Search

This short communication provides a discourse emerged after reading “Stacking ensemble based deep neural networks modeling for effective epileptic seizure detection, Expert Systems with Applications, 148, 113239, 2020.” The discussed paper proposes a novel application of stacking-based ensemble for seizure detection, where several deep neural networks were used as base classifiers. The ensemble design and experimental results presented by the author show some weaknesses, which is indicated by, one of which, an inability of the proposed model to outperform previous studies. In this note, controversy of the discussed paper is explained and an improved version of stacking-based deep neural network is also further introduced and detailed to prevent it in the future.

This paper addresses the problem of how to represent education system measurement definitions and the data used to derive them. ISO 37120 is standard for measuring city performance. It defines 100 indicators of which seven focus on Education. As cities adopt the standard and begin publishing their indicator values, citizens, city bureaucrats and politicians will be able to see how well their education systems perform relative to other cities. Hence, education systems will be subject to greater scrutiny and will have to provide evidence-based explanations for their performance. In order to enable the automated analysis of city performance, four problems must be solved: 1) how are indicator definitions represented? 2) how is the data used to derive an indicator value represented? 3) how is educational system specific knowledge represented, and 4) how is city specific educational system knowledge represented? In this paper we present an Education Ontology designed to represent ISO 37120 education indicators and the information delineated above. The ontology is designed to enable an intelligent agent to perform consistency and root cause analysis of city indicators. It can also be used by cities to publish their education indicators and data, using Semantic Web standards, on their Open Data portals.

Advances in neuroimaging techniques have enabled early diagnosis and better understanding of neuro-psychological disorders using resting state - functional Magnetic Resonance Images (rs-fMRI). However, factors like the large intra-class variability, high homogeneity, limited sample size and varied acquisition methodologies in aggregated datasets have limited their classification performance. These issues especially are highly prevalent in the automatic diagnosis of Attention Deficit Hyperactivity Disorder (ADHD). In this paper, a Discriminative Spatial Filtering Method (DSFM) is developed to improve the separability between the two classes in the dataset by automatically identifying regions of the brain with abnormal activities. Further, DSFM uses an orthogonal projection to extract highly separable features from raw rs-fMRI data of the identified regions of the brain. Using a projection based learning classifier DSFM achieved a classification accuracy of 73.83 % on the publicly available ADHD200 dataset. The results show that, selective inclusion of data from the regions of the brain with discriminant activity identified using DSFM removes the redundant features and improved the classification performance (improvement of 3.5 % over baseline). Further, the brain activity maps derived by inverse mapping of the DSFM spatial filters describe the nature of the abnormal brain activities and identify the corresponding regions responsible for the cognitive symptoms in neuro-psychological disorders like ADHD. The results in this paper clearly show that DSFM is a reliable diagnostic tool with interpretable results critical for the accurate understanding the etiological factors of neuro-psychological disorders like ADHD using rs-fMRI.

Music radio data is currently underutilised in radio program management. Software tools that listen to and analyse music airplay are in many markets nonexistent, limited, or unaffordable. In this paper we present a novel knowledge discovery and visualisation framework for broadcast radio, ZeitMetric. The ZeitMetric framework uses machine learning and music information retrieval techniques to label radio audio automatically for knowledge discovery. The framework incorporates a novel music dataset collection technique (MusiGrab) to leverage online music services for ground-truth data, as well as a novel knowledge visualisation and presentation technique based on self-organizing maps (ZeitViz). The framework is compared to what little literature relating to this topic exists, and a set of requirements for a high-quality broadcast radio knowledge discovery is developed. MusiGrab specifically is compared to an existing static music information retrieval dataset and shown to offer superior results in this context. Future research directions using and extending the framework are also discussed. On acceptance of the paper, code for a use-case of the MusiGrab dataset collection technique will be released on GitHub.

Convolutional neural network is a prominent innovation in computer vision but is often troubled by problems such as dark light, turbidity, blur and high similarity to the background when applied to underwater object detection. Underwater object detection is one of the basic techniques of underwater grasping automation which plays a very important role in ocean detection and fishery of aquatic products. This paper presented an automatic detection method of underwater sea cucumber based on deep learning, which will provide effective technical support for the automated breeding and harvesting of sea cucumber. The Shortcut Feature Pyramid Network (S-FPN) proposed in this paper improves the existing multi-scale feature fusion strategy through shortcut connection. The ablation experimental results show that the mean average precision (mAP) of S-FPN reaches 91.5% which outperforms the baseline Feature Pyramid Network (88.6%), YOLO v3 (83.7%) and SVM-HOG (61.6%). To resolve the problem of complex environmental background interference of ocean floor, we proposed a Piecewise Focal Loss (PFL) function for balancing the positive and negative samples such that the algorithm can focus on the training difficulty of hard (i.e., positive) samples. And the ablation experimental results show that the mAP of PFL reaches 92.3% which outperforms the baseline Cross Entropy (91.5%) and Focal Loss (91.8%). Also, we chose Exponential Linear Unit as the optimization strategy, and Adaptive Moment Estimation as the activation function by ablation research, finally the mAP reached 94%.

The variety and plethora of research papers available on the Web motivated researchers to propose models that could assist users with personalized citation recommendations. In recent years, citation recommendation models using Network Representation Learning (NRL) methods have shown promising results. Nevertheless, existing NRL-based models are limited in terms of exploiting semantic relations and contextual information between the objects of bibliographic papers’ networks. Additionally, these models cannot adequately explore the structure of heterogeneous information networks, topical relevance, and relevant semantics. Consequently, they suffer from network sparsity and inadequate personalization problems. To overcome these shortcomings, we present a network embedding model termed as Global Citation Recommendation employing Generative Adversarial Network (GCR-GAN). The proposed model exploits the Heterogeneous Bibliographic Network (HBN) to generate personalized citation recommendations. In particular, the proposed model utilizes semantic relations corresponding to the objects of the heterogeneous bibliographic network and captures network structure proximity employing the Scientific Paper Embeddings using Citation-informed Transformers (SPECTER) and Denoising Auto-encoder networks to learn semantic-preserving graph representations. Compared to baseline models, the recommendations generated by our model over the DBLP and ACM datasets prove that it outperforms baseline methods by gaining almost 11% and 12% improvement in terms of Mean Average Precision (MAP) and Normalized Discounted Cumulative Gain (nDCG) metrics, respectively. Furthermore, we analyzed the effectiveness of the proposed model considering network sparsity issue, where our model gains almost 7% better recall@100 score against the second-best counterpart.

