Search

In this paper, we describe a review concerning the Quantum Computing (QC) and Deep Learning (DL) areas and their applications in Computational Intelligence (CI). Quantum algorithms (QAs), engage the rules of quantum mechanics to solve problems using quantum information, where the quantum information is concerning the state of a quantum system, which can be manipulated using quantum information algorithms and other processing techniques. Nowadays, many QAs have been proposed, whose general conclusion is that using the effects of quantum mechanics results in a significant speedup (exponential, polynomial, super polynomial) over the traditional algorithms. This implies that some complex problems currently intractable with traditional algorithms can be solved with QA. On the other hand, DL algorithms offer what is known as machine learning techniques. DL is concerned with teaching a computer to filter inputs through layers to learn how to predict and classify information. Observations can be in the form of plain text, images, or sound. The inspiration for deep learning is the way that the human brain filters information. Therefore, in this research, we analyzed these two areas to observe the most relevant works and applications developed by the researchers in the world. [ABSTRACT FROM AUTHOR]

Computational intelligence (CI) has become one of the most useful and successful tools for dealing with uncertainties and complex problems in many fields, such as neural networks, genetic algorithms, and swarm intelligence, artificial intelligence, risk management, financial monitoring, etc. With the development of CI, abundant publications have arisen related to many research directions and hotspots. Based on the technical support from bibliometrics and the corresponding approach as well as the content analysis, this study conducts a science mapping analysis and a coherent knowledge picture of the research field in CI. The research contributes to clear future development trends and provides more ideas for scholars in this field. First, this paper focuses on the fundamental characteristics of CI publications, including annual numbers, term co-occurrence, and hot research directions, as the preliminary exploration of this field. Then, according to the widely used core database, i.e., Web of Science (WoS), and the technologies of software, VOS viewer, and CiteSpace, the productive institutions, authors, and journals are explored. Next, the corresponding internal characteristics of the CI research are analyzed, including the citation features of countries/regions, institutions, journals and authors. Furthermore, to analyze the development trend of research hotspots, the keywords of all CI publications are studied: (a) classifying them into three phases in chronological order aimed; (b) implementing the burst detection algorithm to intuitively reflect the scientific research in the field of CI. Finally, this paper provides a relatively throughout perspective for the CI articles and reviews and discloses the future development trend, which will help the scholars interested in this area conduct deep research. We conclude this bibliometric overview with the limitations and recommendations for future research in the field of CI. [ABSTRACT FROM AUTHOR]

The paper initially proves that locally finite covering (LFC-, for short) rough set structures are interior and closure operators. To be precise, given an LFC-space (U , C) , we prove that the lower H-rough set operator H ∗ is an interior operator and the upper H-rough set operator H ∗ is a closure operator. Besides, we prove a duality of the concept approximations (H ∗ , H ∗) and investigate many theoretical and mathematical properties of the H-rough set operators. After pointing out that Khalimsky (K-, for brevity) topological rough set operators have their own features, we prove that the K-topological lower (resp. upper) approximation operator is not an interior (resp. closure) operator from the viewpoint of K-topology. Besides, we intensively investigate theoretical and mathematical properties of the K-topological rough set operators. This research area can be considered as a part of geometric-based rough set theory. These obtained results can promote the studies of rough set theory associated with information geometry, object classification, artificial or computational intelligence, and so on. In the present paper, each of the sets U, C and X (⊆ U) need not be finite. [ABSTRACT FROM AUTHOR]

Learning to rank (LTR) is one of the problems attracting researchers in information retrieval (IR). The LTR problem refers to ranking the retrieved documents for users in search engines, question answering and product recommendation systems. There is a number of LTR approaches based on machine learning and computational intelligence techniques. Most existing LTR methods have limitations, such as being too slow or not being very effective or requiring a huge computer memory to operate. This paper proposes a LTR method that combines a (1+1)-evolutionary strategy with machine learning. Three variants of the method are investigated: ES-Rank, IESR-Rank and IESVM-Rank. They differ on the chromosome initialisation mechanism for the evolutionary process. ES-Rank simply sets all genes in the initial chromosome to the same value. IESR-Rank uses linear regression, and IESVM-Rank uses support vector machine for the initialisation process. Experimental results from comparing the proposed method to fourteen other approaches from the literature show that IESR-Rank achieves the overall highest performance. Ten problem instances are used here, obtained from four datasets: MSLR-WEB10K, LETOR 3 and LETOR 4. Performance is measured at the top-10 query-document pairs retrieved, using five metrics: mean average precision (MAP), root-mean-square error (RMSE), precision (P@10), reciprocal rank (RR@10) and normalized discounted cumulative gain (NDCG@10). The contribution of this paper is proposing an effective and efficient LTR method combining a list-wise evolutionary technique with point-wise and pair-wise machine learning techniques. [ABSTRACT FROM AUTHOR]

