Journal of Intelligent & Fuzzy Systems - Volume 41, issue 6

Authors: Ngo, Quoc-Dung | Nguyen, Huy-Trung | Nguyen, Le-Cuong

Article Type: Research Article

Abstract: Over the last decade, due to exponential growth in IoT devices and weak security mechanisms, the IoT is now facing more security challenges than ever before, especially botnet malware. There are many security solutions in detecting botnet malware on IoT devices. However, detecting IoT botnet malware, particularly multi-architecture botnets, is challenging. This paper proposes a graphically structured feature extraction mechanism integrated with reinforcement learning techniques in multi-architecture IoT botnet detection. We then evaluate the proposed approach using a dataset of 22849 samples, including actual IoT botnet malware, and achieve a detection rate of 98.03 with low time consumption. The proposed …approach also achieves reliable results in detecting the new IoT botnet (has a new architecture-processor) not appearing in the training dataset at 96.69. To promote future research in the field, we share relevant datasets and source code. Show more

Keywords: IoT security, IoT botnet, reinforcement learning, static analysis, PSI-walk

DOI: 10.3233/JIFS-210699

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6801-6814, 2021

Authors: Song, Runze | Liu, Zhaohui | Wang, Chao

Article Type: Research Article

Abstract: As an advanced machine vision task, traffic sign recognition is of great significance to the safe driving of autonomous vehicles. Haze has seriously affected the performance of traffic sign recognition. This paper proposes a dehazing network, including multi-scale residual blocks, which significantly affects the recognition of traffic signs in hazy weather. First, we introduce the idea of residual learning, design the end-to-end multi-scale feature information fusion method. Secondly, the study used subjective visual effects and objective evaluation metrics such as Visibility Index (VI) and Realness Index (RI) based on the characteristics of the real-world environment to compare various traditional dehazing …and deep learning dehazing method with good performance. Finally, this paper combines image dehazing and traffic sign recognition, using the algorithm of this paper to dehaze the traffic sign images under real-world hazy weather. The experiments show that the algorithm in this paper can improve the performance of traffic sign recognition in hazy weather and fulfil the requirements of real-time image processing. It also proves the effectiveness of the reformulated atmospheric scattering model for the dehazing of traffic sign images. Show more

Keywords: Deep learning, image processing, dehazing of real-world, traffic sign recognition, reformulated atmospheric scattering model

DOI: 10.3233/JIFS-210733

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6815-6830, 2021

Authors: Qarehkhani, Azam | Golsorkhtabaramiri, Mehdi | Mohamadi, Hosein | Yadollahzadeh Tabari, Meisam

Article Type: Research Article

Abstract: Directional sensor networks (DSNs) are classified under wireless networks that are largely used to resolve the coverage problem. One of the challenges to DSNs is to provide coverage for all targets in the network and, at the same time, to maximize the lifetime of network. A solution to this problem is the adjustment of the sensors’ sensing ranges. In this approach, each sensor adjusts its own sensing range dynamically to sense the corresponding target(s) and decrease energy consumption as much as possible through forming the best cover sets possible. In the current study, a continuous learning automata-based method is proposed …to form such cover sets. To assess the proposed algorithm’s performance, it was compared to the results obtained from a greedy algorithm and a learning automata algorithm. The obtained results demonstrated the superiority of the proposed algorithm regarding the maximization of the network lifetime. Show more

Keywords: Directional sensor networks, continuous learning automata, target-coverage, cover set formation

