Showing papers in "International Journal of Computational Intelligence Systems in 2018"

Entropy Measures of Probabilistic Linguistic Term Sets

Abstract: The probabilistic linguistic term sets (PLTSs) are powerful to deal with the hesitant linguistic situation in which each provided linguistic term has a probability. The PLTSs contain uncertainties caused by the linguistic terms and their probability information. In order to measure such uncertainties, three entropy measures are proposed: the fuzzy entropy, the hesitant entropy, and the total entropy. The fuzzy entropy measures the fuzziness of the PLTSs, and the hesitant entropy measures the hesitation of the PLTSs. To facilitate the computation of all uncertainties contained in the PLTSs, the total entropy is proposed. Some properties and some formulas of the entropy measures are introduced. A multi-criteria decision making model based on the PLTSs is introduced by using the proposed entropy measures. An illustrative example is provided and the comparison analysis with the existing method is given.

A Comparison of Outlier Detection Techniques for High-Dimensional Data

Abstract: Outlier detection is a hot topic in machine learning. With the newly emerging technologies and diverse applications, the interest of outlier detection is increasing greatly. Recently, a significant number of outlier detection methods have been witnessed and successfully applied in a wide range of fields, including medical health, credit card fraud and intrusion detection. They can be used for conventional data analysis. However, it is not a trivial work to identify rare behaviors or patterns out from complicated data. In this paper, we provide a brief overview of the outlier detection methods for high-dimensional data, and offer comprehensive understanding of the-state-of-the-art techniques of outlier detection for practitioners. Specifically, we firstly summarize the recent advances on outlier detection for high-dimensional data, and then make an extensive experimental comparison to the popular detection methods on public datasets. Finally, several challenging issues and future research directions are discussed.

Missing values estimation and consensus building for incomplete hesitant fuzzy preference relations with multiplicative consistency

Abstract: This paper proposes a decision support process for incomplete hesitant fuzzy preference relations (HFPRs). First, we present a revised definition of HFPRs, in which the values are not ordered for the hesitant fuzzy element. Second, we propose a method to normalize the HFPRs and estimate the missing elements in incomplete HFPRs based on multiplicative consistency. Based on this, a consensus model with incomplete HFPR is developed. A feedback mechanism is proposed to obtain a best choice with desired consensus level. Multiplicative consistency induced ordered weighted averaging (MC-IOWA) operator is used to aggregate the individual HFPRs into a collective one. A score HFPR is proposed for collective HFPR, and then the hesitant quantifier-guided non-dominance degrees (HQGNDD) of alternatives by using an OWA operator are obtained to rank the alternatives. Finally, a case study for evaluate the qualification of supply chain enterprises is provided to illustrate its application.

Interval-valued Pythagorean Fuzzy Frank Power Aggregation Operators based on An Isomorphic Frank Dual Triple

Abstract: Interval-valued Pythagorean fuzzy sets (PFSs), as an extension of PFSs, have strong potential in the management of complex uncertainty in real-world applications. This study aims to develop several intervalvalued Pythagorean fuzzy Frank power (IVPFFP) aggregation operators with an adjustable parameter via the integration of an isomorphic Frank dual triple. First, a special automorphism on unit interval is introduced to construct an isomorphic Frank dual triple; and this triple is further applied on the definition of interval-valued Pythagorean fuzzy Frank operational laws. Second, two IVPFFP aggregation operators with the inclusion of an adjustable parameter are defined on the basis of the proposed operational laws, and several instrumental properties are then investigated. Furthermore, some limiting cases of the proposed IVPFFP operators are analyzed with respect to the varying adjustable parameter values. Finally, an IVPFFP aggregation operator-based multiple attribute group decision-making model is developed with a practical example furnished to demonstrate its feasibility and efficiency. The power that the adjustable parameter exhibits has been leveraged to affect the final decision results, and the proposed IVPFFP operators are compared with three selected aggregation operators to demonstrate their advantages provided with a practical example.