In the Authorship Attribution (AA) task, the most likely author of textual documents, such as books, papers, news, and text messages and posts are identified using statistical and computational methods. In this paper, a new computational approach is presented for identifying the most likely author of text documents. The proposed solution emphasizes lazy profile-based classification and, by using the Term Frequency-Inverse Document Frequency (TF_IDF) scheme, introduces a new measure for identifying important terms of documents. The importance of the terms is then used to calculate the similarity between an anonymous document and known documents. The proposed solution works with raw text documents and does not require any NLP tools for preprocessing, which makes it language-independent. The efficiency of the proposed solution has been evaluated by conducting several experiments on two English and Persian datasets, each of which contains six corpora with different number of authors. The obtained results demonstrate that the proposed solution outperforms state-of-the-art stylometric features, employed by seven well-known classifiers, by obtaining 0.902 accuracy for the English dataset and 0.931 accuracy for the Persian dataset. In addition, supplementary experiments have been conducted to evaluate the effects of documents’ length on the accuracy of the proposed solution, to examine the computation time of the proposed solution and competitive classifiers, and to identify the most effective stylometric features and classifiers.

Crowdsourcing as an important computing resource for the Crowd-based Cooperative, crowdsourcing workers, due to the complexity of their personnel composition and the ambiguity of group characteristics, make it difficult to accurately recommend suitable worker groups for the task. Recommending crowdsourcing tasks to suitable workers can greatly improve the efficiency and quality of the execution of crowdsourcing tasks. The current crowdsourcing recommendation algorithm is based on the single characteristics of the workers, ignoring the multi-dimensional characteristics of the workers and the community characteristics of the crowd. This paper proposes a worker group recommendation method based on multi-community collaboration (GRMBC) by utilizing the worker characteristics information extracted from crowdsourcing platform. Based on the characteristics of workers'reputation, preference and activity, the method divides the worker group into several characteristics communities with similar behavior to discover the potential multi-community structure in the crowdsourcing worker group, and then selects Top-N worker group for recommendation through the interaction among multi-communities. At the same time, we also proposed two mitigation strategies for the cold start problem of data in crowdsourcing. This paper uses the public data collected by AMT to do the experiments, and compares the aggregation results of the recommendations generated by different algorithms. The results show the recommendations generated by the GRBMC algorithm proposed in this paper performs the best comprehensively. When the GRBMC algorithm that considers both worker attributes and communi-ty characteristics has an accuracy improvement of 0.03 ~ 0.04 compared with not using the recommendation algorithm on each index, the accuracy of 0.01 ~ 0.02 is improved com-pared with the method considering single characteristic. Moreover, compared with the individual recommendation method of the workers, the recommendation method of the workers' group is more in line with the platform requirements, and can well reflect the wisdom of crowd wisdom.

In the field of medicine, decision-making has traditionally been carried out based on the best available scientific information and the experience of specialists using data found in analog formats such as radiographies, medical reports, and handwritten notes, among others. In this sense, the Big Data phenomenon is changing the world of medicine since the technologies that have been developed have made available to researchers and clinicians enormous amounts of data in digital formats that can be used to complement or help in complex tasks such as mentioned decision making. A key element in this process is data analysis techniques, since without them it is not possible to exploit the information. Currently the most used techniques are based on algorithms in the area of artificial intelligence and more specifically machine learning. This paper focuses on a specific domain of medicine, renal replacement therapies for end-stage renal disease, where machine learning is beginning to be used as a complementary tool to predict or make decisions. This paper provides a narrative review of the main machine learning methods that are being used to conduct end-stage renal disease treatment analyses.

This paper develops smoothing data association based on integrated probabilistic data association (FLIPDA-S) tracker to identify a multicopter UAV (MUAV) and estimate its trajectory in an uncertain and cluttered environment. Recently, a number of unidentified MUAVs have been flown in prohibited airspace and a radar system has been utilized to identify the MUAV and track the trajectory. However, target tracking methods could have difficulties to track the target MUAV through randomly distributed and moving objects. The vehicle state estimation (VSE) algorithm, which utilize FLIPDA-S in this paper, adopts UAV kinematics and dynamics for tracking a UAV in clutter and false targets with significant UAV identification performance. The performance of the method has been validated with and without an initial position of the target MUAV. Both numerical simulation and experiments are demonstrated to verify the effectiveness and accuracy of FLIPDA-S with VSE algorithm to track UAVs in an uncertain and clutter environment.

Healthcare waste (HCW) management is a complex and challenging problem. It is one of the priorities in health. An increase in the number of the health services provided leads to an increase in the amount of HCW, which has particularly been noticeable in recent years. Since this is the waste that may pose a risk to humans and the environment, it is necessary to ensure an adequate treatment of the same. HCW management is particularly important in developing countries, due to inappropriate disposal methods, underfunding and a lack of the infrastructure. In order to achieve the cost-effectiveness and sustainability of this area, HCW should be minimized through an adequate treatment of the same. The Public Enterprise Zdravstvo Brcko (Brcko Health System) has intensively been addressing the HCW management issue. They have decided to upgrade the HCW system by purchasing a new infectious waste treatment facility. The paper is aimed at creating a new original integrated multicriteria decision-making model based on D numbers for processing fuzzy linguistic information. This model will serve to support management in the procurement of the mentioned facility. The model integrates the benefits of different approaches and theories. An initial model was formed, consisting of the six potential solutions evaluated based on the 18 criteria classified into the following four groups: social, environmental, economic and technological. Four experts in this field evaluated the criteria and potential solutions. Then, a new Best-Worst Method based on D numbers (BWM-D) was applied in order to determine the significance of the criteria. After that, a Multi-Attributive Border Approximation Area Comparison Based on D numbers (MABAC-D) was developed and applied so as to evaluate and select an infectious waste treatment facility. The results have shown that the alternative A1 gives the best results, whereas the alternative A5 shows the worst results. Finally, a sensitivity analysis was performed to validate the obtained results. In this part of the paper, the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS-D) and VIsekriterijumska Optimizacija Kompromisnog Resenja (VIKOR-D) were developed in order to validate the results. When procuring a new Contagious Waste Treatment System, the characteristics of the available devices need be perceived and all the criteria need be taken into account in order to provide a device which will solve the HCW problem in the best way. This paper has shown how D numbers can be used when making a selection of an HCW management device, and also how all the characteristics of such a device can be perceived and how the device demonstrating the best characteristics can be selected.

Machine Learning (ML) is a branch of artificial intelligence that studies algorithms able to learn autonomously, directly from the input data. Over the last decade, ML techniques have made a huge leap forward, as demonstrated by Deep Learning (DL) algorithms implemented by autonomous driving cars, or by electronic strategy games. Hence, researchers have started to consider ML also for applications within the industrial field, and many works indicate ML as one the main enablers to evolve a traditional manufacturing system up to the Industry 4.0 level. Nonetheless, industrial applications are still few and limited to a small cluster of international companies. This paper deals with these topics, intending to clarify the real potentialities, as well as potential flaws, of ML algorithms applied to operation management. A comprehensive review is presented and organized in a way that should facilitate the orientation of practitioners in this field. To this aim, papers from 2000 to date are categorized in terms of the applied algorithm and application domain, and a keyword analysis is also performed, to details the most promising topics in the field. What emerges is a consistent upward trend in the number of publications, with a spike of interest for unsupervised and especially deep learning techniques, which recorded a very high number of publications in the last five years. Concerning trends, along with consolidated research areas, recent topics that are growing in popularity were also discovered. Among these, the main ones are production planning and control and defect analysis, thus suggesting that in the years to come ML will become pervasive in many fields of operation management.