DOI: 10.3233/JIFS-210759

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6831-6844, 2021

Authors: Han, Guojiang | Chen, Caikou | Xu, Zhixuan | Zhou, Shengwei

Article Type: Research Article

Abstract: Ensemble learning using a set of deep convolutional neural networks (DCNNs) as weak classifiers has become a powerful tool for face expression. Nevertheless, training a DCNNS-based ensemble is not only time consuming but also gives rise to high redundancy due to the nature of DCNNs. In this paper, a novel DCNNs-based ensemble method, named weighted ensemble with angular feature learning (WDEA), is proposed to improve the computational efficiency and diversity of the ensemble. Specifically, the proposed ensemble consists of four parts including input layer, trunk layers, diversity layers and loss fusion. Among them, the trunk layers which are used to …extract the local features of face images are shared by diversity layers such that the lower-level redundancy can be largely reduced. The independent branches enable the diversity of the ensemble. Rather than the traditional softmax loss, the angular softmax loss is employed to extract more discriminant deep feature representation. Moreover, a novel weighting technique is proposed to enhance the diversity of the ensemble. Extensive experiments were performed on CK+ and AffectNet. Experimental results demonstrate that the proposed WDEA outperforms existing ensemble learning methods on the recogntion rate and computational efficiency. Show more

Keywords: Facial expression recognition, ensemble-based CNN, end to end learning, weight matrix unit

DOI: 10.3233/JIFS-210762

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6845-6857, 2021

Authors: Tao, Yujie | Suo, Chunfeng | Wang, Guijun

Article Type: Research Article

Abstract: Piecewise linear function (PLF) is not only a generalization of univariate segmented linear function in multivariate case, but also an important bridge to study the approximation of continuous function by Mamdani and Takagi-Sugeno fuzzy systems. In this paper, the definitions of the PLF and subdivision are introduced in the hyperplane, the analytic expression of PLF is given by using matrix determinant, and the concept of approximation factor is first proposed by using m -mesh subdivision. Secondly, the vertex coordinates and their changing rules of the n -dimensional small polyhedron are found by dividing a three-dimensional cube, and the algebraic cofactor …and matrix norm of corresponding determinants of piecewise linear functions are given. Finally, according to the method of solving algebraic cofactors and matrix norms, it is proved that the approximation factor has nothing to do with the number of subdivisions, but the approximation accuracy has something to do with the number of subdivisions. Furthermore, the process of a specific binary piecewise linear function approaching a continuous function according to infinite norm in two dimensions space is realized by a practical example, and the validity of PLFs to approximate a continuous function is verified by t -hypothesis test in Statistics. Show more

Keywords: Piecewise linear function, mesh subdivision, approximation factor, Mamdani fuzzy system, matrix norm

DOI: 10.3233/JIFS-210770

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6859-6873, 2021

Authors: Ding, Ling | Chen, Xiaojun | Xiang, Yang

Article Type: Research Article

Abstract: Few-shot text classification aims to learn a classifier from very few labeled text data. Existing studies on this topic mainly adopt prototypical networks and focus on interactive information between support set and query instances to learn generalized class prototypes. However, in the process of encoding, these methods only pay attention to the matching information between support set and query instances, and ignore much useful information about intra-class similarity and inter-class dissimilarity between all support samples. Therefore, in this paper we propose a negative-supervised capsule graph neural network (NSCGNN) which explicitly takes use of the similarity and dissimilarity between samples to …make the text representations of the same type closer with each other and the ones of different types farther away, leading to representative and discriminative class prototypes. We firstly construct a graph to obtain text representations in the form of node capsules, where both intra-cluster similarity and inter-cluster dissimilarity between all samples are explored with information aggregation and negative supervision. Then, in order to induce generalized class prototypes based on those node capsules obtained from graph neural network, the dynamic routing algorithm is utilized in our model. Experimental results demonstrate the effectiveness of our proposed NSCGNN model, which outperforms existing few-shot approaches on three benchmark datasets. Show more

Keywords: Graph neural networks, negative supervision, dynamic routing, few-shot learning