A computer aided analysis scheme for detecting epileptic seizure from EEG data

Abstract: This paper presents a computer aided analysis system for detecting epileptic seizure from electroencephalogram (EEG) signal data. As EEG recordings contain a vast amount of data, which is heterogeneous with respect to a time-period, we intend to introduce a clustering technique to discover different groups of data according to similarities or dissimilarities among the patterns. In the proposed methodology, we use K-means clustering for partitioning each category EEG data set (e.g. healthy; epileptic seizure) into several clusters and then extract some representative characteristics from each cluster. Subsequently, we integrate all the features from all the clusters in one feature set and then evaluate that feature set by three well-known machine learning methods: Support Vector Machine (SVM), Naive bayes and Logistic regression. The proposed method is tested by a publicly available benchmark database: 'Epileptic EEG database'. The experimental results show that the proposed scheme with SVM classifier yields overall accuracy of 100% for classifying healthy vs epileptic seizure signals and outperforms all the recent reported existing methods in the literature. The major finding of this research is that the proposed K-means clustering based approach has an ability to efficiently handle EEG data for the detection of epileptic seizure.

A dynamic multi-attribute group emergency decision making method considering experts’ hesitation

Abstract: This work was partly supported by the Young Doctoral Dissertation Project of Social Science Planning Project of Fujian Province (Project No. FJ2016C202), National Natural Science Foundation of China (Project No. 71371053, 61773123), Spanish National Research Project ( Project No. TIN2015-66524-P), and Spanish Ministry of Economy and Finance Postdoctoral Fellow (IJCI-2015- 23715).

Assessment of Healthcare Waste Treatment Alternatives Using an Integrated Decision Support Framework

Abstract: Healthcare waste (HCW) management has become a major environmental and public-health concern especially in developing countries, and therefore, it has been receiving increasing attention from both industrial practitioners and researcher in recent years. Selection of the optimal treatment technology for HCW is regarded as an intricate multi-criteria decision-making problem involving conflicting and intertwined qualitative as well as quantitative evaluative criteria. To address this decision problem, we develop an integrated decision support framework based on decision-making trial and evaluation laboratory (DEMATEL), intuitionistic fuzzy ANP, and intuitionistic fuzzy AHP. The DEMATEL method determines the influences of main factors and criteria and produces a network relationship map while the ANP method calculates the degree of interrelationship among evaluative criteria and obtains their relative weights. The AHP method assesses the HCW treatment alternatives over evaluative criteria. The experts’ opinions are collected in form of intuitionistic fuzzy preference relations as they are effective in capturing uncertainty and hesitancy involved in decision-makers’ judgment.We also develop a priority method to derive nonfuzzy weights from the intuitionistic fuzzy preference relations. To validate the feasibility of the proposed approach, a case study is carried out on the selection of optimum HCW treatment technology for Chhattisgarh, India. The analyses of the results indicate that the proposed integrated multi-criteria decision making (MCDM) framework effectively handles the issue of HCW treatment technology selection in uncertain environments. © 2019 The Authors. Published by Atlantis Press SARL. This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

The linguistic intuitionistic fuzzy set TOPSIS method for linguistic multi-criteria decision makings

Abstract: In the paper, we express uncertain assessments information in linguistic multi-criteria decision makings (LMCDMs) as linguistic intuitionistic fuzzy sets, i.e., the decision maker provides membership and nonmembership fuzzy linguistic terms to represent uncertain assessments information of alternatives in LMCDMs, and present Hamming distance between two linguistic intuitionistic fuzzy sets. Then we propose the linguistic intuitionistic fuzzy set TOPSIS method for LMCDMs, compared with the traditional TOPSIS method, we provide different the positive ideal solution, the negative ideal solution and the relative closeness degrees of alternatives, in addition, we design an algorithm to finish the linguistic intuitionistic fuzzy set TOPSIS method for LMCDMs. We utilize a LMCDM problem to illustrate the performance, usefulness and effectiveness of the linguistic intuitionistic fuzzy set TOPSIS method, and compare it with the hesitant fuzzy linguistic multi-criteria optimization and compromise solution (HFL-VIKOR) method, the symbolic aggregation-based method and the hesitant fuzzy linguistic TOPSIS (HFL-TOPSIS) method in the example, results show that the linguistic intuitionistic fuzzy set TOPSIS method is a useful and alternative method for LMCDMs.