Electric vehicles are environmental transportation modes that are widely applied in green logistics systems. To guarantee the energy efficiency, the impacts of customer service modes and recharging strategies need to be integrated into the optimization of electric logistics resource. This paper proposes an electric vehicle routing problem with mixed backhauls, time windows, and recharging strategies (EVRPMBTW-RS), minimizing the total travel cost with sophisticated constraints on the time-dependent pickup and delivery requests, limited recharging station capacity, and battery remaining capacity of electric vehicles. Mixed service sequences of linehaul and backhaul customers is allocated for the routing planning, with the synchronous optimization of recharging strategies including the selection of recharging stations and determination of recharging time. A time-discretized multi-commodity network flow model is constructed based on an extended space–time-state modeling framework, which is formulated as a quadratic 0–1 programming model by using the augmented Lagrangian relaxation technique. After the dualization and linearized transformation, we decompose the model into a sequence of least-cost path subproblems based on the alternating direction multiplier method (ADMM). The subproblems are alternately minimized and solved using the time-dependent forward dynamic programming algorithm. The solution quality can be guaranteed through calculating the optimality gap between the best lower bound and upper bound for each iteration. The proposed solution approach is examined on examples of a simple 7-node network and real-world Yizhuang road network. This paper provides a theoretical foundation for the route optimization method of electric logistics vehicles, and contributes to improve the operational efficiency of electric logistics systems.

Emergency medical services (EMS) are essential components for post-disaster rescue activities. Considering injury heterogeneity and deterioration over time of the casualties, the uncertainty in the number of the casualties can effectively improve the medical service performance. This paper develops a robust optimization model for combined facility location and casualty transportation under uncertainty in the number of casualties. We divide casualties into two types: mild casualties who are transported to on-site clinics with rescue vehicles and serious casualties who are transported to general hospitals with helicopters. Meanwhile, we consider the Injury Severity Score (ISS) increment to describe the injury deterioration of the casualties over time. The objective is to minimize the total weighted ISS increment of mild casualties and serious casualties. Then, we employ the robust optimization method to deal with the uncertainty and derive the robust counterpart of the proposed model in this paper. Case studies based on Yushu Earthquake show that the model can get the optimal emergency facility location and casualty transportation scheme to minimize the total weighted ISS increment. The total weighted ISS increment increases as the constraint violation probability decreases, which reflects the trade-off between performance and robustness. Sensitivity analyses show that the greater uncertainty of the casualty number is, the greater the impact on the total weighted ISS increment is, the more conservative the decision scheme is. The capacity of general hospitals has a greater effect on the objective value compared with the capacity of on-site clinics and the decision scheme of the robust optimization model has a greater optimality compared with the deterministic model when the problem size is magnified.

Customer purchase prediction aims to predict customers' future purchases, and the prediction results are of great importance for conducting future commercial activities. To obtain accurate predictions of customer purchases, this paper develops a machine learning framework based on historical behavioural data. First, considering the sparsity of behavioural data, this paper proposes a feature combination method based on the improved factorization machine algorithm. Second, due to the imbalance of customer purchase data, this paper proposes an imbalanced prediction method based on the maximized marginal category and cost-sensitive ensemble learning. Finally, a real-word travel service purchase dataset is adopted to test the feasibility of the proposed prediction framework. The experimental results and comparative analysis verify the validity of the proposed model.

Traditional social community discovery methods concentrate mainly on static social networks, but the analysis of dynamic networks is a prerequisite for real-time and personalized social services. Through the study of community changes, the community structure in a dynamic network can be tracked over time, which helps in the mining of dynamic network information. In this paper, we propose a method of tracking dynamic community evolution that is based on resistance distance. Specifically, we model the time-varying features of dynamic networks using the convergence of a resistance-based distance. In our model, the heterogeneity of neighboring nodes can be obtained in the local topology of nodes by analyzing the resistance distance between nodes. We design a community discovery algorithm that essentially discovers community structures on dynamic networks by identifying the so-called core node. During the process of community evolution analysis, both the dynamic contribution of ordinary nodes and core nodes in each community are considered. In addition, to avoid the inclusion of spurious communities in the community structure, we define the notion of noise community and account for it in our algorithm. Experimental results show that the method proposed in this paper can yield better accuracy than other existing methods.

The paper addresses the problem of trust management of cloud, fog and IoT services in dynamically changing environments. The continuous dynamic environment is one of the challenges that trustworthy services management in the cloud, fog and IoT settings faces. Services in such an environmental context have difficulty securing an acceptable quality of service (QoS). This article proposes a trust management framework that establishes service trust by considering the direct trust from the truster (subjective trust), aggregating referrals about the service in a collusion-resistant manner (objective trust), and bootstrapping new services. We introduce a subjective trust model based on the formalism of dependency networks to dynamically predict the provided QoS in response to context environment changes. The proposed approach leverages the dependency relations that exist among the QoS metrics and environmental context variables. The novelty at the subjective trust level lies in considering the dynamic cyclic dependency relations that enhances the prediction accuracy. However, subjective trust based on direct interactions could be insufficient to make the trust estimate credible. Hence, on top of the subjective layer, we propose an objective trust management model resilient to collusion attacks by leveraging the power of mass collaboration among referees. Finally, we propose a bootstrapping mechanism that is resilient to the white-washing attacks by observing the behaviours of newcomer services with no trust resources using the concept of social adoption to estimate their initial trust values. Experiments conducted on real-life and synthetic datasets demonstrate the effectiveness of our approach compared with state-of-the-art approaches. We used the statistical log score to assess the model’s prediction accuracy and employed the estimation error of the objective trust as indicated by referees relative to the one calculated by the multi-round simulation. The ROC (Receiver Operating Characteristic) curves are finally used to measure the accuracy of the classifier used in the proposed trust bootstrapping mechanism in providing accurate initial trust values. The main findings of the paper are around the new trust model of cloud, fog and IoT services considering their dynamically changing environments. The first finding is that the prediction of the provided QoS shows better results when it is dynamic and responds to context environment changes by leveraging the dynamic dependency network linking the QoS metrics and context variables of the environment. The second finding is that the objective trust performs better when it is resilient to collusion attacks by leveraging the power of mass collaboration among referees. The third finding is that the bootstrapping mechanism that observes the behaviours of new comer services with no trust resources using the concept of social adoption to estimate their initial trust values excels by being resilient to white watching attacks.

In computer vision, significant advances have been made on object detection with the rapid development of deep convolutional neural networks (CNN). This paper provides a comprehensive review of recently developed deep learning methods for small object detection. We summarize challenges and solutions of small object detection, and present major deep learning techniques, including fusing feature maps, adding context information, balancing foreground-background examples, and creating sufficient positive examples. We discuss related techniques developed in four research areas, including generic object detection, face detection, object detection in aerial imagery, and segmentation. In addition, this paper compares the performances of several leading deep learning methods for small object detection, including YOLOv3, Faster R-CNN, and SSD, based on three large benchmark datasets of small objects. Our experimental results show that while the detection accuracy on small objects by these deep learning methods was low, less than 0.4, Faster R-CNN performed the best, while YOLOv3 was a close second.