DOI: 10.3233/JIFS-210795

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6875-6887, 2021

Authors: Alshammari, Ibtesam | Parimala, Mani | Jafari, Saeid

Article Type: Research Article

Abstract: Imprecision in the decision-making process is an essential consideration. In order to navigate the imprecise decision-making framework, measuring tools and methods have been developed. Pythagorean fuzzy soft sets are one of the new methods for dealing with imprecision. Pythagorean fuzzy soft topological spaces is an extension of intuitionistic fuzzy soft topological spaces. These sets generalizes intuitionistic fuzzy sets for a broader variety of implementations. This work is a gateway to study such a problem. The concept of Pythagorean fuzzy soft topological spaces(PyFSTS), interior, closure, boundary, neighborhood of Pythagorean fuzzy soft spaces PyFSS, base and subspace of PyFSTSs are presented and …its properties are figured out. We established an algorithm under uncertainty based on PyFSTS for multi-attribute decision-making (MADM) and to validate this algorithm, a numerical example is solved for suitable brand selection. Finally, the benefits, validity, versatility and comparison of our proposed algorithms with current techniques are discussed.The advantage of the proposed work is to detect vagueness with more sizably voluminous valuation space than intuitionistic fuzzy sets. Show more

Keywords: Pythagorean fuzzy soft sets, Pythagorean fuzzy soft topology, Pythagorean fuzzy soft interior and soft closure, multi-attribute decision making

DOI: 10.3233/JIFS-210805

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6889-6897, 2021

Authors: Yuan, Ling | Pan, Zhuwen | Sun, Ping | Wei, Yinzhen | Yu, Haiping

Article Type: Research Article

Abstract: Click-through rate (CTR) prediction, which aims to predict the probability of a user clicking on an ad, is a critical task in online advertising systems. The problem is very challenging since(1) an effective prediction relies on high-order combinatorial features, and(2)the relationship to auxiliary ads that may impact the CTR. In this paper, we propose Deep Context Interaction Network on Attention Mechanism(DCIN-Attention) to process feature interaction and context at the same time. The context includes other ads in the current search page, historically clicked and unclicked ads of the user. Specifically, we use the attention mechanism to learn the interactions between …the target ad and each type of auxiliary ad. The residual network is used to model the feature interactions in the low-dimensional space, and with the multi-head self-attention neural network, high-order feature interactions can be modeled. Experimental results on Avito dataset show that DCIN outperform several existing methods for CTR prediction. Show more

Keywords: Click-through rate, attention mechanism, residual network, feature interaction, context

DOI: 10.3233/JIFS-210830

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6899-6914, 2021

Authors: Tang, Han | Dalin,

Article Type: Research Article

Abstract: Uncertain time series analysis has been developed for studying the imprecise observations. In this paper, we propose a nonlinear model called uncertain max-autoregressive (UMAR) model. The unknown parameters in model are estimated by the least squares estimation. Then the residual analysis is presented. In many cases, there are some outliers in the time series due to short-term change in the underlying process. The UMAR model offers an alternative for detecting outliers in the imprecise observations. Based on the previous theoretical results, the UMAR model is used to forecast the future. Finally, an example suggests that the new proposed time series …model works well compared to the uncertain autoregressive (UAR) model. Show more

Keywords: Uncertain time series analysis, principle of least squares, residual analysis, outlier detection, confidence interval

DOI: 10.3233/JIFS-210848

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6915-6922, 2021

Authors: Maity, Suman | De, Sujit Kumar | Pal, Madhumangal | Mondal, Sankar Prasad

Article Type: Research Article

Abstract: This article deals with an economic order quantity inventory model of imperfect items under non-random uncertain demand. Here we consider the customers screen the imperfect items during the selling period. After a certain period of time, the imperfect items are sold at a discounted price. We split the model into three cases, assuming that the demand rate increases, decreases, and is constant in the discount period. Firstly, we solve the crisp model, and then the model is converted into a fuzzy environment. Here we consider the dense fuzzy, parabolic fuzzy, degree of fuzziness and cloudy fuzzy for a comparative study. …The basic novelty of this paper is that a computer-based algorithm and flow chart have been given for the solution of the proposed model. Finally, sensitivity analysis and graphical illustration have been given to check the validity of the model. Show more

Keywords: Imperfect inventory, dense fuzzy number, parabolic fuzzy number, cloudy fuzzy number, degree of fuzziness, optimization

DOI: 10.3233/JIFS-210856

Citation: Journal of Intelligent & Fuzzy Systems, vol. 41, no. 6, pp. 6923-6934, 2021