Genetic Algorithm Approaches for Improving Prediction Accuracy of Multi-criteria Recommender Systems

Abstract: We often make decisions on the things we like, dislike, or even don’t care about. However, taking the right decisions becomes relatively difficult from a variety of items from different sources. Recommender systems are intelligent decision support software tools that help users to discover items that might be of interest to them. Various techniques and approaches have been applied to design and implement such systems to generate credible recommendations to users. A multi-criteria recommendation technique is an extended approach for modeling user’s preferences based on several characteristics of the items. This research presents genetic algorithm-based approaches for predicting user preferences in multi-criteria recommendation problems. Three genetic algorithms’ methods, namely standard genetic algorithm, adaptive genetic algorithm, and multi-heuristic genetic algorithms are used to conduct the experiments using a multi-criteria dataset for movies recommendation. The empirical results of the comparative analysis of their performance are presented in this study.

Credit risk prediction in an imbalanced social lending environment

Abstract: Credit risk prediction is an effective way of evaluating whether a potential borrower will repay a loan, particularly in peer-to-peer lending where class imbalance problems are prevalent. However, few credit risk prediction models for social lending consider imbalanced data and, further, the best resampling technique to use with imbalanced data is still controversial. In an attempt to address these problems, this paper presents an empirical comparison of various combinations of classifiers and resampling techniques within a novel risk assessment methodology that incorporates imbalanced data. The credit predictions from each combination are evaluated with a G-mean measure to avoid bias towards the majority class, which has not been considered in similar studies. The results reveal that combining random forest and random under-sampling may be an effective strategy for calculating the credit risk associated with loan applicants in social lending markets.

Tire Defects Classification Using Convolution Architecture for Fast Feature Embedding

Abstract: Convolutional Neural Network (CNN) has become an increasingly important research field in machine learning and computer vision. Deep image features can be learned and subsequently used for detection, classification and retrieval tasks in an end-to-end model. In this paper, a supervised feature embedded deep learning based tire defects classification method is proposed. We probe into deep learning based image classification problems with application to real-world industrial tasks. Combined regularization techniques are applied for training to boost the performance. Experimental results show that our scheme receives satisfactory classification accuracy and outperforms state-of-the-art methods.

A Genetic Algorithm with New Local Operators for Multiple Traveling Salesman Problems

Abstract: Multiple Traveling Salesman Problem (MTSP) is able to model and solve various real-life applications such as multiple scheduling, multiple vehicle routing and multiple path planning problems, etc. While Traveling Salesman Problem (TSP) focuses on searching a path of minimum traveling distance to visit all cities exactly once by one salesman, the objective of the MTSP is to find m paths for m salesmen with a minimized total cost the sum of traveling distances of all salesmen through all of the respective cities covered. They have to start from a designated depot which is the departing and returning location of all salesmen. Since the MTSP is a NP-hard problem, a new effective Genetic Algorithm with Local operators (GAL) is proposed in this paper to solve the MTSP and generate high quality solution within a reasonable amount of time for real-life applications. Two new local operators, Branch and Bound (BaB) and Cross Elimination (CE), are designed to speed up the convergence of the search process and improve the solution quality. Results demonstrate that GAL finds a better set of paths with a 9.62% saving on average in cost comparing to two existing MTSP algorithms.

SCAN: Semantic Context Aware Network for Accurate Small Object Detection

Abstract: Recent deep convolutional neural network-based object detectors have shown promising performance when detecting large objects, but they are still limited in detecting small or partially occluded ones—in part because such objects convey limited information due to the small areas they occupy in images. Consequently, it is difficult for deep neural networks to extract sufficient distinguishing fine-grained features for high-level feature maps, which are crucial for the network to precisely locate small or partially occluded objects. There are two ways to alleviate this problem: the first is to use lower-level but larger feature maps to improve location accuracy and the second is to use context information to increase classification accuracy. In this paper, we combine both methods by first constructing larger and more meaningful feature maps in top-down order and concatenating them and subsequently fusing multilevel contextual information through pyramid pooling to construct context aware features. We propose a unified framework called the Semantic Context Aware Network (SCAN) to enhance object detection accuracy. SCAN is simple to implement and can be trained from end to end. We evaluate the proposed network on the KITTI challenge benchmark and present an improvement of the precision.

