribution system, the individual calculation sections are assigned to the neurons of the topological map of Kohonen Network drawn up for nominal diameters. By comparing the diameter used for the calculation, with the diameter obtained on the topological map, the accuracy of the chosen diameter can be assessed. The topological map, created as a result of labelling the neurons of the output layer of the Kohonen Network, graphically shows the position of the classified diameter, relative to those diameters with similar input values. The position of a given diameter, relative to other diameters, may suggest the need to change the diameter of the pipe.
- Published
- 2021
15. Mapping Recent Lava Flows at Mount Etna Using Multispectral Sentinel-2 Images and Machine Learning Techniques
- Author
-
Claudia Corradino, Gaetana Ganci, Annalisa Cappello, Giuseppe Bilotta, Alexis Hérault, and Ciro Del Negro
- Subjects
volcano hazard monitoring ,satellite remote sensing ,artificial networks ,automatic detection ,Sentinel Mission ,Pléiades Mission ,Science - Abstract
Accurate mapping of recent lava flows can provide significant insight into the development of flow fields that may aid in predicting future flow behavior. The task is challenging, due to both intrinsic properties of the phenomenon (e.g., lava flow resurfacing processes) and technical issues (e.g., the difficulty to survey a spatially extended lava flow with either aerial or ground instruments while avoiding hazardous locations). The huge amount of moderate to high resolution multispectral satellite data currently provides new opportunities for monitoring of extreme thermal events, such as eruptive phenomena. While retrieving boundaries of an active lava flow is relatively straightforward, problems arise when discriminating a recently cooled lava flow from older lava flow fields. Here, we present a new supervised classifier based on machine learning techniques to discriminate recent lava imaged in the MultiSpectral Imager (MSI) onboard Sentinel-2 satellite. Automated classification evaluates each pixel in a scene and then groups the pixels with similar values (e.g., digital number, reflectance, radiance) into a specified number of classes. Bands at the spatial resolution of 10 m (bands 2, 3, 4, 8) are used as input to the classifier. The training phase is performed on a small number of pixels manually labeled as covered by fresh lava, while the testing characterizes the entire lava flow field. Compared with ground-based measurements and actual lava flows of Mount Etna emplaced in 2017 and 2018, our automatic procedure provides excellent results in terms of accuracy, precision, and sensitivity.
- Published
- 2019
- Full Text
- View/download PDF
16. Data-driven modelling of the FRC network for studying the fluid flow in the conduit system.
- Author
-
Savinkov, Rostislav, Kislitsyn, Alexey, Watson, Daniel J., Loon, Raoul van, Sazonov, Igor, Novkovic, Mario, Onder, Lucas, and Bocharov, Gennady
- Subjects
- *
IMMUNE system , *RETICULAR formation , *LYMPH nodes , *COMPUTATIONAL biology , *STANDARD deviations , *GEOMETRIC modeling - Abstract
The human immune system is characterized by enormous cellular and anatomical complexity. Lymph nodes are key centers of immune reactivity, organized into distinct structural and functional modules including the T-cell zone, fibroblastic reticular cell (FRC) network and the conduit system. A thorough understanding of the modular organization is a prerequisite for lymphoid organ tissue-engineering. Due to the biological complexity of lymphoid organs, the development of mathematical models capable of elaborating the lymph node architecture and functional organization, has remained a major challenge in computational biology. Here, we present a computational method to model the geometry of the FRC network and fluid flow in the conduit system. It differs from the blood vascular network image-based reconstruction approaches as it develops the parameterized geometric model using the real statistics of the node degree and the edge length distributions. The FRC network model is then used to analyze the fluid flow through the underlying conduit system. A first observation is that the pressure gradient is approximately linear, which suggests homogeneity of the network. Furthermore, calculated permeability values ( ≈ 0.0033 μ m 2 ) show the generated network is isotropic, while investigating random variations of pipe radii (with a given mean and standard deviation) shows a significant effect on the permeability. This framework can now be further explored to systematically correlate fundamental characteristics of the FRC conduit system to more global material properties such as permeability. [ABSTRACT FROM AUTHOR]
- Published
- 2017
- Full Text
- View/download PDF
17. An Improved Spectral Clustering Community Detection Algorithm Based on Probability Matrix
- Author
-
Shuxia Ren, Tao Wu, and Shubo Zhang
- Subjects
Spectral clustering algorithm ,Article Subject ,Markov chain ,Computer science ,010102 general mathematics ,Stochastic matrix ,Artificial networks ,02 engineering and technology ,01 natural sciences ,Spectral clustering ,Graph ,Modeling and Simulation ,QA1-939 ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,0101 mathematics ,Cluster analysis ,Algorithm ,Mathematics - Abstract
The similarity graphs of most spectral clustering algorithms carry lots of wrong community information. In this paper, we propose a probability matrix and a novel improved spectral clustering algorithm based on the probability matrix for community detection. First, the Markov chain is used to calculate the transition probability between nodes, and the probability matrix is constructed by the transition probability. Then, the similarity graph is constructed with the mean probability matrix. Finally, community detection is achieved by optimizing the NCut objective function. The proposed algorithm is compared with SC, WT, FG, FluidC, and SCRW on artificial networks and real networks. Experimental results show that the proposed algorithm can detect communities more accurately and has better clustering performance.
- Published
- 2020
18. A Puzzle concerning Compositionality in Machines
- Author
-
Ryan M. Nefdt
- Subjects
Cognitive science ,Philosophy of mind ,Philosophy of science ,Relation (database) ,Computer science ,business.industry ,Principle of compositionality ,Deep learning ,Artificial networks ,Philosophy ,Artificial Intelligence ,Theory of computation ,Deep neural networks ,Artificial intelligence ,business - Abstract
This paper attempts to describe and address a specific puzzle related to compositionality in artificial networks such as Deep Neural Networks and machine learning in general. The puzzle identified here touches on a larger debate in Artificial Intelligence related to epistemic opacity but specifically focuses on computational applications of human level linguistic abilities or properties and a special difficulty with relation to these. Thus, the resulting issue is both general and unique. A partial solution is suggested.