Heuristic computational intelligence techniques are widely used in combinatorial optimization problems, essentially in large size configurations. Bio-inspired heuristics such as PSO, FA or FPA showed their capacities to solve such problems. Bi-heuristic optimization consists of using a couple of techniques and a collaboration mechanism. This paper reviews the major contributions in solving TSP with bi-heuristics and presents a new hybridization scheme based on FPA, ACO with Ls, ant supervised by flower pollination with local search, ASFPA-Ls; as well as the impact of social and cognitive PSO for the ant supervised by PSO with local search, ASPSO-Ls. AS-chaotic-PSO-Ls which stands for ant supervised by chaotic PSO local search is also investigated. The meta-heuristic algorithms (FPA, PSO or chaotic PSO) and ant colony optimization are used with a hierarchical collaboration schema in addition to a local search mechanism. In this work, the local search strategy, Ls, used is the 2-opt method; the proposals are called, respectively, ASFPA-Ls, cognitive ant supervised by PSO with local search, Co-ASPSO-Ls, social-ASPSO-Ls and AS-chaotic-PSO-Ls; where ACO is coupled with a local search heuristic mechanism called 2-opt, and a set of meta-heuristics, FA, FPA and PSO asked to adapt ACO parameters while running. Comparative experimental investigations are conducted using the TSP test bench. Proposed hybridizations attended fair solutions for the TSP problems used in the experimental investigations including berlin52, St70, eil76, rat 99, eil101, KroA100, Ch150 and kroA200. A good balance performance/time is found with the social ant supervised by PSO with local search, So-ASPSO-Ls. The cognitive ant supervised by PSO with local search, Co-ASPSO-Ls comes in the second position in terms of time effectiveness with close performances to AS-chaoticPS0-LS, all proposed approaches returned low rate errors or BKS for used test benches: berlin52, st70, eil76, rat99, kroA100 and kroA200. [ABSTRACT FROM AUTHOR]

In the era of robotics, path planning and navigation of bipeds in multifaceted environments have always remained as one of the most promising areas of research. In this paper, a novel hybridized navigational controller is proposed using the logic of both statistical technique and computational intelligence method for path planning of bipeds. The proposed path controller is a hybridization of regression analysis with adaptive particle swarm optimization. The inputs given to the regression controller are in the forms of obstacle distances, and the output of the regression controller is interim turning angle. The output interim turning angle is again fed to the adaptive particle swarm optimization controller along with other inputs. The output of the adaptive particle swarm optimization controller termed as final turning angle acts as the directing factor for smooth navigation of bipeds in a cluttered environment. The proposed navigational controller is experimented for a bipedal robot in simulation and experimental environments. The outputs from various simulations and experimental results are compared and good agreement between navigational parameters like path length and path time is observed. Again a better efficacy has been observed by comparing the designed hybridization technique with an existing approach which validates the novel approach. [ABSTRACT FROM AUTHOR]

Nowadays, due to the constant increase in size and complexity of the software systems imposed by their evolution, developing qualitative software systems becomes a highly important task. To achieve this goal, early detection of software defects is a must. The paper proposes an approach to generate rules for software defect prediction. In this respect, a Software Defects Rules Discovery (SDRD) algorithm was put forward. This one uses the ant colony system method to discover the best solution based on code metrics values. We conducted 20 experiments in total (five experiments with three metrics and 15 experiments with combinations of two metrics). The results revealed that the metrics that correlate with the dependent variable are CBO (Coupling Between Objects), RFC (Response For a Class) and NPM (Number of Private Methods), and that from all the combinations of two metrics, for the five projects, the best obtained rule is formed with RFC and NPM metrics. [ABSTRACT FROM AUTHOR]

Vehicle traffic congestion is an increasing concern in metropolitan areas, with negative health, environment and economical implications. In recent times, computational intelligence (CI), a set of nature-inspired computational approaches and algorithms, has been used in vehicle routing and congestion mitigation research (also referred to as CI-based vehicle traffic routing systems—VTRSs). In this paper, we conduct a critique of existing literature on CI-based VTRSs and discuss identified limitations, evaluation process of existing approaches and research trends. We also identify potential research opportunities. [ABSTRACT FROM AUTHOR]