CPP-ELM: Cryptographically Privacy-Preserving Extreme Learning Machine for Cloud Systems

Abstract: The training techniques of the distributed machine learning approach replace the traditional methods with a cloud computing infrastructure and provide flexible computing services to clients. Moreover, machine learning-based classification methods are used in many diverse applications such as medical predictions, speech/face recognition, and financial applications. Most of the application areas require security and confidentiality for both the data and the classifier model. In order to prevent the risk of confidential data disclosure while outsourcing the data analysis, we propose a privacy-preserving protocol approach for the extreme learning machine algorithm and give private classification protocols. The proposed protocols compute the hidden layer output matrix H in an encrypted form by using a distributed multi-party computation (or cloud computing model) approach. This paper shows how to build a privacy-preserving classification model from encrypted data.

Optimization of quality measures in association rule mining: an empirical study

Abstract: In the association rule mining field many different quality measures have been proposed over time with the aim of quantifying the interestingness of each discovered rule. In evolutionary computation, many of these metrics have been used as functions to be optimized, but the selection of a set of suitable quality measures for each specific problem is not a trivial task. The aim of this paper is to review the most widely used quality measures, analyze their properties from an empirical standpoint and, as a result, ease the process of selecting a subset of them for tackling the task of mining association rules through evolutionary computation. The experimental analysis includes twenty metrics, thirty datasets and a diverse set of algorithms to describe which quality measures are related (or unrelated) so they should (or should not) be used at time. A series of recomendations are therefore provided according to which quality measures are easily optimized, what set of measures should be used to optimize the whole set of metrics, or which measures are hardly optimized by any other.

Pessimistic Bilevel Optimization: A Survey

Abstract: Bilevel optimization are often addressed in an organizational hierarchy in which the upper level decision maker is the leader and the lower level decision maker is the follower. The leader frequently cannot obtain complete information from the follower. As a result, the leader most tends to be risk-averse, and then would like to create a safety margin to bound the damage resulting from the undesirable selection of the follower. Pessimistic bilevel optimization represents an attractive tool to model risk-averse hierarchy problems, and would provide strong ability of analysis for the risk-averse leader. Since to the best of our knowledge, there is not a comprehensive review on pessimistic bilevel optimization, the goal of this paper is to provide a extensive review on pessimistic bilevel optimization from basic definitions and properties to solution approaches. Some real applications are also proposed. This survey will directly support researchers in understanding theoretical research results, designing solution algorithms and applications in relation to pessimistic bilevel optimization.

A Pythagorean Fuzzy TOPSIS Method Based on Novel Correlation Measures and Its Application to Multiple Criteria Decision Analysis of Inpatient Stroke Rehabilitation

Abstract: Department of Nursing, Linkou Chang Gung Memorial Hospital, Department of Nursing, Chang Gung University of Science and Technology, No.5, Fuxing St., Guishan District, Taoyuan City 333, Taiwan Department of Nursing, Linkou Chang Gung Memorial Hospital, Department of Nursing, Chang Gung University of Science and Technology, No.5, Fuxing St., Guishan District, Taoyuan City 333, Taiwan Department of Nursing, Linkou Chang Gung Memorial Hospital, Department of Nursing, Chang Gung University of Science and Technology, No.5, Fuxing St., Guishan District, Taoyuan City 333, Taiwan Graduate Institute of Business and Management, Chang Gung University, Department of Industrial and Business Management, Chang Gung University, Department of Nursing, Linkou Chang Gung Memorial Hospital, No. 259, Wenhua 1st Rd., Guishan District, Taoyuan City 33302, Taiwan