In the interest of developing an intelligent manufacturing environment with an agile, efficient, and optimally utilized transportation system, mobile robots need to achieve a certain level of autonomy as they play an important role in carrying out transportation tasks. Bearing this in mind, in the paper we propose a novel stereo visual servoing method for nonholonomic mobile robot control based on semantic segmentation. Semantic segmentation provides a rich body of information required for an adequate decision-making process in a clustered, dynamic, and ever-changing manufacturing environment. The innovative idea behind the new visual servoing system is to utilize semantic information of the scene for visual servoing, as well as for other mobile robot tasks, such as obstacle avoidance, scene understanding, and simultaneous localization and mapping. Semantic segmentation is carried out by exploiting fully convolutional neural networks. The new visual servoing algorithm utilizes an intensity-based image registration procedure, which results in the image transformation matrix. The transformation matrix encompasses the relations of images taken at the current and desired pose, and that information is directly used for visual servoing. The developed algorithm is deployed on our own developed wheeled differential drive mobile robot RAICO (Robot with Artificial Intelligence based COgnition). The experimental evaluation is carried out in the 3D simulation environment and in the laboratory model of the real manufacturing environment. The experimental results show that the accuracy of the proposed approach is improved when compared to the state-of-the-art approaches while being robust to the partial occlusions of the scene and illumination changes.

The novel of coronavirus (COVID-19) has suddenly and abruptly changed the world as we knew at the start of the 3rd decade of the 21st century. Particularly, COVID-19 pandemic has negatively affected financial econometrics and stock markets across the globe. Artificial Intelligence (AI) and Machine Learning (ML)-based prediction models, especially Deep Neural Network (DNN) architectures, have the potential to act as a key enabling factor to reduce the adverse effects of the COVID-19 pandemic and future possible ones on financial markets. In this regard, first, a unique COVID-19 related PRIce MOvement prediction ( COVID19 PRIMO ) dataset is introduced in this paper, which incorporates effects of social media trends related to COVID-19 on stock market price movements. Afterwards, a novel hybrid and parallel DNN-based framework is proposed that integrates different and diversified learning architectures. Referred to as the COVID-19 adopted Hybrid and Parallel deep fusion framework for Stock price Movement Prediction ( COVID19-HPSMP ), innovative fusion strategies are used to combine scattered social media news related to COVID-19 with historical mark data. The proposed COVID19-HPSMP consists of two parallel paths (hence hybrid), one based on Convolutional Neural Network (CNN) with Local/Global Attention modules, and one integrated CNN and Bi-directional Long Short term Memory (BLSTM) path. The two parallel paths are followed by a multilayer fusion layer acting as a fusion center that combines localized features. Performance evaluations are performed based on the introduced COVID19 PRIMO dataset illustrating superior performance of the proposed framework.

The chemometric analysis of chromatographic data is commonly used for discriminating bee propolis based on their geographical origin. Traditional machine learning methods for this purpose include principal component analysis and hierarchical clustering. When viewed as a time-series, the key discriminatory features in chromatographic data are the peaks, which should have similar location patterns for the propolis of the same origin. However, the peaks between same-origin samples are not always exactly aligned. Without proper alignment, samples from the same origin may be perceived by the clustering method to be very different. In this paper, we propose a novel dynamic time warping kernel principal component analysis (DTW-KPCA) method for the chemometric discrimination of propolis. The proposed method uses a Gaussian dynamic time warping kernel to measure the similarity between chromatographic time-series which incorporates time-series alignment. Results show a better clustering of propolis samples compared to the currently used methods, as measured by the silhouette coefficient. Hence, the proposed method enables a more reliable clustering as to the origin of propolis samples.

Recommender systems use intelligent algorithms to learn a user’s preferences and provide them relevant suggestions. Lack of sufficient ratings – also known as data sparsity problem – often results in poor recommendation performance. The existing recommendation methods have mainly focused on designing recommenders with high accuracy without paying much attention to the reliability of the recommendations. On the other hand, the users’ preferences may vary over time and considering the time factor in the design process is crucial, which has been largely ignored in most of the existing recommenders. To deal with these issues, a novel recommendation method is proposed in this paper which incorporates temporal reliability and confidence measures into the recommendation process. First, the effectiveness of the users’ rating is measured using a probabilistic approach and ineffective rating profiles are enriched by adding some implicit ratings to them. The quality of the predictions is evaluated using a temporal reliability measure taking into account the changes of users’ preferences over time. Then, the ratings with low reliability values are recalculated using a novel procedure, which updates the target user’s neighborhood by removing ineffective users. This leads to a temporal confidence measure that is used to update the neighborhood to provide more reliable and accurate recommendations. The superiority of the proposed method over state-of-the-art recommendation methods is shown by conducting extensive experiments on three benchmark datasets.

In the last decades, a wide portfolio of Feature Weighting (FW) methods have been proposed in the literature. Their main potential is the capability to transform the features in order to contribute to the Machine Learning (ML) algorithm metric proportionally to their estimated relevance for inferring the output pattern. Nevertheless, the extensive number of FW related works makes difficult to do a scientific study in this field of knowledge. Therefore, in this paper a global taxonomy for FW methods is proposed by focusing on: (1) the learning approach (supervised or unsupervised), (2) the methodology used to calculate the weights (global or local), and (3) the feedback obtained from the ML algorithm when estimating the weights (filter or wrapper). Among the different taxonomy levels, an extensive review of the state-of-the-art is presented, followed by some considerations and guide points for the FW strategies selection regarding significant aspects of real-world data analysis problems. Finally, a summary of conclusions and challenges in the FW field is briefly outlined.

In this paper, a knowledge-based differential covariance matrix adaptation cooperative algorithm (DCMAC) is proposed for continuous problems. On the basis of combining successful history-based adaptive DE variants with linear population size reduction (LSHADE) and covariance matrix adaptation evolutionary strategy (CMA-ES), DCMAC proposes a strategy based on knowledge reward and punishment to achieve the purpose of collaborative optimization. Afterward, an adaptive learning mechanism is introduced to optimize the parameters to balance the exploitation and exploration of DCMAC. This process enables the algorithm to have global search capability. Finally, the niching-based population size reduction mechanism is introduced to improve the local search ability of the DCMAC. A weighted mutation strategy with dynamic greedy p value and covariance matrix adaptation (CMA) sampled based on differential vector are presented. Meanwhile, the knowledge acquired in the previous iteration process is adopted in the algorithm to select a mutation strategy for generating the new candidate solutions in the next iteration. The niching population size reduction mechanism is introduced to maintain the diversity of the population and compared with the other classical population size reduction methods. The optimal combination of parameters in the DCMAC algorithm is testified by the design of the experiment. Furthermore, the DCMAC is testified on the CEC2017 benchmark functions. The effectiveness and efficiency of the DCMAC are demonstrated by the experimental results in solving complex continuous problems.