In previous studies, customer survey data were commonly adopted to perform the modelling of customer satisfaction (CS). However, it could be time-consuming to conduct surveys and obtain their data. On the other hand, respondents' responses are quite often confined by preset questions. Nowadays, a huge number of customer online reviews on products can be found on various websites. The reviews can be extracted easily in a very short time. Customers can freely express their concerns and views of products in their online reviews. Those reviews provide a valuable source of information for manufacturers to improve their existing products and develop their new products. Previous studies have attempted to develop CS models based on survey data by using various computational intelligence techniques. However, no attempt at developing explicit CS models based on online reviews was reported in the literature. In this paper, a methodology for the modelling of CS based on customer online reviews and a multigene genetic programming-based fuzzy regression (MGGP-FR) approach is proposed. In the proposed methodology, relevant textual reviews of products are extracted from e-commerce websites. Then, opinion mining is conducted on the reviews and sentiments scores of customer concerns are derived. A MGGP-FR approach is then introduced to develop CS models based on the derived sentiment scores. A case study on developing CS models for electronic hairdryers is conducted to illustrate the proposed methodology and validate the effectiveness of MGGP-FR in the modelling of CS. The validation results show MGGP-FR outperforms the other three modelling approaches, fuzzy regression, genetic programming, and genetic programming-based fuzzy regression, in the CS modelling in terms of prediction accuracy. [ABSTRACT FROM AUTHOR]

Financial forecasting using computational intelligence nowadays remains a hot topic. Recent improvements in deep neural networks allow us to predict financial market behavior better than traditional machine learning approaches. In this paper, we propose three novel deep learning-based financial forecasting frameworks, all of which considerably outperform existing approaches, yielding a much better annual financial return on DOW-30 stocks and Exchange-Traded Funds (ETFs) tested between January 1, 2007, and December 31, 2016. The first framework Convolutional Neural Networks with Technical Indicator Clustering (CNN-TIC) creates images with multiple channels corresponding to the technical indicator clusters and employs the take profit and stop loss techniques to obtain a superior annual financial return. The second model Evolutionary Optimized CNN-TIC (EO-CNN-TIC) computes the optimal values in the take profit and stop loss techniques using one of the recently created evolutionary optimization algorithms, Cuckoo Search. Finally, the third model Residual Network with Technical Analysis (ResNet-TA) applies residual blocks to the convolutional part of the neural network architecture to extract more useful features from deeper layers. Both CNN-TIC and EO-CNN-TIC are based on clustering the technical indicators by their similarity in behavior and creating separate five distinct images based on the five clusters, while ResNet-TA takes advantage of going deeper in the network with residual blocks. All three models further improve their performances by hyperparameter tuning. On DOW-30 stocks, we were able to achieve annual returns of 20.45% , 29.54% , and 36.70% for CNN-TIC, EO-CNN-TIC, and ResNet-TA, whereas for ETFs, 16.56% , 19.20% , and 32.09% annual returns were observed, respectively. We conclude with future work that can be done in order to further improve the computational and financial performances of the models. [ABSTRACT FROM AUTHOR]

Rough set theory introduced by Pawlak (Int J Comput Inf Sci 11:341–356, 1982), multi-set theory proposed by Blizard (Notre Dame J Form Log 30:36–65, 1989) and soft set theory introduced by Molodtsov (Comput Math Appl 37(4–5):19–31, 1999) are fundamental concepts in computational intelligence, which have a myriad of applications in modeling uncertainties and decision making under uncertainty. In this paper, the idea of soft multi-rough set (SMRS) is introduced as a hybrid model of soft set, multi-set and rough set. The SMRS provides roughness of a multi-set in terms of soft multi-approximation space. The novel concept of soft multi-rough topology (SMR-topology) is defined to discuss topological structure of SMRSs by using pairwise SMR-approximations. The proposed models of SMRS and SMR-topology are suitable for modeling uncertainties in the real-life circumstances. SMR-topology is the generalization of crisp topology, soft topology and soft rough set topology. Some fundamental properties of SMR-topology and their related results are studied. Some algorithms for are developed for multi-criteria decision making based on soft multi-sets, soft multi-rough sets and soft multi-rough topology. Based on proposed algorithms, the applications of SMRSs and SMR-topology toward diagnosis of depression and diabetes are illustrated by the numerical examples. A comparison analysis of proposed methods with some existing methods is also given to justify their reliability, feasibility and flexibility. [ABSTRACT FROM AUTHOR]