A New Approach for Condition Monitoring and Detection of Rail Components and Rail Track in Railway

Abstract: Computer vision-based tracking and fault detection methods are increasingly growing method for use on railway systems. These methods make detection of components of the railways and fault detection and condition monitoring process can be performed using data obtained by means of computers. In this study, methods are proposed for fault detection on railway components and condition monitoring. With cameras placed on the bottom and the top of the experimental vehicle the images are taken. The camera at the top, overhead rails are positioned to see a way for war and the camera is fixed to the bottom mounted to see clearly railway components. Images from cameras placed on the bottom, Canny edge extraction and Hough transform methods are applied. The types of the components and faults are determined by using classification algorithm with decision trees using the obtained data. The condition monitoring has done by the camera is positioned on the upper part of the vehicle. By processing the taken images with processing methods, inclination angle of the rails and direction of railways are detected. Thus, during the course of the vehicle is obtained information of the direction of railway. Real images are used in the operation of railways belonging to the experimental environment. On these images, to identify the components of the proposed method using the railways and rail direction determination is made. The results obtained are given at the end of the study. The experimental results are analyzed, it is observed that the proposed method accurate and effective results.

An improved chaotic firefly algorithm for global numerical optimization

Abstract: Firefly algorithm (FA) is a prominent metaheuristc technique. It has been widely studied and hence there are a lot of modified FA variants proposed to solve hard optimization problems from various areas. In this paper an improved chaotic firefly algorithm (ICFA) is proposed for solving global optimization problems. The ICFA uses firefly algorithm with chaos (CFA) as the parent algorithm since it replaces the attractiveness coefficient by the outputs of the chaotic map. The enhancement of the proposed approach involves introducing a novel search strategy which is able to obtain a good ratio between exploration and exploitation abilities of the algorithm. The impact of the introduced search operator on the performance of the ICFA is evaluated. Experiments are conducted on nineteen well-known benchmark functions. Results reveal that the ICFA is able to significantly improve the performance of the standard FA, CFA and four other recently proposed FA variants.

Fault diagnosis of sucker rod pumping systems based on Curvelet Transform and sparse multi-graph regularized extreme learning machine

Abstract: A novel approach is proposed to complete the fault diagnosis of pumping systems automatically. Fast Discrete Curvelet Transform is firstly adopted to extract features of dynamometer cards that sampled from sucker rod pumping systems, then a sparse multi-graph regularized extreme learning machine algorithm (SMELM) is proposed and applied as a classifier. SMELM constructs two graphs to explore the inherent structure of the dynamometer cards: the intra-class graph expresses the relationship among data from the same class and the inter-class graph expresses the relationship among data from different classes. By incorporating the information of the two graphs into the objective function of extreme learning machine (ELM), SMELM can force the outputs of data from the same class to be as same as possible and simultaneously force results from different classes to be as separate as possible. Different from previous ELM models utilizing the structure of data, our graphs are constructed through sparse representation instead of K-nearest Neighbor algorithm. Hence, there is no parameter to be decided when constructing graphs and the graphs can reflect the relationship among data more exactly. Experiments are conducted on dynamometer cards acquired on the spot. Results demonstrate the efficacy of the proposed approach for faults diagnosis in sucker rod pumping systems.

Design of PID controller for sun tracker system using QRAWCP approach

Abstract: In this paper, a direct formula is proposed for design of robust PID controller for sun tracker system using quadratic regulator approach with compensating pole (QRAWCP). The main advantage of the proposed approach is that, there is no need to use recently developed iterative soft computing techniques which are time consuming, computationally inefficient and also there is need to know boundary of search space. In order to show the superiority of the proposed approach, performance of the sun tracker system is compared with the recently applied tuning approaches for sun tracker systems such as particle swarm optimization, firefly algorithm and cuckoo search algorithm. The performance of the existing and proposed approaches are verified in time domain, frequency domain and also using integral performances indices. It is found that the performance is improved in transient, robustness, and uncertainty aspects in comparison to recently proposed soft computing approaches.