The objective of the paper is to detect sarcasm in human communication. The methodology uses basic cognitive features of human utterances by capturing three modes of data viz., voice, text, and temporal facial features. The captured data is unstructured as it consists of parameters of feelings and emotions to generate sarcasm which affects expressions through glottal and facial organs. The data capturing method is equally challenging as compared to the method of data processing to acquire features. The significant work is aligned to make natural decisions in the prediction processes using cognitive information in the data lineage. Sarcasm detection in natural human communication is a challenging process. The Linguistic features of natural language processing (NLP) methods help identify sentiment as negative and positive sentences based on polarity using the pre-labelled samples. The multiclass neural network model is used as a soft cognition method for the detection of sarcasm under cloud resources. Identified cognitive features have information like voice cues and eye movements, they tend to influence the decision of detecting sarcasm. The visual data are found to be quite interesting and can establish a strong platform in the area of NLP for further research work.

In this era of big data, processing large scale data efficiently and accurately has become a challenging problem. Ensemble classification is a type of supervised learning that uses multiple experts to generate the final output. It provides a way to classify data more accurately. As a result of using multiple classifiers, they are often more complicated than single classifiers, especially for big data problems. Apache Spark is a unified analytics engine for big data processing which provides a scalable framework to analyze the data. In this paper, we first extend our previous work and design a distributed heterogeneous ensemble classifier inspired by the boosting approach, which is capable of dealing with big datasets. Using heterogeneous classifiers makes it possible to have more diverse classifiers, and consequently, a more accurate classifier is obtained. Then, we present the Spark version of the proposed approach to speed up our heterogeneous ensemble classifier using the MapReduce paradigm. In order to evaluate our approach, we have applied it to seven big datasets. Extensive experimental results indicate the superiority of the proposed method over the existing ensemble algorithms implemented by Spark MLlib in terms of the classification accuracy, performance, and scalability.

Image clustering is one of the most significant problems in computer vision and data mining. To mitigate the influence brought by appearance variation, many scholars attempt to cluster images with multiple features, a.k.a, multi-view image clustering. However, the majority of existing methods simply consider context information across images or content information within images, failing to combine both sides. We in this paper propose a novel auto-weighted multi-view content-context information bottleneck (AMC2IB) method for image clustering. The AMC2IB method can simultaneously utilize the content and context information for partitioning images. The “content” characterizes the intrinsic information within each image, e.g., appearance feature of color or shape; The “context” describes the close relationships among images of each view, e.g., inter-image similarity. The mechanism of maximum entropy is also introduced to automatically learn the view weight, and thus the importance of different views can be integrated for effective clustering. Additionally, to remove the extra weight regularization parameter in AMC2IB, we further propose an auto-weighted multi-view content-context information bottleneck without weight regularization (AMC2IBW) method. Afterwards, the above problems can be formulated as information loss functions by maintaining the context and content information maximally when the input images are compressed. Eventually, a new alternating iterative method is designed for the optimization of both proposed objective functions. Experimental results on five real-world multi-view image datasets show the superiority of the proposed method.

The unidentified number of unmanned aerial vehicles (UAVs) can execute aggressive maneuvers in the restricted and the cluttered environment. Therefore, it is difficult to detect and track the uncertain motion of the UAV target in such complex environment. In addition, multi-target tracking (MTT) algorithms such as joint data association approach faces various computational complexities that could exceeds the available computation resources. This paper develops a novel smoothing data association idea in a linear multi-target (LM) tracking based on integrated probabilistic data association (sLM-IPDA) algorithm that acts like a single target tracker in the MTT situation. The significant detection and tracking performance of a UAV are validated without a-prior information of the UAV’s initial position. The forward and backward tracks are initialized separately using sensor measurements received in each scan. The sLM-IPDA estimates the backward multi-tracks simultaneously associating backward tracks in a subsequent predicted forward track for fusion. Thus, a forward track state estimate is obtain using the smoothing (fusion) measurements. This significantly improves estimation accuracy for large number of cross-over targets in heavy clutter. Numerical assessments of the sLM-IPDA are verified using both simulation and experiment to demonstrate the application of the proposed algorithm.

Moth–Flame Optimization (MFO) algorithm is widely used to solve real parameter global optimization problems. However, the original MFO algorithm has some problems. First, the global search ability of the algorithm is insufficient due to the lack of high-quality flames or the lack of diversity of the population. Moreover, due to the reduction of the number of flames mechanism and the simple transverse orientation flight mode of moths, the algorithm is easy to fall into local optimum. To overcome these difficulties, a new MFO called ODSFMFO is proposed in this paper, which consists of an improved flame generation mechanism based on Opposition-Based Learning (OBL) and Differential Evolution (DE) algorithm, and an enhanced local search mechanism based on Shuffled Frog Leaping Algorithm (SFLA) and death mechanism. The ODSFMFO firstly obtains a high-quality population through opposition-based learning, then enhances the diversity of the population through differential evolution algorithm, to improve the global search ability of the algorithm, and then, employs an improved shuffled frog leaping algorithm as a local search algorithm, and uses the death mechanism to eliminate the individuals with low fitness value, which not only make the algorithm easy to jump out of the local optimum, but also further improve the convergence speed. To demonstrate the effectiveness of the proposed algorithm, the ODSFMFO and other fourteen well-known optimization algorithms are tested and compared on 28 benchmark tasks of CEC 2013 and 30 benchmark tasks of CEC 2017. Simulation results show that the proposed algorithm obtains better results compared with other competitive algorithms.

When reading an article, especially a professional article, we often encounter words or phrases that we don't recognize. They may be specific domain terms or emerging entities. When we can't guess their meaning from the article, we generally refer to the terminology dictionary or search for related content on the Internet to understand them. Some researchers have used natural language generation (NLG) models to explain these unknown phrases in recent years automatically. Still, current NLG models have difficulties generating long sentences with good coherence, and they are difficult to generate multiple sentences that describe unknown phrases from different angles. Therefore, this paper proposes a model that can judge whether an existing sentence can explain a certain phrase, called TransExplain. TransExplain can use LSTM, convolutional neural network, and attention mechanism to extract multiple sentence features and map them to a fixed-dimensional semantic feature vector. By calculating the cosine similarity between the semantic feature vector and the unknown phrase vector, it is judged whether the sentence can explain the semantics of the unknown phrase. And a loss function called positive and negative means square error is introduced to improve the model's ability that distinguishes negative examples. For this task, we provided a Chinese dataset containing phrases and explanation pairs in 7 important domains. On this dataset, TransExplain can achieve better results than previous similar tasks.