- Published
- 2020
19. Overlapping community detection based on the union of all maximum spanning trees
- Author
-
Dounia Lotfi, Mohamed El Marraki, and Khawla Asmi
- Subjects
Modularity (networks) ,Ground truth ,Theoretical computer science ,Spanning tree ,Social network ,business.industry ,Computer science ,Node (networking) ,Artificial networks ,0102 computer and information sciences ,02 engineering and technology ,Library and Information Sciences ,01 natural sciences ,Execution time ,Local community ,010201 computation theory & mathematics ,020204 information systems ,0202 electrical engineering, electronic engineering, information engineering ,business ,Information Systems - Abstract
Purpose The state-of-the-art methods designed for overlapping community detection are limited by their high execution time as in CPM or the need to provide some parameters like the number of communities in Bigclam and Nise_sph, which is a nontrivial information. Hence, there is a need to develop the accuracy that represents the primordial goal, where the actual state-of-the-art methods do not succeed to achieve high correspondence with the ground truth for many instances of networks. The paper aims to discuss this issue. Design/methodology/approach The authors offer a new method that explore the union of all maximum spanning trees (UMST) and models the strength of links between nodes. Also, each node in the UMST is linked with its most similar neighbor. From this model, the authors extract local community for each node, and then they combine the produced communities according to their number of shared nodes. Findings The experiments on eight real-world data sets and four sets of artificial networks show that the proposed method achieves obvious improvements over four state-of-the-art (BigClam, OSLOM, Demon, SE, DMST and ST) methods in terms of the F-score and ONMI for the networks with ground truth (Amazon, Youtube, LiveJournal and Orkut). Also, for the other networks, it provides communities with a good overlapping modularity. Originality/value In this paper, the authors investigate the UMST for the overlapping community detection.
- Published
- 2020
20. TSCDA: A novel greedy approach for community discovery in networks
- Author
-
M. Dehghan Chenary, A. Ferdowsi, and Alireza Khanteymoori
- Subjects
Computer science ,business.industry ,Node (networking) ,media_common.quotation_subject ,Artificial networks ,computer.software_genre ,Measure (mathematics) ,Two stages ,Local search (optimization) ,Quality (business) ,Data mining ,business ,computer ,Network analysis ,media_common - Abstract
In this paper, we introduce a new approach for detecting community structures in networks. The approach is subject to modifying one of the connectivity-based community quality functions based on considering the impact that each community’s most influential node has on the other vertices. Utilizing the proposed quality measure, we devise an algorithm that aims to detect high-quality communities of a given network based on two stages: finding a promising initial solution using greedy methods and then refining the solutions in a local search manner.The performance of our algorithm has been evaluated on some standard real-world networks as well as on some artificial networks. The experimental results of the algorithm are reported and compared with several state-of-the-art algorithms. The experiments show that our approach is competitive with the other well-known techniques in the literature and even outperforms them. This approach can be used as a new community detection method in network analysis.
- Published
- 2021
21. Use of Bayesian Networks to Analyze Port Variables in Order to Make Sustainable Planning and Management Decision
- Author
-
Beatriz Molina Serrano, Nicoleta González-Cancelas, Francisco Soler-Flores, Samir Awad-Nuñez, and Alberto Camarero Orive
- Subjects
Bayesian Networks ,graph theory ,sustainability ,port management ,artificial networks ,Transportation and communication ,K4011-4343 ,Management. Industrial management ,HD28-70 ,Transportation and communications ,HE1-9990 - Abstract
In the current economic, social and political environment, society demands a greater variety of outcomes from the public logistics sector, such as efficiency, efficiency of managed resources, greater transparency and business performance. All of them are an indispensable counterpart for its recognition and support. In case of port planning and management, many variables are included. Use of Bayesian Networks allows to classify, predict and diagnose these variables and even to estimate the subsequent probability of unknown variables, basing on the known ones. Research includes a data base with more than 40 variables, which have been classified as smart port studies in Spain. Then a network was generated using a non-cyclic conducted grafo, which shows port variable relationships. As conclusion, economic variables are cause of the rest of categories and they represent a parent role in the most of cases. Furthermore, if environmental variables are known, subsequent probability of social variables can be estimated.
- Published
- 2018
- Full Text
- View/download PDF
22. Artificial networks for spectral resolution of antibiotic residues in bovine milk; solidification of floating organic droplet in dispersive liquid-liquid microextraction for sample treatment
- Author
-
Heba T. Elbalkiny and Ali M. Yehia
- Subjects
Residue (complex analysis) ,Chromatography ,medicine.drug_class ,Chemistry ,Liquid Phase Microextraction ,Sample (material) ,Antibiotics ,Extraction (chemistry) ,Artificial networks ,Veterinary Drugs ,Oxytetracycline ,Atomic and Molecular Physics, and Optics ,Analytical Chemistry ,Anti-Bacterial Agents ,Milk ,medicine ,media_common.cataloged_instance ,Liquid liquid ,Animals ,Humans ,European union ,Instrumentation ,Spectroscopy ,media_common ,medicine.drug - Abstract
The intensive use of antibiotics in livestock practice has a negative impact on human health and increases the antibiotic resistance. In this study feasible data interpretation algorithm along with efficient extraction protocol were combined for selective analysis of three antibiotics in milk samples. Trimethoprim, sulphamethoxazole and oxytetracycline are widely used antibiotics in veterinary pharmaceuticals. The studied antibiotics were efficiently extracted from milk samples with solidification of floating organic droplet in dispersive liquid-liquid microextraction. This extraction protocol was optimized not only to maximize extraction recoveries but also to approach the lower residue limits specified by European Union. Artificial neural networks succeeded in resolving spectral overlap between the studied drugs. The network architecture was optimized and validated for accurate and precise analysis. The proposed method outweighs the reported chromatographic methods for being simple and inexpensive and compared favorable to official methods.
- Published
- 2021
23. Dynamic task-belief is an integral part of decision-making
- Author
-
Lily E. Kramer, Marlene R. Cohen, and Cheng Xue
- Subjects
Neurons ,biology ,media_common.quotation_subject ,General Neuroscience ,Decision Making ,Artificial networks ,Macaca mulatta ,Macaque ,Task (project management) ,Electrophysiology ,Cognition ,Perception ,biology.animal ,Feature (machine learning) ,Animals ,Natural (music) ,Noise (video) ,Nerve Net ,Visual Cortex ,Cognitive psychology ,media_common - Abstract
SummaryNatural decisions involve two seemingly separable processes: inferring the relevant task (task-belief) and performing the believed-relevant task. The assumed separability has led to the traditional practice of studying task-switching and perceptual decision-making individually. Here, we used a novel paradigm to manipulate and measure macaque monkeys’ task-belief, and demonstrated inextricable neuronal links between flexible task-belief and perceptual decision-making. We showed that in animals, but not artificial networks that performed as well or better than the animals, stronger task-belief is associated with better perception. Correspondingly, recordings from neuronal populations in cortical areas 7a and V1 revealed that stronger task-belief is associated with better discriminability of the believed-relevant but not the believed-irrelevant feature. Perception also impacts belief updating: noise fluctuations in V1 help explain how task-belief is updated. Our results demonstrate that complex tasks and multi-area recordings can reveal fundamentally new principles of how biology affects behavior in health and disease.