Area of computational intelligence is gaining researcher's attention in ongoing trend of technology and evolution due to their high capability to deliver near-optimal solutions. A new hierarchy of algorithms has been proposed in the paper, and they have been organized on the basis of their inspiration sources. The broad two domains of the algorithms are modeling of human mind and nature-inspired intelligence. Nature-inspired computational algorithms being heuristic algorithms are robust and have optimization capability to solve obscure and substantiated problems. The heuristic techniques aim on finding the best possible solution to the query in a satisfiable amount of time. The computational intelligence methods inspired from nature have further been categorized into artificial immune systems, evolutionary algorithms, swarm intelligence, artificial neural networks and geoscience-based algorithms. Geoscience-based domain is the least explored domain in which the algorithms can be developed based on geographic phenomenon taking place on the earth's surface. An extensive tabular comparison is done among algorithms of all the domains on the basis of various attributes. Also, variants of the algorithms and their implementation in a specific application have been examined. The efficiency and performance of selected algorithms have been compared on clustering and traveling salesman problem for better understanding. [ABSTRACT FROM AUTHOR]

Medical decision support systems have been a core of intense research for years. The ongoing study shows that artificial intelligence has been accustomed to probe risk factors for hypertension. Factors, like health-damaging personal behaviors and changes in lifestyle and environment, are major contributors to chronic diseases. The goal of this research was to forecast the risk of developing hypertension by revealing hidden patterns in medical datasets. Quality of the data is the key to enhance the performance of learning model. But most healthcare data suffer from class imbalance problem, which induce the need for an intelligent model which can learn from such grimy data. This paper incorporates a novel approach by combing learning model and rule-based mining to offer decision support. Typically, the proposed work comprises two main implications. First suggests an intelligent learning model using boosting-based support vector machine to diagnose and expose multi-class categories in the imbalanced datasets. Finally, the enhanced predictive model is built upon the classification solution which will portray the innate data similarities. An intelligent fuzzy-based approach was employed to recognize frequent behavioral patterns. Based on these rules, valid decisions could be made to prevent hypertension. The suggested enhanced model is evaluated using a real-time hypertension dataset obtained through primary health centers. With the combination of ensemble strategies, the proposed intelligent learning model attains high classification accuracy for the imbalanced dataset above the traditional model. Thus, the efficient integration of personalized behavior with health data could provide a better understanding regarding patient health. In future this can serve as an eye toward personalized medicine. [ABSTRACT FROM AUTHOR]

Workers healthcare gained a lot of attention recently as many countries are increasingly concerning about welfare. This paper faces the problem of predicting occupational disease risks by means of computational intelligence and pattern recognition techniques. Specifically, three different machine learning approaches are compared: the first one is based on the k-means algorithm, in charge to determine a set of meaningful labelled clusters as the final model. The latter two are based on fully supervised techniques, namely Support Vector Machines and K-Nearest Neighbours. Real data regarding both the worker and the workplace by mixing numerical and categorical attributes have been used for testing. The three approaches are automatically tuned by means of genetic algorithms in order to simultaneously find the optimal hyperparameters for the classification systems and the optimal ad-hoc dissimilarity measure weights in order to maximize the classification performances. Computational results show that the three approaches are rather comparable in terms of performances, but a clustering-based approach allows a deeper knowledge discovery phase, helpful for further risk assessment and forecasting. [ABSTRACT FROM AUTHOR]

Fuzzy clustering especially fuzzy $$C$$ -means (FCM) is considered as a useful tool in the processes of pattern recognition and knowledge discovery from a database; thus being applied to various crucial, socioeconomic applications. Nevertheless, the clustering quality of FCM is not high since this algorithm is deployed on the basis of the traditional fuzzy sets, which have some limitations in the membership representation, the determination of hesitancy and the vagueness of prototype parameters. Various improvement versions of FCM on some extensions of the traditional fuzzy sets have been proposed to tackle with those limitations. In this paper, we consider another improvement of FCM on the picture fuzzy sets, which is a generalization of the traditional fuzzy sets and the intuitionistic fuzzy sets, and present a novel picture fuzzy clustering algorithm, the so-called FC-PFS. A numerical example on the IRIS dataset is conducted to illustrate the activities of the proposed algorithm. The experimental results on various benchmark datasets of UCI Machine Learning Repository under different scenarios of parameters of the algorithm reveal that FC-PFS has better clustering quality than some relevant clustering algorithms such as FCM, IFCM, KFCM and KIFCM. [ABSTRACT FROM AUTHOR]