In this paper a semi-supervised method for the detection of anomalies in both texture- and object-based product images is presented. The method exploits a pre-trained Convolutional Neural Network (CNN) autoencoder that is blended with a statistical-based transformation of the neural network embedding layer in order to remove anomalies from the input image. The “cleaned” version of the input image is then compared with the input image itself in order to spatially localize the anomalies. The method does not require a specific training of the CNN to be applied to a new class of product, but it requires a very fast domain adaptation based on only “anomaly-free” examples. Experiments conducted on a publicly available dataset made of fifteen texture- and object-based classes show that overall performance is better than the state of the art of about 4%. In the case of texture-based classes the proposed method outperforms the state of the art of about 13%. In the case of object-based classes, the proposed method reaches overall the same performance of the state of the art. In this case, apart from 3 cases, that is “bottle”, “transistor” and “metal nut”, the proposed method performs better than the state of the art in 7 object classes out of 10.

Learning the distributed representation of a sentence is a fundamental operation for a variety of natural language processing tasks, such as text classification, machine translation, and text semantic matching. Tree-structured dynamic compositional networks have achieved promising performance in sentence representation due to its ability in capturing the richness of compositionality. However, existing dynamic compositional networks are mostly based on binarized constituency trees which cannot represent the inherent structural information of sentences effectively. Moreover, syntactic tag information, which is demonstrated to be useful in sentence representation, has been rarely exploited in existing dynamic compositional models. In this paper, a novel LSTM structure, ARTree-LSTM, is proposed to handle general constituency trees in which each non-leaf node can have any number of child nodes. Based on ARTree-LSTM, a novel network model, Tag-Enhanced Dynamic Compositional Neural Network (TE-DCNN), is proposed for sentence representation learning, which contains two ARTree-LSTMs, i.e. tag-level ARTree-LSTM and word-level ARTree-LSTM. The tag-level ARTree-LSTM guides the word-level ARTree-LSTM in conducting dynamic composition. Extensive experiments demonstrate that the proposed TE-DCNN achieves state-of-the-art performance on text classification and text semantic matching tasks.

Efficient and accurate semantic segmentation is particularly important in scene understanding for autonomous driving. Although Deep Convolutional Neural Networks(DCNNs) approaches have made a significant improvement for semantic segmentation. However, state-of-the-art models such as Deeplab and PSPNet have complex architectures and high computation complexity. Thus, it is inefficient for realtime applications. On the other hand, many works compromise the performance to obtain real-time inference speed which is critical for developing a light network model with high segmentation accuracy. In this paper, we present a computationally efficient network named DSANet, which follows a two-branch strategy to tackle the problem of real-time semantic segmentation in urban scenes. We first design a Semantic Encoding Branch, which employs channel split and shuffle to reduce the computation and maintain higher segmentation accuracy. Also, we propose a dual attention module consisting of dilated spatial attention and channel attention to make full use of the multi-level feature maps simultaneously, which helps predict the pixel-wise labels in each stage. Meanwhile, Spatial Encoding Network is used to enhance semantic information and preserve the spatial details. To better combine context information and spatial information, we introduce a Simple Feature Fusion Module. We evaluated our model with state-of-the-art semantic image semantic segmentation methods using two challenging datasets. The proposed method achieves an accuracy of 69.9% mean IoU and 71.3% mean IoU at speed of 75.3 fps and 34.08 fps on CamVid and Cityscapes test datasets respectively.

Recently, many intelligent algorithms have been proposed to find the best solution for complex engineering problems. These algorithms can search volatile and multi-dimensional solution spaces and find optimal answers timely. In this paper, a new meta-heuristic method is proposed that inspires the behavior of the swarm of birds called Coot. The Coot algorithm imitates two different modes of movement of birds on the water surface: in the first phase, the movement of birds is irregular, and in the second phase, the movements are regular. The swarm moves towards a group of leading leaders to reach a food supply; the movement of the end of the swarm is in the form of a chain of coots, each of coot which moves behind its front coots. The algorithm then runs on a number of test functions, and the results are compared with well-known optimization algorithms. In addition, to solve several real problems, such as Tension/Compression spring, Pressure vessel design, Welded Beam Design, Multi-plate disc clutch brake, Step-cone pulley problem, Cantilever beam design, reducer design problem, and Rolling element bearing problem this algorithm is used to confirm the applicability of this algorithm. The results show that this algorithm is capable to outperform most of the other optimization methods. The source code is currently available for public from: https://www.mathworks.com/matlabcentral/fileexchange/89102-coot-optimization-algorithm .

The Electrocardiogram (ECG) arrhythmia classification has become an interesting research area for researchers and developers as it plays a vital role in early prevention and diagnosis of cardiovascular diseases. In ECG signal classification, the feature extraction and selection processes are critical steps. Thus, in this paper, different ECG signal descriptors based on one-dimensional local binary pattern (LBP), wavelet, higher-order statistical (HOS), and morphological information are introduced for feature extraction. For feature selection and classification processes, a new hybrid ECG arrhythmia classification approach called MRFO-SVM that combines a metaheuristic algorithm termed Manta ray foraging optimization (MRFO) with support vector machine (SVM) is proposed to automatically determine the relevance features of LBP, HOS, wavelet and magnitude values. In MRFO-SVM approach, the MRFO is utilized to optimize the parameters of SVM and to select the significant features subset that provides the best classification performance, meanwhile SVM is used for classification purposes. The proposed MRFO-SVM approach is trained on the MIT-BIH Arrhythmia database containing four abnormal and one normal heartbeats. The experimental results of ECG arrhythmia classification using the proposed MRFO-SVM revealed with evidence its superiority with overall classification accuracy of 98.26% over seven well-known metaheuristic algorithms.

Feature selection aims at finding the minimum number of features that result in high classification accuracy. Accordingly, the feature selection is considered as a multi-objective problem. However, most existing approaches treat feature selection as single-objective problem, and they are often divided into two main categories: filter and wrapper methods. Filters are known as fast methods but less accurate, while wrappers are computationally expensive but with high classification performance. This paper proposes a novel hybrid filter-wrapper feature selection approach using whale optimization algorithm (WOA). The proposed method is a multi-objective algorithm in which a filter and wrapper fitness functions are optimized simultaneously. Our algorithm’s efficiency is demonstrated through an extensive comparison with seven well-known algorithms on twelve benchmark datasets. Experimental results show the ability of the proposed algorithm to obtain several subsets that include smaller number of features with excellent classification accuracy.

Hearing impaired individuals can easily overcome the barriers in communicating with other members of the society via computer technology. In this study, the recognition of dynamic words in Turkish Sign Language (TSL) with two hands was studied using the Leap Motion Controller (LMC) device. 50 dynamic words were determined considering the similarities and differences among themselves, and a dataset was created using 4 signers. For the system proposed in this paper, a comprehensive feature extraction process is executed. Applying feature selection algorithms and PCA, LDA and PCA+LDA dimension reduction methods to this dataset, new datasets with less dimension were obtained. For the first time, ELM architectures were operated as a classifier in a sign language recognition system. Recognition performance was tested with 5 different ELM networks and 2 classical classifiers and the results were compared. In addition, comparisons of classical and ELM based classifiers were presented. The 10-fold cross validation method was used to test the validity of the proposed system and the accuracy of the results obtained. Based on the results obtained by a comprehensive analysis, it was observed that the ML-KELM classifier maintains its performance rate and gives the highest performance rate. At the same time, it has been observed that ML-KELM classifier has a stable structure, which offers less user intervention.