- Published
- 2022
24. Electrofacies analysis for coal lithotype profiling based on high-resolution wireline log data.
- Author
-
Roslin, A. and Esterle, J.S.
- Subjects
- *
COAL mining , *GEOPHYSICS , *DATA analysis , *ARTIFICIAL neural networks , *CLUSTER analysis (Statistics) - Abstract
The traditional approach to coal lithotype analysis is based on a visual characterisation of coal in core, mine or outcrop exposures. As not all wells are fully cored, the petroleum and coal mining industries increasingly use geophysical wireline logs for lithology interpretation.This study demonstrates a method for interpreting coal lithotypes from geophysical wireline logs, and in particular discriminating between bright or banded, and dull coal at similar densities to a decimetre level. The study explores the optimum combination of geophysical log suites for training the coal electrofacies interpretation, using neural network conception, and then propagating the results to wells with fewer wireline data. This approach is objective and has a recordable reproducibility and rule set.In addition to conventional gamma ray and density logs, laterolog resistivity, microresistivity and PEF data were used in the study. Array resistivity data from a compact micro imager (CMI tool) were processed into a single microresistivity curve and integrated with the conventional resistivity data in the cluster analysis. Microresistivity data were tested in the analysis to test the hypothesis that the improved vertical resolution of microresistivity curve can enhance the accuracy of the clustering analysis. The addition of PEF log allowed discrimination between low density bright to banded coal electrofacies and low density inertinite-rich dull electrofacies.The results of clustering analysis were validated statistically and the results of the electrofacies results were compared to manually derived coal lithotype logs. [ABSTRACT FROM AUTHOR]
- Published
- 2016
- Full Text
- View/download PDF
25. Paths
- Author
-
Gulyás, András, Heszberger, Zalán, and Biró, József
- Subjects
Popular Science in Mathematics ,Mathematics ,network science ,graph theory ,path ,shortest path ,internet ,brain ,routing policy ,social networks ,artificial networks ,computer networks ,road networks ,open access ,Popular and recreational mathematics ,bic Book Industry Communication::P Mathematics & science::PD Science: general issues::PDZ Popular science::PDZM Popular mathematics - Abstract
This open access book explores the amazing similarity between paths taken by people and many other things in life, and its impact on the way we live, teach and learn. Offering insights into the new scientific field of paths as part of the science of networks, it entertainingly describes the universal nature of paths in large networked structures. It also shows the amazing similarity in the ways humans and other – even nonliving – things navigate in a complex environment, to allow readers to easily grasp how paths emerge in many walks of life, and how they are navigated. Paths is based on the authors recent research in the area of paths on networks, which points to the possible birth of the new science of “paths” as a natural consequence ‘and extension) of the science of “networks.” The approach is essentially story-based, supported by scientific findings, interdisciplinary approaches, and at times, even philosophical points of view. It also includes short illustrative anecdotes showing the amazing similarities between real-world paths and discusses their applications in science and everyday life. Paths will appeal to network scientists and to anyone interested in popular science. By helping readers to step away from the “networked” view of many recent popular scientific books and start to think of longer paths instead of individual links, it sheds light on these problems from a genuinely new perspective. --------------------------------------------------------------------------------- The path is the goal. The essence behind this short sentence is known to many people around the world, expressed through the interpretations of some of the greatest thinkers like Lao-Tze and Gandhi. It means that it is the journey that counts, not the destination. When speaking about such subjective and intangible things, philosophy and religion are some of the only approaches that are addressed. In this book, the authors address this conventional wisdom from the perspective of natural science. They explore a sequence of steps that leads the reader closer to the nature of paths and accompany him on the search for “the path to paths”.
- Published
- 2021
- Full Text
- View/download PDF
26. Fractality in Water Distribution Networks: Application to Criticality Analysis and Optimal Rehabilitation
- Author
-
David Butler, Bogumil Ulanicki, and Kegong Diao
- Subjects
Criticality ,Optimization ,Theoretical computer science ,Distribution networks ,Computer science ,Geography, Planning and Development ,Fractality ,Artificial networks ,Complexity ,Failure mode, effects, and criticality analysis ,Fractal ,Feature (computer vision) ,Water distribution networks ,Mathematics::Metric Geometry ,Water Science and Technology - Abstract
The file attached to this record is the author's final peer reviewed version. The Publisher's final version can be found by following the DOI link. Fractals have been identified as a common feature in many natural and artificial networks that exhibit self-similarity of the topological patterns, i.e. different parts of the system have similar structures to each other as well as to the whole system. This study investigates the fractality in water distribution networks (WDNs) and the application of the fractal property in WDNs analysis. Specifically, we explore the existence of fractal topological patterns in eight real-world WDNs of different complexities by using the box-covering algorithm. The results demonstrate all of the studied WDNs are fractal. Moreover, the application of the fractal property is demonstrated via critical pipe identification and optimal rehabilitation of benchmark real-world WDNs. All results show that the fractal-based approach can achieve better or equally good solutions compared with conventional methods in a much more efficient way, e.g. via automation of some processes or significant reduction in the search space/components to consider.
- Published
- 2021
27. Biological Signals Identification by a Dynamic Recurrent Neural Network: from Oculomotor Neural Integrator to Complex Human Movements and Locomotion
- Author
-
Guy Cheron, Ana Maria Cebolla, Ana Bengoetxea, Françoise Leurs, Jean-Philippe Draye, Bernard Dan, and Pablo d’Alcantara
- Subjects
Identification (information) ,Recurrent neural network ,Artificial neural network ,business.industry ,Computer science ,Integrator ,Artificial networks ,Control engineering ,Artificial intelligence ,Functional organization ,business ,Field (computer science) ,Humanoid robot - Abstract
The recent advances in the application of artificial neural networks in the biological field have been inspired by the functional organization of real biological structures (Draye et al.,1997a; Anastasio & Gad, 2007). The fascination exerted by the oculomotor system upon both engineers and neuroscientists have played an important role in this issue. In particular, since the definitive evidence of the existence of a neural integrator in the brainstem (Cheron et al., 1986a; Cannon & Robinson, 1987; Robinson, 1989 for a review) performing mathematical integration of the eye velocity into eye position signals, numerous artificial networks have been developed allowing a better understanding of the fundamental question of how the brain control movement. Such bio-mimetic strategy has recently permitted to elaborate different dynamic recurrent neural networks (DRNN) specifically dedicated to the command of humanoid robot (Tani et al., 2008). Hierarchical neuralinspired modules have also been proposed forming cascades of forward dynamics models (Jordan & Rumelhart, 1992; Kawato et al., 1987; Tani, 2003) in which top-down and bottomup influences allowed generating behavioural primitives. This Chapter describes the main steps performed in the development of our DRNN from the neural integrator models to those applied in the field of human movement control.