Intrusion detection systems are devoted to monitor a network with aims at finding and avoiding anomalous events. In particular, we focus on misuse detection systems, which are trained to identify several known types of attacks. These can be unauthorized accesses, or denial of service attacks, among others. Whenever it scans a trace of a suspicious event, it is programmed to trigger an alert and/or to block this dangerous access to the system. Depending on the security policies of the network, the administrator may seek different requirements that will have a strong dependency on the behavior of the intrusion detection system. For a given application, the cost of raising false alarms could be higher than carrying out a preventive access lock. In other scenarios, there could be a necessity of correctly identifying the exact type of cyber attack to proceed in a given way. In this paper, we propose a multi-objective evolutionary fuzzy system for the development of a system that can be trained using different metrics. By increasing the search space during the optimization of the model, more accurate solutions are expected to be obtained. Additionally, this scheme allows the final user to decide, among a broad set of solutions, which one is better suited for the current network characteristics. Our experimental results, using the well-known KDDCup'99 problem, supports the quality of this novel approach in contrast to the state-of-the-art for evolutionary fuzzy systems in intrusion detection, as well as the C4.5 decision tree. [ABSTRACT FROM AUTHOR]

The increased demand of systems able to generate reports in natural language from numerical data involves the search for new solutions. This paper presents an adaptation of standard natural language generation methodologies to generate customized linguistic descriptions of data. Namely, we merge one of the most well-known architectures in the natural language generation research field together with our previous architecture for generating linguistic descriptions of complex phenomena. The latter is supported by the computational theory of perceptions which comes from the fuzzy sets and systems research field. We include a practical case of use dealing with the problem of inefficient consumption of energy at households. It generates natural language recommendations adapted to each household to promote a more responsible consumption. The proposal reveals opportunities of collaboration between the different research communities that are involved. [ABSTRACT FROM AUTHOR]

In differential evolution (DE) research, many successful empirical guidelines in selecting appropriate trial vector generation strategies and control parameter values for various problems have been investigated. The comprehensive exploration of the experience can be an effective way to develop an advanced DE variant. In this paper, an improved DE approach with time-frame strategy adaptation called the time-frame adaptive differential evolution (TFADE) is proposed. It employs diverse trial vector generation strategies with various control parameter values that can be adaptively determined to generate promising solutions and dynamically adjusted to deal with premature convergence during evolution, according to successful experience over a period of preceding generations called the time frame. In the experimental study, TFADE is compared with 4 commonly used conventional DEs, 3 outstanding state-of-the-art adaptive DEs, and 2 novel non-DE approaches, evaluated by a test suite of 25 benchmark functions. The experimental results show that the performance of TFADE is significantly superior to these competitors. [ABSTRACT FROM AUTHOR]

This paper studies a hybrid artificial bee colony (ABC) algorithm for finding high quality solutions of the job-shop scheduling problem with no-wait constraint (JSPNW) with the objective of minimizing makespan among all the jobs. JSPNW is an extension of well-known job-shop scheduling problem subject to the constraint that no waiting time is allowed between operations for a given job. ABC algorithm is a swarm intelligence technique based on intelligent foraging behavior of honey bee swarm. The proposed hybrid approach effectively coordinates the various components of ABC algorithm such as solution initialization, selection and determination of a neighboring solution with the local search in such a way that it leads to high quality solutions for the JSPNW. The proposed approach is compared with the two best approaches in the literature on a set of benchmark instances. Computational results show the superiority of the proposed approach over these two best approaches. [ABSTRACT FROM AUTHOR]

This paper proposes a new architecture and learning algorithms for a hybrid cascade neural network with pool optimization in each cascade. The proposed system is different from existing cascade systems in its capability to operate in an online mode, which allows it to work with non-stationary and stochastic nonlinear chaotic signals with the required accuracy. Compared to conventional analogs, the proposed system provides computational simplicity and possesses both tracking and filtering capabilities. [ABSTRACT FROM AUTHOR]