Multi-feature based event recommendation in Event-Based Social Network*

Abstract: As a new type of heterogeneous social network, Event-Based Social Network (EBSN) has experienced rapid development after its appearance. In EBSN, the interaction data between users and events is relatively sparse because of the short life cycle of events, which brings great challenges to event recommendation. In this paper, a multiple features based event recommendation method is proposed, which makes full use of various information in the network to mine users’ preference for event recommendation. Firstly, a heterogeneous information network model is constructed based on the intrinsic structure characteristics. Then multiple features about topology, temporal, spatial and semantic are extracted to measure the user’s event preference, and a linear scoring model is designed to acquire user’s preference score on events. At last, the bayesian personalized ranking method is used to learn the feature weights by using user-event pairs in scoring model and events are recommended to users according to the descending score order. Experiments are carried out on two real EBSN data sets, the results show that our approach can effectively alleviate the data sparseness problem and achieve better recommendation results.

A novel MAGDM approach with proportional hesitant fuzzy sets

Abstract: In this paper, we propose an extension of hesitant fuzzy sets, i.e., proportional hesitant fuzzy sets (PHFSs), with the purpose of accommodating proportional hesitant fuzzy environments. The components of PHFSs, which are referred to as proportional hesitant fuzzy elements (PHFEs), contain two aspects of information provided by a decision-making team: the possible membership degrees in the hesitant fuzzy elements and their associated proportions. Based on the PHFSs, we provide a novel approach to addressing fuzzy multi-attribute group decision making (MAGDM) problems. Different from the traditional approach, this paper first converts fuzzy MAGDM (expressed by classical fuzzy numbers) into proportional hesitant fuzzy multi-attribute decision making (represented by PHFEs), and then solves the latter through the proposal of a proportional hesitant fuzzy TOPSIS approach. In this process, preferences of the decision-making team are calculated as the proportions of the associated membership degrees. Finally, a numerical example and a comparison are provided to illustrate the reliability and effectiveness of the proposed approach.

Spontaneous Concept Learning with Deep Autoencoder

Abstract: In this study we investigate information processing in deep neural network models. We demonstrate that unsupervised training of autoencoder models of certain class can result in emergence of compact and structured internal representation of the input data space that can be correlated with higher level categories. We propose and demonstrate practical possibility to detect and measure this emergent information structure by applying unsupervised clustering in the activation space of the focal hidden layer of the model. Based on our findings we propose a new approach to training neural network models based on emergent in unsupervised training information landscape, that is iterative, driven by the environment, requires minimal supervision and with intriguing similarities to learning of biologic systems. We demonstrate its viability with originally developed method of spontaneous concept learning that yields good classification results while learning new higher level concepts with very small amounts of supervised training data.

CPO: A Crow Particle Optimization Algorithm

Abstract: Particle swarm optimization (PSO) is the most well known of the swarm-based intelligence algorithms and is inspired by the social behavior of bird flocking. However, the PSO algorithm converges prematurely, which rapidly decreases the population diversity, especially when approaching local optima. Recently, a new metaheuristic algorithm called the crow search algorithm (CSA)was proposed. The CSA is similar to the PSO algorithm but is based on the intelligent behavior of crows. Themain concept behind the CSA is that crows store excess food in hiding places and retrieve it when needed. The primary advantage of the CSA is that it is rather simple, having just two parameters: flight length and awareness probability. Thus, the CSA can be applied to optimization problems very easily. This paper proposes a hybridization algorithm based on the PSO algorithm and CSA, known as the crow particle optimization (CPO) algorithm. The two main operators are the exchange and local search operators. It also implements a local search operator to enhance the quality of the best solutions from the two systems. Simulation results demonstrated that the CPO algorithm exhibits a significantly higher performance in terms of both fitness value and computation time compared to other algorithms. © 2019 The Authors. Published by Atlantis Press SARL. This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