- Published
- 2021
28. Locating the Rumor Source in Social Networks using Timestamps
- Author
-
Anupam Biswas and Ravi Kishore Devarapalli
- Subjects
Root (linguistics) ,Signal processing ,business.industry ,Computer science ,Node (networking) ,Artificial networks ,Network structure ,Timestamp ,Rumor ,Scale (map) ,business ,Computer network - Abstract
We discuss the problem of locating the rumored origin in the social networks. Earlier works focused on network structure and sensor nodes; we concentrate on the timestamp of infected nodes. We recommend a two-stage algorithm to find the root node in a network using neighbor nodes' timestamp. In the first stage, discover the neighbor nodes with the minimal timestamp. Second, find the origin node using the minimal timestamp of the friend nodes. Simulations illustrate that our proposed two-stage algorithm works well and determines the root node within few hops. We evaluate our algorithm on small- scale networks, large-scale networks, and artificial networks. Our numerical results also show that our recommended two-stage algorithm operates well to discover the root node accurately, with a higher success rate and fewer distance error hops.
- Published
- 2021
29. Evolving the olfactory system with machine learning
- Author
-
Yi Sun, Richard Axel, Peter Y. Wang, Larry F. Abbott, and Guangyu Robert Yang
- Subjects
Glomerulus (olfaction) ,Olfactory system ,medicine.anatomical_structure ,Odor ,Artificial neural network ,Computer science ,Convergent evolution ,medicine ,Artificial networks ,Receptor type ,Neuroscience - Abstract
SummaryThe convergent evolution of the fly and mouse olfactory system led us to ask whether the anatomic connectivity and functional logic of olfactory circuits would evolve in artificial neural networks trained to perform olfactory tasks. Artificial networks trained to classify odor identity recapitulate the connectivity inherent in the olfactory system. Input units are driven by a single receptor type, and units driven by the same receptor converge to form a glomerulus. Glomeruli exhibit sparse, unstructured connectivity to a larger, expansion layer of Kenyon cells. When trained to both classify odor identity and to impart innate valence onto odors, the network develops independent pathways for identity and valence classification. Thus, the defining features of fly and mouse olfactory systems also evolved in artificial neural networks trained to perform olfactory tasks. This implies that convergent evolution reflects an underlying logic rather than shared developmental principles.
- Published
- 2021
30. Modeling the Local and Global Evolution Pattern of Community Structures for Dynamic Networks Analysis
- Author
-
Huaming Wu, Minghu Tang, Pengfei Jiao, Wei Yu, Wenjun Wang, and Yueheng Sun
- Subjects
Structure (mathematical logic) ,Dynamic network analysis ,General Computer Science ,Orthogonal non-negative matrix factorization (ONMF) ,Computer science ,Perspective (graphical) ,General Engineering ,Community structure ,Artificial networks ,computer.software_genre ,structure prediction ,Non-negative matrix factorization ,temporal community detection ,Snapshot (computer storage) ,General Materials Science ,Global evolution ,Data mining ,lcsh:Electrical engineering. Electronics. Nuclear engineering ,computer ,lcsh:TK1-9971 ,evolutionary pattern extraction - Abstract
Exploring and understanding the temporal structure of dynamic networks attract extensive attention over the past few years. Most of these current research focuses on temporal community detection, evolution analysis or link prediction from a mission-oriented perspective. In fact, these three tasks should be not isolated but mutually reinforcing. Transforming these three tasks into a unified framework, it is crucial to extract the evolution pattern, which helps to understand the time-varying characteristics of temporal structure in essence. In addition, to the best of our knowledge, there is no work focusing on modeling and uncovering the local and global evolution pattern hidden in temporal community structure, simultaneously. In this paper, we propose a novel framework based on Orthogonal Nonnegative Matrix Factorization to Explore the Evolution Pattern (ONMF-EEP) for analyzing and predicting the time-varying structures in dynamic networks from local and global perspectives. The nature of this framework assumes that community structures are subject to a local evolution pattern (LEP) at each snapshot, and these LEPs are from a common global evolution pattern (GEP). The framework can synchronously detect temporal community structure, extract evolution pattern, and predict structure including communities and future snapshot links. The extensive experiments on real-world networks and artificial networks demonstrate that our proposed framework is highly effective on the tasks of dynamic network analysis.
- Published
- 2019
31. Community Detection in Complex Networks by Detecting and Expanding Core Nodes Through Extended Local Similarity of Nodes
- Author
-
Asgarali Bouyer, Kamal Berahmand, and Mahdi Vasighi
- Subjects
Computer science ,business.industry ,Artificial networks ,02 engineering and technology ,Complex network ,computer.software_genre ,01 natural sciences ,Graph ,Human-Computer Interaction ,Global information ,Central node ,Modeling and Simulation ,0103 physical sciences ,Outlier ,Computer data storage ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,Data mining ,010306 general physics ,business ,Local algorithm ,computer ,Social Sciences (miscellaneous) - Abstract
As the community detection is able to facilitate the discovery of hidden information in complex networks, it has been drawn a lot of attention recently. However, due to the growth in computational power and data storage, the scale of these complex networks has grown dramatically. In order to detect communities by utilizing global approaches, it is required to have all the global information of the whole network; something which is impossible, because of the rapid growth in the size of the networks. In this paper, a local approach has been proposed based on the detection and expansion of core nodes. First, a community’s central node (core node) which has a high level of embeddedness is detected based on the similarity between graph’s nodes. By using this, the total weights of a weighted graph’s edges created. Following by that, the expansion of these nodes will be considered, by utilizing the concept of node’s membership based on the definition of strong community for weighted graphs. It can be seen that in detecting communities, the more accurate the weights of edges detected based on the node similarity, the more precise the local algorithm will be. In fact, the algorithm has the ability to detect all the graph’s communities in a network using local information as well as identifying various roles of nodes, either being (core or outlier). Test results on both real-world and artificial networks prove that the quality of the communities which are detected by the proposed algorithm is better than the results which are achieved by other state-of-the-art algorithms in the complex networks.