More accurate prediction of software maintenance effort contributes to better management and control of software maintenance. Several research studies have recently investigated the use of computational intelligence models for software maintainability prediction. The performance of these models, however, may vary from dataset to dataset. Consequently, ensemble methods have become increasingly popular as they take advantage of the capabilities of their constituent computational intelligence models toward a dataset to come up with more accurate or at least competitive prediction accuracy compared to individual models. This paper investigates and empirically evaluates different homogenous and heterogeneous ensemble methods in predicting software maintenance effort and change proneness. Three major empirical studies were designed and conducted taken into consideration different design such as the types of the investigated ensembles methods, types of prediction problems, used datasets, and other experimental setup. Overall empirical evidence obtained from the three studies confirms that some ensemble methods provide more accurate or at least competitive prediction accuracy compared to individual models across datasets, and thus they are more reliable. [ABSTRACT FROM AUTHOR]

In this paper, we revisit the mean-variance model of Markowitz and the construction of the risk-return efficient frontier. A few other models, such as the mean absolute deviation, the minimax and maximin, and models with diagonal quadratic form as objectives, which use alternative metrics for risk are also introduced. Then we present a neurodynamic model for solving these kinds of problems. By employing Lyapunov function approach, it is also shown that the proposed neural network model is stable in the sense of Lyapunov and it is globally convergent to an exact optimal solution of the original problem. The validity and transient behavior of the neural network are demonstrated by using several examples of portfolio selection. [ABSTRACT FROM AUTHOR]

Wireless sensor networks (WSNs) is an emerging technology in several application domains, ranging from urban surveillance to environmental and structural monitoring. Computational intelligence (CI) techniques are particularly suitable for enhancing these systems. However, when embedding CI into wireless sensors, severe hardware limitations must be taken into account. In this paper we investigate the possibility to perform an online, distributed optimization process within a WSN. Such a system might be used, for example, to implement advanced network features like distributed modelling, self-optimizing protocols, and anomaly detection, to name a few. The proposed approach, called DOWSN (distributed optimization for WSN) is an island-model infrastructure in which each node executes a simple, computationally cheap (both in terms of CPU and memory) optimization algorithm, and shares promising solutions with its neighbors. We perform extensive tests of different DOWSN configurations on a benchmark made up of 15 continuous optimization problems; we analyze the influence of the network parameters (number of nodes, inter-node communication period and probability of accepting incoming solutions) on the optimization performance. Finally, we profile energy and memory consumption of DOWSN to show the efficient usage of the limited hardware resources available on the sensor nodes. [ABSTRACT FROM AUTHOR]

In this paper we consider the problem of finding a global optimum of a multimodal function applying path relinking. In particular, we target unconstrained large-scale problems and compare two variants of this methodology: the static and the evolutionary path relinking (EvoPR). Both are based on the strategy of creating trajectories of moves passing through high-quality solutions in order to incorporate their attributes to the explored solutions. Computational comparisons are performed on a test-bed of 19 global optimization functions previously reported with dimensions ranging from 50 to 1,000, totalizing 95 instances. Our results show that the EvoPR procedure is competitive with the state-of-the-art methods in terms of the average optimality gap achieved. Statistical analysis is applied to draw significant conclusions. [ABSTRACT FROM AUTHOR]

Fuzzy sets, soft sets and N-soft set are essential concepts for fuzzy modeling, decision making under uncertainty and computational intelligence. Nevertheless, all these models have some limitations imposed on the alternatives and attributes. These mathematical models were unable to handle the situations whenever decision makers need to assign non-binary evaluations to both attributes and alternatives. To overcome these problems, Riaz et al. (Symmetry 13(5):1–31, 2021) proposed the idea of M-parameterized N-soft set (MPNSS), through examining independent non-binary evaluations to both parameters and alternatives. MPNSSs are highly convenient in representation of ambiguous and uncertain data in decision analysis during ranking, rating and grading the objects. The proposed model has greater efficiency to handle the imprecision rather than existing mathematical models. Additionally, we develop multi-attribute decision-making (MADM) techniques on account of M-parameterized N-soft sets (MPNSSs), M-parameterized N-soft aggregation operators (MPNSAO) and M-parameterized N-soft weighted aggregation operators (MPNSWAO). For these objectives, we define fundamental operations on MPNSSs and their properties. Meanwhile, several related results are well proven and some algorithms for MADM are also developed. Moreover, the applications of MADM techniques corresponding to proposed algorithms are explained by illustrative material. To explain the presumption, authenticity and effectiveness of the suggested course of actions, we compared the techniques with few existing MADM methods. [ABSTRACT FROM AUTHOR]