The Control of Greenhouses Based on Fuzzy Logic Using Wireless Sensor Networks

Abstract: Greenhouses cannot be easily controlled because their climate parameters are interrelated. This study contributes to increasing the quality and yield of greenhouses by saving time, energy, light and water consumption via measuring and controlling the climate parameters that are effective in forming climate factors in greenhouses. The greenhouse climate variables including temperature, relative humidity, soil moisture and light intensity were measured by a realistic sensor application. In this way, several sensor nodes, that belong to the nodal packages were distributed to a wireless sensor network (WSN) constructed in a star topology. In addition, the data obtained from the nodes, have been controlled and monitored with the fuzzy logic-based control strategy proposed as a developing, smart and remotely accessible Android-based interface. The proposed method has been analyzed, and its performances have been evaluated in terms of the benefits of both the user and the greenhouse.

A Multiple Attribute Decision Making Approach Based on New Similarity Measures of Interval-valued Hesitant Fuzzy Sets

Abstract: Hesitant fuzzy sets, as an extension of fuzzy sets to deal with uncertainty, have attracted much attention since its introduction, in both theory and application aspects. The present work is focused on the interval-valued hesitant fuzzy sets (IVHFSs) to manage additional uncertainty. Now that distance and similarity as a kind of information measures are essential and important numerical indexes in fuzzy set theory and all their extensions, the present work aims at investigating distance and similarity measures in the IVHFSs and then employing them into multiple attribute decision making application. To begin with, II-type generalized interval-valued hesitant fuzzy distance is firstly introduced in the IVHFS, along with its properties and its relationships with the traditional Hamming-Distance and the Euclidean distance. Afterwards, another interval-valued hesitant fuzzy Lp distance based on Lp metric is proposed and its relationship with the Hausdorff distance is discussed. In addition, different from most of similarity measures with dependent on the corresponding distances, a new similarity measure based on set-theoretic approach for IVHFSs is introduced and its properties are discussed; especially, a relative similarity measure is proposed based on the positive ideal IVHFS and the negative ideal IVHFS. Finally, we describe how the IVHFS and its relative similarity measure can be applied to multiple attribute decision making. A numerical example is then provided to illustrate the effectiveness of the proposed method.

Bare bones particle swarm optimization with adaptive chaotic jump for feature selection in classification

Abstract: Feature selection (FS) is a crucial data pre-processing process in classification problems. It aims to reduce the dimensionality of the problem by eliminating irrelevant or redundant features while achieve similar or even higher classification accuracy than using all the features. As a variant of particle swarm optimization (PSO), Bare bones particle swarm optimization (BBPSO) is a simple but very powerful optimizer. However, it also suffers from premature convergence like other PSO algorithms, especially in high-dimensional optimization problems. In order to improve its performance in FS problems, this paper proposes a novel BBPSO based FS method called BBPSOACJ. An adaptive chaotic jump strategy is designed to help the stagnated particles make a large change in their searching trajectory. It can enrich the search behavior of BBPSO and prevent the particles from being trapped into local attractors. A new global best updating mechanism is employed to reduce the size of obtained feature subset. The proposed BBPSO-ACJ is compared with eight evolutionary computation (EC) based wrapper methods and two filter methods on nine benchmark datasets with different number of dimensions and instances. The experimenta l results indicate that the proposed method can select the most discriminative features from the entire feature set and achieve significantly better classification performance than other comparative methods.

A Driving Behavior Awareness Model based on a Dynamic Bayesian Network and Distributed Genetic Algorithm

Abstract: It is necessary for automated vehicles (AVs) and advanced driver assistance systems (ADASs) to have a better understanding of the traffic environment including driving behaviors. This study aims to build a driving behavior awareness (DBA) model that can infer driving behaviors such as lane change. In this study, a dynamic Bayesian network DBA model is proposed, which includes three layers, namely, the observation, hidden and behavior layer. To enhance the performance of the DBA model, the network structure is optimized by employing a distributed genetic algorithm (GA). Using naturalistic driving data in Beijing, the comparison between the optimized model and other non-optimized models such as the hidden Markov model (HMM) and HMM with a mixture of Gaussian outputs (GM-HMM) indicates that the optimized model could estimate driving behaviors earlier and more accurately.