- Published
- 2018
32. Modeling and analyzing users’ behavioral strategies with co-evolutionary process
- Author
-
Toshiharu Sugawara, Yutaro Miura, and Fujio Toriumi
- Subjects
Facebook ,Computer science ,Process (engineering) ,media_common.quotation_subject ,Internet privacy ,Public goods game ,02 engineering and technology ,lcsh:QA75.5-76.95 ,020204 information systems ,Reading (process) ,Genetic algorithm ,0202 electrical engineering, electronic engineering, information engineering ,media_common ,Agent-based simulation ,lcsh:T58.5-58.64 ,lcsh:Information technology ,business.industry ,Artificial networks ,Computer Science Applications ,Co-evolution ,Human-Computer Interaction ,Ego networks ,Social networking services ,Free rider problem ,Modeling and Simulation ,Social relationship ,020201 artificial intelligence & image processing ,lcsh:Electronic computers. Computer science ,business ,Information Systems - Abstract
Social networking services (SNSs) are constantly used by a large number of people with various motivations and intentions depending on their social relationships and purposes, and thus, resulting in diverse strategies of posting/consuming content on SNSs. Therefore, it is important to understand the differences of the individual strategies depending on their network locations and surroundings. For this purpose, by using a game-theoretical model of users called agents and proposing a co-evolutionary algorithm called multiple-world genetic algorithm to evolve diverse strategy for each user, we investigated the differences in individual strategies and compared the results in artificial networks and those of the Facebook ego network. From our experiments, we found that agents did not select the free rider strategy, which means that just reading the articles and comments posted by other users, in the Facebook network, although this strategy is usually cost-effective and usually appeared in the artificial networks. We also found that the agents who mainly comment on posted articles/comments and rarely post their own articles appear in the Facebook network but do not appear in the connecting nearest-neighbor networks, although we think that this kind of user actually exists in real-world SNSs. Our experimental simulation also revealed that the number of friends was a crucial factor to identify users’ strategies on SNSs through the analysis of the impact of the differences in the reward for a comment on various ego networks.
- Published
- 2021
33. The New Abnormal: Network Anomalies in the AI Era
- Author
-
Francesca Soro, Danilo Giordano, Thomas Favale, Luca Vassio, Zied Ben Houidi, and Idilio Drago
- Subjects
reinforcement learning ,business.industry ,Computer science ,Artificial networks ,Denial-of-service attack ,anomalies, representation learning, GANs, autoencoders, reinforcement learning ,Machine learning ,computer.software_genre ,representation learning ,autoencoders ,Reinforcement learning ,anomalies ,Anomaly detection ,Artificial intelligence ,GANs ,business ,computer ,Feature learning ,Internet monitoring ,Anomalies ,Representation learning ,GANs, Autoencoders - Abstract
Anomaly detection aims at finding unexpected patterns in data. It has been used in several problems in computer networks, from the detection of port scans and distributed denial‐of‐service (DDoS) attacks to the monitoring of time series collected from Internet monitoring systems. Data‐driven approaches and machine learning have seen widespread application on anomaly detection too, and this trend has been accelerated by the recent developments on Artificial Intelligence (AI) research. This chapter summarizes ongoing recent progresses on anomaly detection research. In particular, we evaluate how developments on AI algorithms bring new possibilities for anomaly detection. We cover new representation learning techniques such as Generative Artificial Networks and Autoencoders, as well as techniques that can be used to improve models learned with machine learning algorithms, such as reinforcement learning. We survey both research works and tools implementing AI algorithms for anomaly detection. We found that the novel algorithms, while successful in other fields, have hardly been applied to networking problems. We conclude the chapter with a case study that illustrates a possible research direction.
- Published
- 2021
34. Analysis and control of agreement and disagreement opinion cascades
- Author
-
Anastasia S. Bizyaeva, Naomi Ehrich Leonard, Shinkyu Park, and Alessio Franci
- Subjects
Physics - Physics and Society ,0209 industrial biotechnology ,Computer science ,media_common.quotation_subject ,Control (management) ,Network structure ,FOS: Physical sciences ,02 engineering and technology ,Physics and Society (physics.soc-ph) ,020901 industrial engineering & automation ,Artificial Intelligence ,0202 electrical engineering, electronic engineering, information engineering ,FOS: Mathematics ,Mathematics - Optimization and Control ,Computer communication networks ,media_common ,business.industry ,Artificial networks ,Agreement ,Optimization and Control (math.OC) ,Key (cryptography) ,020201 artificial intelligence & image processing ,Artificial intelligence ,Centrality ,business ,Opinion formation - Abstract
We introduce and analyze a continuous time and state-space model of opinion cascades on networks of large numbers of agents that form opinions about two or more options. By leveraging our recent results on the emergence of agreement and disagreement states, we introduce novel tools to analyze and control agreement and disagreement opinion cascades. New notions of agreement and disagreement centrality, which depend only on network structure, are shown to be key to characterizing the nonlinear behavior of agreement and disagreement opinion formation and cascades. Our results are relevant for the analysis and control of opinion cascades in real-world networks, including biological, social, and artificial networks, and for the design of opinion-forming behaviors in robotic swarms. We illustrate an application of our model to a multi-robot task-allocation problem and discuss extensions and future directions opened by our modeling framework.
- Published
- 2021
- Full Text
- View/download PDF
35. Application of Image Recognition Based on Deep Learning Model
- Author
-
Qiushi Guan, Qiuzi He, Yuqi Wang, Hao He, Jie Zhao, and Xinyu Huang
- Subjects
Normalization (statistics) ,Speedup ,Artificial neural network ,Computer science ,business.industry ,Deep learning ,Computer Science::Neural and Evolutionary Computation ,Feature extraction ,Artificial networks ,Data set ,Computer Science::Computer Vision and Pattern Recognition ,Computer vision ,Artificial intelligence ,business ,MNIST database - Abstract
Traditional image recognition is mainly based on shallow neural networks, mainly including artificial networks, but these shallow neural networks have limited feature extraction capabilities. This paper can speed up the operation efficiency of the model, eliminate over-fitting, and improve the recognition accuracy by adding batch normalization layers to the neural network. The experiments are based on the MNIST data set, and have fully verified the recognition ability depth model proposed in this article.