Pile foundations are often subjected to lateral pressures in addition to axial loads due to earthquakes, ground pressure, and wind pressures in various buildings. As a result, pile foundations have gotten more research than any other type of foundation. Furthermore, estimating the lateral load capacity of a pile (LLCP) accurately is a difficult undertaking, and there has been relatively little study in this field. To overcome these problems, in this study, the adaptive neuro-fuzzy inference system (ANFIS) was ustilized to construct forecasting models for the indirect assessment of LLCP embedded in clay in this work. The fuzzy c-means clustering technique (FCM) and the subtractive clustering method (SCM) were implemented as ANFIS models. The data from open-source literature were used to evaluate the two ANFIS models. In these models, pile length (L), pile diameter (D), undrained shear strength of soil (Su), and eccentricity of load (e) were used as the inputs, while the measured LLCP in clay was the output. To compare the performance of the estimating models, several statistical performance measures were used. The modeling results show that the relationships determined for estimating the LLCP in clay by ANFIS models (ANFIS-SCM and ANFIS-FCM) are accurate and close to the real value. It can also be concluded that the use of ANFIS models to predict the LLCP in clay is very efficient. [ABSTRACT FROM AUTHOR]

Machine learning-based computational intelligence methods are widely used to analyze large-scale data sets in this age of big data. Extracting useful predictive modeling from these types of data sets is a challenging problem due to their high complexity. Analyzing large amount of streaming data that can be leveraged to derive business value is another complex problem to solve. With high levels of data availability (i.e., Big Data), automatic classification of them has become an important and complex task. Hence, we explore the power of applying MapReduce-based distributed AdaBoosting of extreme learning machine (ELM) to build a predictive bag of classification models. Accordingly, (1) data set ensembles are created; (2) ELM algorithm is used to build weak learners (classifier functions); and (3) builds a strong learner from a set of weak learners. We applied this training model to the benchmark knowledge discovery and data mining data sets. [ABSTRACT FROM AUTHOR]

The strength behavior of a rock mass is often controlled by different discontinuities therein, along which the structural failures may threaten rock structures. Thus, the shear strength criteria for rock joints have always been a fundamental topic in rock mechanics. Knowing the different aspects of the rock joint shear strength based on the previously presented criteria, it is time to try different approaches other than the regression analysis in order to model the nonlinear behavior of rock joints. This research focuses on the estimation of the shear strength of rock joints using the computational intelligence. A total of 84 direct shear tests were first performed on replicas of natural rock fractures with various mechanical and morphological characteristics under several normal stress levels. Then, an adaptive neuro-fuzzy inference system (ANFIS) combined with a teaching–learning-based optimization (TLBO) algorithm was used to establish a new shear strength criterion. The results demonstrated that the ANFIS–TLBO criterion provided an accurate estimation of rock joint shear strength. A comparison between the ANFIS–TLBO criterion and the Barton's empirical equation revealed a better performance of the suggested model in mapping the experimental data. The ANFIS–TLBO model's residual indicated a random pattern, and its histogram exhibited a symmetric bell-shaped distribution around zero, supporting the appropriateness of the model. [ABSTRACT FROM AUTHOR]

Business games have been widely used as differentiated pedagogical tools to provide experiential learning for business students. However, a critical problem with these tools is the issue of how to give feedback to students during the runtime of the simulation, especially in view of the high number of players involved in the game and the large amount of data generated in the simulations. In this scenario, intelligent mechanisms are desirable to make knowledge-based inferences, providing information which can assist both the players and the instructors facilitating the gaming process. In this work, we present an innovative knowledge-based approach focused on business games. Firstly, we apply data mining techniques to identify the behavioral patterns of players, based on their previous decisions stored in the database of a business game called business management simulator (BMS) that is used as a support tool for teaching concepts of production management, sales and business strategies. Secondly, based on these patterns, we develop a fuzzy inference system (FIS) to predict players' performance based on their decisions in the game. Experimental results from a comparison of the real performance of players with the performance calculated by the proposed FIS show that this approach is very useful in the business game analyzed here, since it can help students during the simulation runtime, allowing them to improve their decisions. It is also clear that the proposed approach can be easily adapted to other business games, and particularly those with a similar structure to that of BMS. [ABSTRACT FROM AUTHOR]