Parkinson’s Disease (PD) is a progressive neurological disorder that affects the functioning of the brain. As PD progresses, there arises a problem of maintaining coordination between brain and different parts of the body. Subjects have the problem of performing daily activities. Its early diagnosis is important to improve the quality of life of Parkinson’s patients. Most of Parkinson’s patients are likely to have voice disorders in the early stages. Therefore, the analysis of voice recordings with machine learning-based models can improve the diagnosis process. Voice signals are non-linear, non-stationary signals which exhibit oscillatory behavior. In this paper, a hybrid approach is proposed to extract features from resonance-based and time-frequency based information. A combination of Resonance based Sparse Signal Decomposition (RSSD) + Time-Frequency (T-F) algorithm is proposed. Two datasets are collected for the study. The first dataset D1 is a public dataset consisting of 16 PD + 21 Healthy Controls (HC) and the second dataset D2 is collected from 20 HC to explore the effect of diversity in the diagnosis process. To explore the potential of deep learning techniques in diagnosing speech impairments in PD patients, the potential of Convolution Neural Network (CNN) is also explored. To evaluate the method, we conducted several experiments with state-of-the-art feature extraction and classification techniques. The proposed hybrid approach distinguished PD and HC with a validation accuracy of 99.37%. We have also explored the effect of diversity in the data collection process and it is observed that diversity in the socio-cultural group plays very important role to put the system in the clinical practice.

Knowledge Distillation has been established as a highly promising approach for training compact and faster models by transferring knowledge from more heavyweight and powerful models, so as to satisfy the computation and storage requirements of deploying state-of-the-art deep neural models on embedded systems. However, conventional knowledge distillation requires multiple stages of training rendering it a computationally and memory demanding procedure. In this paper, a novel single-stage self knowledge distillation method is proposed, namely Online Subclass Knowledge Distillation (OSKD), that aims at revealing the similarities inside classes, improving the performance of any deep neural model in an online manner. Hence, as opposed to existing online distillation methods, we are able to acquire further knowledge from the model itself, without building multiple identical models or using multiple models to teach each other, rendering the OSKD approach more effective. The experimental evaluation on five datasets indicates that the proposed method enhances the classification performance, while comparison results against existing online distillation methods validate the superiority of the proposed method.

Implementing ERP (Enterprise Resource Planning) system successfully comes with a set of costly and complex challenges. The purpose of this paper is to offer an effective cloud-based ERP software evaluation framework to gain a better understanding of these complexities and address such challenges by identifying the selection criteria. The PFS (Pythagorean Fuzzy Set) is the new extension of Intuitionistic fuzzy sets, the Simple Additive Weighting (SAW) is an easy-to-use technique to determine the weights of the criteria, and the VIKOR technique is a compromise ranking methodology. In the recommended novel integrated MCDM (Multi-Criteria Decision-Making) methodology, Decision-Makers (DMs) must choose the best one among a set of candidate alternatives that satisfies the evaluation criteria. The GDM (Group Decision Making) is preferred here to deal with hesitancy in the evaluation of alternatives under group preferences. The sensitivity analysis resulted from various scenarios reveals that the proposed method is quite sensitive to criteria weights. The service and support, cost, and functionality are found as the essential evaluation criteria for the Turkish firm. These findings emphasize the significance of a qualified group of DMs for the evaluation process. The case study findings demonstrate that the developed approach is practical and validated for solving software development projects. This study presents a novel integrated methodology based on the PFS, SAW, and VIKOR to measure the performance of software development projects, to deal with the ambiguity and imprecision caused by the subjective judgments of DMs. The research contributes to the literature by delivering a combined decision methodology for the first time and by applying it to evaluate the software development projects.

Previous methods for representing scene images based on deep learning primarily consider either the foreground or background information as the discriminating clues for the classification task. However, scene images also require additional information (hybrid) to cope with the inter-class similarity and intra-class variation problems. In this paper, we propose to use hybrid features in addition to foreground and background features to represent scene images. We suppose that these three types of information could jointly help to represent scene image more accurately. To this end, we adopt three VGG-16 architectures pre-trained on ImageNet, Places, and Hybrid (both ImageNet and Places) datasets for the corresponding extraction of foreground, background and hybrid information. All these three types of deep features are further aggregated to achieve our final features for the representation of scene images. Extensive experiments on two large benchmark scene datasets (MIT-67 and SUN-397) show that our method produces the state-of-the-art classification performance., Comment: Submitted to Expert Systems with Applications (ESWA), 28 pages and 17 images

In recent years, the development of algorithms that assist in communicate with deaf people is an important challenge. The development of automatic systems to translate sign language is a current research topic. However, this involves several processes that range from video capture, pre-processing to identification or classification of the signal. The development of systems capable of extracting discriminative features that enhance the power of generalization of a classifier is even a very challenging problem. The meaning of a sign is the combination of the hand movement, hand shape, and the point of contact of the hand in the body. This paper presents a method to detect and translate hand gestures. First, we obtain 15 frames per word, obtaining 3 regions of interest (hands and face) from which we obtain geometric features. Finally we use several classifier techniques and present the experimental results.

Learning from data stemming from real-world problems is inherently challenging and difficult due to the numerous intrinsic characteristics present in datasets. The problem of class imbalance is known to significantly impair classification performance and has attracted increasing attention from researchers. On the other hand, some studies suggest that the detrimental effects of class imbalance occur only when the dataset encompasses other intrinsic characteristics such as small disjuncts, class overlapping, noise or data rarity. However, the literature is often ambiguous in terms of understanding and distinguishing the influence of these characteristics on the behaviour of standard classification algorithms. This paper provides a contemporary empirical study of the behaviour and performance of five well-known classifiers on a large number of imbalanced datasets exhibiting numerous combinations of the stated characteristics. The aim of the study is to identify and rank difficulty factors when learning from imbalanced data, depending on the type of classification algorithm used. In general, the obtained results suggest that if classifiers conceptually have no problem with class separation into sub-concepts, noise is the characteristic that most impairs their performance, closely followed by class overlapping and class imbalance. To alleviate these problems, oversampling and undersampling procedures were tested and directions are given for selecting appropriate techniques when dealing with the problem of class imbalance.