- Published
- 2021
36. Virus-Information Coevolution Spreading Dynamics on Multiplex Networks
- Author
-
Xiaolin Qin, Jian Wang, and Hongying Fang
- Subjects
Multidisciplinary ,General Computer Science ,Article Subject ,Computer science ,Dynamics (mechanics) ,Artificial networks ,Complex system ,QA75.5-76.95 ,01 natural sciences ,Virus ,010305 fluids & plasmas ,Electronic computers. Computer science ,0103 physical sciences ,Multiplex ,010306 general physics ,Biological system ,Coevolution - Abstract
Virus and information spreading dynamics widely exist in complex systems. However, systematic study still lacks for the interacting spreading dynamics between the two types of dynamics. This paper proposes a mathematical model on multiplex networks, which considers the heterogeneous susceptibility and infectivity in two subnetworks. By using a heterogeneous mean-field theory, we studied the dynamic process and outbreak threshold of the system. Through extensive numerical simulations on artificial networks, we find that the virus’s spreading dynamics can be suppressed by increasing the information spreading probability, decreasing the protection power, or decreasing the susceptibility and infectivity.
- Published
- 2021
37. 3D Printed Self Driving Electric Vehicle
- Author
-
Abdullah S. Karar, S. alkork, Sherif Said, M. Sheikh, and Murat Ötkür
- Subjects
3d printed ,Acceleration ,business.product_category ,Chassis ,Tractive force ,Self driving ,Computer science ,Electric vehicle ,Artificial networks ,business ,Convolutional neural network ,Automotive engineering - Abstract
With the development in the areas such as high performance graphic processing units (GPU), sensor technologies and artificial networks the goal of self-driving vehicles has never been closer. With the promise of massive decrease in the road accidents, self-driving cars have been the center of attention for the past few years. In this paper we discuss the major components of a self-driving vehicles along with the methodologies required to produce a self-driving vehicle. With the added advent of the 3D printed body our aim is to rapid prototype a 3D printed Electric Vehicle with the goal to test and improve the current self-driving algorithms. The chassis of the car, with an initial body design was simulated and tested along with some basic parameters such as vehicle speed acceleration and traction force. The paper also presents the simulated results of an end-to-end approach to steering vehicle using convolutional neural networks, that would later be the basis of the self-driving algorithm.
- Published
- 2020
38. EL IMPACTO DE LA ESTRUCTURA DE LAS REDES SOCIALES SOBRE EL ACCESO DE LOS INDIVIDUOS AL MERCADO LABORAL.
- Author
-
GARCÍA-VALDECASAS, JOSÉ IGNACIO
- Abstract
Copyright of Revista Internacional de Sociología is the property of Consejo Superior de Investigaciones Cientificas and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2014
- Full Text
- View/download PDF
39. Detecting Overlapping Communities via Expanding Core Regions
- Author
-
Zhongzheng Zhang, Jianwu Li, Qingyao Liu, and Dingda Yang
- Subjects
Theoretical computer science ,Computer science ,Artificial networks ,Community structure ,Complex network ,Social circle ,Vertex (geometry) - Abstract
A social circle usually has some cores. Inspired by this phenomenon, we consider that a community in a complex network is also formed around one or several core vertices, called core region. We define the core score of a vertex to reflect its ability to attract other vertices. Utilizing the core scores of vertices, we address how to find the core region of one community, and then we apply local expansion and optimization to detect communities based on these core regions. Besides detecting community structure, our method can detect some abnormal vertices that are the hubs of network. Experimental results based on artificial networks and real-world networks show that our method is more effective than some usual methods.
- Published
- 2020
40. On modelling of artificial networks arising in applications
- Author
-
Felix Sadyrbaev and Svetlana Atslega
- Subjects
Computer science ,Distributed computing ,Artificial networks - Published
- 2020
41. Real-time burn depth assessment using artificial networks: a large-scale, multicentre study
- Author
-
Fangfang Li, Yu Zhang, Jiao Zhou, Chun Huang, Lu Kai, Zhiyou He, T. Li, Yuan Wang, Canqun Yang, Zuo Ke, Pihong Zhang, Xiang Chen, and P. Xie
- Subjects
Adult ,Burn injury ,medicine.medical_specialty ,China ,Time Factors ,Scale (ratio) ,Critical Care and Intensive Care Medicine ,Convolutional neural network ,030207 dermatology & venereal diseases ,03 medical and health sciences ,0302 clinical medicine ,Computer Systems ,medicine ,Medical imaging ,Humans ,Medical physics ,Wound Healing ,Burn depth ,business.industry ,Artificial networks ,030208 emergency & critical care medicine ,General Medicine ,Feature (computer vision) ,Clinical diagnosis ,Emergency Medicine ,Surgery ,business ,Burns - Abstract
Introduction Early judgment of the depth of burns is very important for the accurate formulation of treatment plans. In medical imaging the application of Artificial Intelligence has the potential for serving as a very experienced assistant to improve early clinical diagnosis. Due to lack of large volume of a particular feature, there has been almost no progress in burn field. Methods 484 early wound images are collected on patients who discharged home after a burn injury in 48 h, from five different levels of hospitals in Hunan Province China. According to actual healing time, all images are manually annotated by five professional burn surgeons and divided into three sets which are shallow(0–10 days), moderate(11–20 days) and deep(more than 21 days or skin graft healing). These ROIs were further divided into 5637 patches sizes 224 × 224 pixels, of which 1733 shallow, 1804 moderate, and 2100 deep. We used transfer learning suing a Pre-trained ResNet50 model and the ratio of all images is 7:1.5:1.5 for training:validation:test. Results A novel artificial burn depth recognition model based on convolutional neural network was established and the diagnostic accuracy of the three types of burns is about 80%. Discussion The actual healing time can be used to deduce the depth of burn involvement. The artificial burn depth recognition model can accurately infer healing time and burn depth of the patient, which is expected to be used for auxiliary diagnosis improvement.