The present study proposes a new self-tuning least squares support vector machine, called MSOS-SVM, for modeling the pavement rutting behavior of asphalt mixtures. MSOS-SVM combines the least squares support vector machine (LS-SVM), the symbiotic organisms search (SOS), and chaotic maps. In this system, the LS-SVM is used to establish the relationship model between the flow number obtained from laboratory tests and the parameters specified in the asphalt mix design. SOS is used to find the best LS-SVM tuning parameters. Meanwhile, chaotic system is used to enhance the exploration and exploitation process of SOS. A total of 118 historical cases were used to establish the intelligence-prediction model. The results validate the ability of MSOS-SVM to model the pavement rutting behavior of asphalt mixtures to a relatively high level of accuracy as measured using four error indicators. The present study demonstrates that the proposed computational intelligence system is a highly beneficial decision-making tool for road designers and engineers. [ABSTRACT FROM AUTHOR]

Efficient clustering is a well-documented NP-hard optimization problem in wireless sensor networks (WSNs). Variety of computational intelligence techniques including evolutionary algorithms, reinforcement learning, artificial immune systems and recently, artificial bee colony (ABC) metaheuristic have been applied for efficient clustering in WSNs. Due to ease of use and adaptive nature, ABC arose much interest over other population-based metaheuristics for solving optimization problems in WSNs. However, its search equation contributes to its insufficiency due to comparably poor exploitation cycle and requirement of certain control parameters. Thus, we propose an improved artificial bee colony (iABC) metaheuristic with an improved solution search equation to improve exploitation capabilities of existing metaheuristic. Further, to enhance the global convergence of the proposed metaheuristic, an improved population sampling technique is introduced through Student's t-distribution, which require only one control parameter to compute and store and therefore increase efficiency of proposed metaheuristic. The proposed metaheuristic maintain a good balance between exploration and exploitation search abilities with least memory requirements; moreover, the use of first-of-its-kind compact Student's t-distribution makes it suitable for limited hardware requirements of WSNs. Additionally, an energy-efficient clustering protocol based on iABC metaheuristic is presented, which inherits the capabilities of the proposed metaheuristic to obtain optimal cluster heads along with an optimal base station location to improve energy efficiency in WSNs. Simulation results show that the proposed clustering protocol outperforms other well-known protocols on the basis of packet delivery, throughput, energy consumption, network lifetime and latency as performance metric. [ABSTRACT FROM AUTHOR]

In this study, we have introduced a hybrid version of artificial bee colony (ABC) and shuffled frog-leaping algorithm (SFLA). The hybrid version is a two-phase modification process. In the first phase to increase the global convergence, the initial population is produced using randomly generated and chaotic system, and then in the second phase to balance two antagonist factors, i.e., exploration and exploitation capabilities, population is portioned into two groups (superior and inferior) based on their fitness values. ABC is applied to the first group, whereas SFLA is applied to the second group of population. The proposed version is named as Shuffled-ABC. The proposal is implemented and tested on constrained benchmark consulted from CEC 2006 and five chemical engineering problems where constraints are handled using penalty function methods. The simulated results illustrate the efficacy of the proposal. [ABSTRACT FROM AUTHOR]

Several properties of transition metal nitrides (TMN) that make them useful in many applications are closely related to the state of their surfaces. Meanwhile, high melting points which characterize these materials make the determination of their surface energies experimentally difficult. This work presents a computational intelligence technique using support vector regression (SVR) to establish, for the first time, a complete database of average surface energies of all members of TMN series. SVR-based model was developed by training and testing SVR with best parameters obtained through test-set-cross-validation technique using thirty-five experimental data of periodic metals. The developed SVR-based model was used to estimate average surface energies of 3d, 4d and 5d-TMN, and the obtained results agree well with the existing theoretical values. Simple and effective computational approach of the developed model together with its accurate estimation of average surface energies of all the members of TMN series contributes to the uniqueness of this developed model over the existing theoretical methods. [ABSTRACT FROM AUTHOR]

top end graphics card (GPU) plus a suitable SIMD interpreter can deliver a several hundred fold speed up, yet cost less than the computer holding it. We give highlights of AI and computational intelligence applications in the new field of general purpose computing on graphics hardware (GPGPU). In particular, we surveyed genetic programming (GP) use with GPU. We gave several applications from Bioinformatics and showed that how the fastest GP is based on an interpreter rather than compilation. Finally using GP to generate GPU CUDA kernel C++ code is sketched. [ABSTRACT FROM AUTHOR]