Real-time vehicle type classification has become more popular and significant in recent years owing to its potential applications. In this paper, a real-time Convolutional Neural Network is proposed for vehicle type classification. This novel CNN structure is a combination of classic CNN layers along with squeeze-and-excitation (SE) modules and Haar wavelet as the pooling layer. This structure improves the performance of the CNN classifier by emphasizing informative feature maps and decreasing the entropy of the network. A cross-entropy loss function is proposed for better performance. A new pooling method is also introduced based on the Haar transform. Network parameters and the number of layers are selected in such a way to be real-time. Experimental results on two vehicle datasets show that the overall performance of this model, including recognition time and accuracy, is better than the others. For example, using DWT instead of Max-pooling improved the recognition rate on the IRVD dataset from 97.12% to 99.06%. The number of training parameters of the proposed model is about 5 million which is much less than popular networks like VGG (128 m), ResNet152 (58 m), DarkNet (40 m), and Inception-V3 (24 m). This made it very faster so that its recognition time is only 42 ms on CPU which is suitable for real-time applications.

Intelligent system applications in computer vision suffer detection and identification problems in ill lighting conditions (i.e., non-uniform illumination), where under-exposed and over-exposed regions coexist in the captured images. Processing on these images results in over and under enhancement with colour and contrast distortions. The traditional methods design some handcrafted constraints and rely on image pairs and priors, whereas existing deep learning-based methods rely on large scale and even paired training data. But these method’s capacity is limited to specific scenes (i.e., lighting conditions). In this paper, we present a deep-hybrid ill-light image enhancement method and propose a contrast enhancement strategy based on the decomposition of the input images into reflection J and illumination T. A Divide to Glitter network (D2G-Net) is designed to learn from the few-shots of training samples and do not require paired and large quantity training data. D2G-Net is comprised of a multilayer Division-Net for image division and a Glitter-Net to amplify the illumination map. We propose to regularize learning using a correlation consistency of decomposition extracted from the input data itself. Extensive experiments are organized under ill-lighting conditions, where a new test dataset is also proposed with robust lighting variation to evaluate the performance of the proposed method. Experimental results prove that our method has superior performance for preserving structural and texture details compared to state-of-the-art approaches, which suggests that our method is more practical in interactive computer vision and intelligent expert system applications.

World health organization's estimates reveal that air pollution kills almost 6.5 million people in the world every year. As human beings, on average, spend 80–90% of their routine time in closed spaces, indoor air pollution has been a prime concern for their health and well-being. Primary and secondary pollutants are present in indoor environments, and they leave a considerable impact on human health. However, PM10 has attracted special scientific and legislative attention due to its close association with chronic health problems such as respiratory illness, lung cancer, and asthma attacks. Therefore, it is critical to develop a reliable model to analyze PM10 pollutants in the indoor environment so that building occupants can take relevant preventive measures. This paper focuses on the monitoring and predicting PM10 pollutant concentration in the indoor environments using the Fuzzy Inference System Tree (FIST) model. The forecasting model was trained using five different input parameters (PM2.5, CO2, VOC, temperature, and humidity) while considering PM10 as the target variable. The system performance was measured in terms of four performance indicators where MSE = 1.8126; MAE = 1.1821; MAPE = 4.4372%; RMSE = 1.3463 using normalized data. The model performance was further improved using Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). Results show that the proposed aggregated FIS optimized with GA (RMSE = 1.2101) outperformed the PSO (RMSE = 1.2202) based model in terms of performance indicators. The proposed model can be installed in real-time environments to forecast PM10 concentration for improved public health and well-being.

The Extreme learning machine (ELM) designed by Huang et al. is proved to be a fast and good classifier over a decade, but existing ELM is non-deterministic in nature as well as kernel dependent and needs attention to optimize the selection of kernels. In ELM feature space is obtained with the help of single kernel function. The choice of kernel depends on perceptiveness of classification problem. So a generalized framework with deterministic nature along with optimized kernel is ought to be designed that can be applied to large domain of real world heterogeneous pattern classification problems. This paper presents a deterministic extreme learning machine for neural network with feedforward architecture which is formulated with multiple kernel learning. We further enhance this approach by incorporating Gray level co-occurrence matrix (GLCM) for multi-modal feature extraction. Two formulation of kernel extreme learning machine are introduced, with target kernel function as a linear combination of different base kernels. The first one is based on deterministic multiple kernel learning while the second one uses deterministic multiple kernel learning along with GLCM for extracting the invariant feature vectors. The performance of proposed algorithms are analyzed on pattern recognition problem for face recognition by changing the number of training set, types of kernel used and coefficients used for combining base kernels. The superior recognition rate is achieved for prominent multi-class face databases, when compared with contemporary methods that prove the efficacy of proposed algorithms.

One of the most studied application areas of intelligent systems is the classification area, and this is because classification covers a wide range of real-world problems. Some examples are fault-diagnosis, image segmentation, medical diagnosis, among others. In most cases, the intelligent systems designed for the solution of this kind of problems are based on supervised learning, which is based on learning how to classify with previous datasets for finding relations between the inputs and outputs. The main focus of the present paper is the supervised generation of general type-2 fuzzy classifiers with a new strategy for modeling data uncertainty. The proposed methodology includes a mix of concepts, such as the use of embedded type-1 membership functions, statistical concepts such as the quartiles, and nature inspired optimization methods. The classifiers generated with the proposed methodology are compared with respect to other general type-2 fuzzy classifiers based on symmetric uncertainty to evaluate their performance, in this way obtaining interesting results for medical diagnosis with benchmark data sets.

The technology improvement has radically changed how tourists perform their journey offering new services to enhance their cultural experience and to easily retrieve required information. In this paper we propose a novel framework, that models intangible and tangible cultural objects into a unified data model, for supporting tourists journey. In particular, a Micro-service architecture has been designed to provide several services whose tourists can access through a conversational agent based on the Seq2Seq model. Furthermore, we propose also an Enterprise Service Bus to ingest events automatically from promotional website or manually from public and/or private organization. We have evaluated the proposed framework according to the following two experiments: (i) the efficiency and efficacy of the Chatbot engine, showing that the use of GRU cells allows to obtain better results in terms of loss and accuracy with respect to LSTM one and (ii) the efficacy of the proposed framework according to the NASA-TLX asking to 50 users to interact with the Chatbot, demonstrating that the proposed approach outperforms the state-of-the art ones combining natural processing language, user profile information and historical activities.

As the intersection of finance and statistics, financial surveillance is a new interdisciplinary field of research. In this field, statistical process control methods are applied to monitor financial indices. The final aim is to detect out-of-control conditions and trigger a signal as soon as possible. These early signals can help practitioners in making on-time decisions. In this paper, a new method based on a support vector machine is proposed to detect upward and downward shifts with step and trend patterns in auto-correlated financial processes. These processes are modeled by the autoregressive moving average (ARMA) and generalized autoregressive conditional heteroskedasticity (GARCH) variance (ARMA-GARCH) time series model. Autocorrelation structure in data with changing volatility makes pattern recognition difficult. As such, some features are selected to extract different properties of the considered patterns. Moreover, a new feature named the maximum degree between the horizontal line and the consecutive observations line is used for the better distinction of the step and trend shift patterns. To evaluate the performance of the proposed method, we performed a comprehensive simulation study. Moreover, to illustrate the proposed method's application, it was applied to classify patterns in the OPEC crude oil basket return and the Tehran stock exchange index.