- Published
- 2020
42. Tailored ensembles of neural networks optimize sensitivity to stimulus statistics
- Author
-
Jens Wilting, Johannes Zierenberg, Anna Levina, and Viola Priesemann
- Subjects
Computational Neuroscience ,Neurons, networks, dynamical systems ,Artificial neural network ,Quantitative Biology::Neurons and Cognition ,Dynamic range ,Computer science ,business.industry ,Artificial networks ,Reservoir computing ,FOS: Physical sciences ,Pattern recognition ,Disordered Systems and Neural Networks (cond-mat.dis-nn) ,Stimulus (physiology) ,Condensed Matter - Disordered Systems and Neural Networks ,FOS: Biological sciences ,Quantitative Biology - Neurons and Cognition ,Neurons and Cognition (q-bio.NC) ,Artificial intelligence ,business ,Biological network - Abstract
The dynamic range of stimulus processing in living organisms is much larger than a single neural network can explain. For a generic, tunable spiking network we derive that while the dynamic range is maximal at criticality, the interval of discriminable intensities is very similar for any network tuning due to coalescence. Compensating coalescence enables adaptation of discriminable intervals. Thus, we can tailor an ensemble of networks optimized to the distribution of stimulus intensities, e.g., extending the dynamic range arbitrarily. We discuss potential applications in machine learning., 6 pages plus supplemental material
- Published
- 2020
43. Dendritic normalisation improves learning in sparsely connected artificial neural networks
- Author
-
Hermann Cuntz and Alex D. Bird
- Subjects
medicine.anatomical_structure ,Artificial neural network ,Quantitative Biology::Neurons and Cognition ,Computer science ,business.industry ,Artificial networks ,medicine ,Artificial intelligence ,Neuron ,business - Abstract
Inspired by the physiology of neuronal systems in the brain, artificial neural networks have become an invaluable tool for machine learning applications. However, their biological realism and theoretical tractability are limited, resulting in poorly understood parameters. We have recently shown that biological neuronal firing rates in response to distributed inputs are largely independent of size, meaning that neurons are typically responsive to the proportion, not the absolute number, of their inputs that are active. Here we introduce such a normalisation, where the strength of a neuron’s afferents is divided by their number, to various sparsely-connected artificial networks. The learning performance is dramatically increased, providing an improvement over other widely-used normalisations in sparse networks. The resulting machine learning tools are universally applicable and biologically inspired, rendering them better understood and more stable in our tests.
- Published
- 2020
- Full Text
- View/download PDF
44. Evolutionary Mechanisms of Network Motifs in PPI Networks
- Author
-
Jinhu Lü and Pei Wang
- Subjects
Natural selection ,Computer science ,Mechanism (biology) ,Gene duplication ,Artificial networks ,Computational biology ,Enhanced Data Rates for GSM Evolution ,Divergence (statistics) ,Biological network - Abstract
Duplication and divergence are two basic evolutionary mechanisms of bio-molecular networks. Real-world bio-molecular networks and their statistical characteristics can be well mimicked by artificial algorithms based on the two mechanisms. Bio-molecular networks consist of network motifs, which act as building blocks of large-scale networks. A fundamental question is how network motifs are evolved from long time evolution and natural selection. By considering the effect of various duplication and divergence strategies, it is founded that the underlying duplication scheme of the real-world undirected bio-molecular networks would rather follow the anti-preference strategy than the random one. The anti-preference duplication mechanism and the dimerization can lead to the formation of various motifs and robustly conserve proper quantities of motifs in the artificial networks as that in the real-world ones. Furthermore, the anti-preference mechanism and edge deletion divergence can robustly preserve the sparsity of the networks. The investigations reveal the possible evolutionary mechanisms of network motifs and have potential implications in the design, synthesis, and reengineering of biological networks for biomedical purpose.
- Published
- 2020
45. El efecto de la estructura de las redes sociales sobre la confianza : un análisis de simulaciones computarizadas y evaluación de la tesis de Coleman
- Author
-
José Ignacio García Valdecasas Medina
- Subjects
Redes artificiales ,Món petit ,Sociology and Political Science ,Simulació informatitzada ,Social Sciences ,Gini index ,HM401-1281 ,mundo pequeño ,Clustering coefficient ,Mundo pequeño ,Coeficiente de conglomerado ,coeficiente de conglomerado ,Sociology (General) ,Experiments virtuals ,Índex de Gini ,Experimentos virtuales ,Tesi de Coleman ,Índice de Gini ,Simulación computarizada ,Computer simulation ,Artificial networks ,índice de Gini ,Coeficient de conglomerat ,Coleman's theory ,Virtual experiments ,experimentos virtuales ,simulación computarizada ,Tesis de Coleman ,Small world ,Xarxes artificials ,redes artificiales ,Social Sciences (miscellaneous) - Abstract
El objetivo del artículo es analizar el impacto de distintos tipos de estructuras y de diferentes propiedades estructurales de las redes sociales sobre la tendencia a confiar en el interior de ellas. Para llevar a cabo dicho objetivo, se ha realizado una serie de experimentos virtuales con redes artificiales a través de técnicas de simulación computarizadas. Los resultados ponen de manifiesto el profundo efecto de las propiedades estructurales (densidad de red, índice de globalización de vínculos e índice de Gini de la distribución de vínculos) y del tipo de estructura (regular, aleatoria y mundo pequeño) de las redes sociales sobre la tendencia a confiar entre los agentes de una red. Además, se evalúa la tesis de Coleman sobre el control y la vigilancia de los jóvenes para no abandonar los estudios a partir de los datos suministrados por los experimentos virtuales realizados con las redes artificiales anteriores. L'objectiu de l'article és analitzar l'impacte de diferents tipus d'estructures i de diferents propietats estructurals de les xarxes socials sobre la tendència a confiar a l'interior d'aquestes. Per dur a terme aquest objectiu, s'ha fet una sèrie d'experiments virtuals amb xarxes artificials a través de tècniques de simulació informatitzades. Els resultats posen de manifest el profund efecte de les propietats estructurals (densitat de xarxa, índex de globalització de vincles i índex de Gini de la distribució de vincles) i del tipus d'estructura (regular, aleatòria i món petit) de les xarxes socials sobre la tendència a confiar entre els agents d'una xarxa. A més, s'avalua la tesi de Coleman sobre el control i la vigilància dels joves per no abandonar els estudis a partir de les dades subministrades pels experiments virtuals realitzats amb les xarxes artificials anteriors. The aim of this article is to analyze the impact of different social network structures and structural properties on the tendency to trust inside networks. To this end, a series of virtual experiments with artificial networks have been carried out using computer simulations. The results underscore the profound effect of the structural properties (network density, link globalization index and Gini index of the distribution of links) and type of structure (regular, random and small world) of social networks on the tendency to trust among the agents of a network. In addition, Coleman's theory on the control and surveillance of young people is evaluated to study the data provided by the virtual experiments with the previous artificial networks.
- Published
- 2020
46. A Comparative Analysis of Community Detection Methods in Massive Datasets
- Author
-
Deepa Kumari, Subhrakanta Panda, and B. S. A. S. Rajita
- Subjects
Random graph ,Social network ,Computer science ,business.industry ,Artificial networks ,02 engineering and technology ,computer.software_genre ,Boom ,Synthetic data ,020204 information systems ,Scalability ,0202 electrical engineering, electronic engineering, information engineering ,020201 artificial intelligence & image processing ,Data mining ,business ,computer - Abstract
Nowadays there is a boom in social network data streaming from various fields of interest related to finance, engineering, medicine, and general sciences. All these data are modeled as graphs for better analysis. Community detection is one such mechanism for the analysis of such massive data. Many community detection algorithms exist in literature. The existing algorithms are compared by using either real-world or artificial networks (modeled as graphs) but not both. This paper aims to make a comparative study of two popular existing community detection algorithms both on real-world and synthetic data and verify their performance. The approach in this paper makes good use of recent advances in graphical modeling of different social networks. We generated a random graph that represents most of the observed properties of a real-world dataset. The experimental results are tabulated and the computed metrics help in inferring the suitability or scalability of an algorithm for small or massive datasets.
- Published
- 2020
47. Identifying Condition-Specific Modules by Clustering Multiple Networks
- Author
-
Penggang Sun, Guimin Qin, and Xiaoke Ma
- Subjects
0301 basic medicine ,Structure (mathematical logic) ,Modularity (networks) ,business.industry ,Applied Mathematics ,Artificial networks ,Cancer therapy ,Machine learning ,computer.software_genre ,03 medical and health sciences ,030104 developmental biology ,Discriminative model ,Genetics ,Sensitivity (control systems) ,Artificial intelligence ,Cluster analysis ,business ,computer ,Biotechnology ,Mathematics - Abstract
Condition-specific modules in multiple networks must be determined to reveal the underlying molecular mechanisms of diseases. Current algorithms exhibit limitations such as low accuracy and high sensitivity to the number of networks because these algorithms discover condition-specific modules in multiple networks by separating specificity and modularity of modules. To overcome these limitations, we characterize condition-specific module as a group of genes whose connectivity is strong in the corresponding network and weak in other networks; this strategy can accurately depict the topological structure of condition-specific modules. We then transform the condition-specific module discovery problem into a clustering problem in multiple networks. We develop an efficient heuristic algorithm for the S pecific M odules in M ultiple N etworks ( SMMN ), which discovers the condition-specific modules by considering multiple networks. By using the artificial networks, we demonstrate that SMMN outperforms state-of-the-art methods. In breast cancer networks, stage-specific modules discovered by SMMN are more discriminative in predicting cancer stages than those obtained by other techniques. In pan-cancer networks, cancer-specific modules are more likely to associate with survival time of patients, which is critical for cancer therapy.
- Published
- 2018
48. Simulating network intervention strategies: Implications for adoption of behaviour
- Author
-
Jennifer Badham, Ruth F. Hunter, and Frank Kee
- Subjects
medicine.medical_specialty ,030505 public health ,Operationalization ,Knowledge management ,Sociology and Political Science ,Social Psychology ,Computer science ,business.industry ,Communication ,Public health ,Psychological intervention ,Artificial networks ,Emotional contagion ,agent-based modelling ,03 medical and health sciences ,0302 clinical medicine ,SDG 3 - Good Health and Well-being ,Intervention (counseling) ,medicine ,Network interventions ,030212 general & internal medicine ,0305 other medical science ,business - Abstract
This study uses simulation over real and artificial networks to compare the eventual adoption outcomes of network interventions, operationalized as idealized contagion processes with different sets of seeds. While the performance depends on the details of both the network and behaviour adoption mechanisms, interventions with seeds that are central to the network are more effective than random selection in the majority of simulations, with faster or more complete adoption throughout the network. These results provide additional theoretical justification for utilizing relevant network information in the design of public health behavior interventions.
- Published
- 2018
49. Influential nodes ranking in complex networks: An entropy-based approach
- Author
-
Adel Fatemi, Amir Sheikhahmadi, and Ahmad Zareie
- Subjects
General Mathematics ,Applied Mathematics ,Topological information ,Artificial networks ,General Physics and Astronomy ,Statistical and Nonlinear Physics ,02 engineering and technology ,Complex network ,Information theory ,computer.software_genre ,01 natural sciences ,Evolving networks ,0103 physical sciences ,0202 electrical engineering, electronic engineering, information engineering ,Entropy (information theory) ,020201 artificial intelligence & image processing ,Data mining ,010306 general physics ,computer ,Mathematics - Abstract
Measurement of the spreading capability of nodes has been one of the most attractive challenges in the field of social networks. Because of the huge number of nodes in a network, it has appealed to many researchers to find an accurate measure which can potentially detect the spreading capability and rankings of nodes. Most of the available methods determine the spreading capability of nodes based on their topological locations. In this paper, however, we have proposed a new measure based on the basic notions in information theory to detect the spreading capability of nodes in networks on the basis of their topological information. The simulation and experimental results of investigating real-world and artificial networks show that the proposed measure is more accurate and efficient than the similar ones.
- Published
- 2017
50. Ranking influential nodes in social networks based on node position and neighborhood
- Author
-
Zhixiao Wang, Yan Xing, Changjiang Du, and Jianping Fan
- Subjects
Correctness ,Computational complexity theory ,Cognitive Neuroscience ,Artificial networks ,Monotonic function ,02 engineering and technology ,computer.software_genre ,01 natural sciences ,Computer Science Applications ,Artificial Intelligence ,0103 physical sciences ,0202 electrical engineering, electronic engineering, information engineering ,Entropy (information theory) ,020201 artificial intelligence & image processing ,Data mining ,010306 general physics ,computer ,Mathematics - Abstract
Ranking influential nodes of networks is very meaningful for many applications, such as disease propagation inhibition and information dissemination control. Taking multiple attributes into consideration is a hopeful strategy. However, traditional multi-attribute ranking methods have some defects. Firstly, the computational complexity of these methods is usually higher than O(n), inapplicable to large scale social networks. Secondly, contributions of different attributes are viewed as equally important, leading to the limited improvement in performance. This paper proposes a multi-attribute ranking method based on node position and neighborhood, with low computational complexity O(n). The proposed method utilizes iteration information in the K-shell decomposition to further distinguish the node position and also fully considers the neighborhood’s effect upon the influence capability of a node. Furthermore, the entropy method is used to weight the node position and neighborhood attributes. Experiment results in terms of monotonicity, correctness and efficiency have demonstrated the good performance of the proposed method on both artificial networks and real world ones. It can efficiently and accurately provide a more reasonable ranking list than previous approaches.
- Published
- 2017
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.