Journal of Intelligent & Fuzzy Systems - Volume 40, issue 3

Authors: Li, Bing | Xiao, Binqing | Yang, Yang

Article Type: Research Article

Abstract: This study identifies credit risk sources, credit scoring index classification modes and modelling methods, and constructs a credit scoring system for small and micro businesses (SMBs) with soft information. Through the analysis and comparison of neural network models, this study demonstrates the superiority of the back-propagation neural network (BPNN) models for loan classification prediction. There are three contributions and innovations as follows. (1) Based on the actual demands and default characteristics of SMBs, this study adds the behavioural variables of loan managers to strengthen the role of soft information in credit scoring model. (2) It develops a hybrid analysis and …comparison framework based on the BPNN model. It identifies that the BPNN model is more suitable for approving SMB loans, as it can precisely identify the second type of error. (3) Using the precious ledger data of SMB loans from a rural commercial bank in Jiangsu province, China, this study compares the prediction accuracy of the credit scoring model based on BPNN using two-level and five-level loan classifications. Further, it illustrates the applicability of the BPNN model. By connecting the practical operations of credit scoring and quantitative models, this paper supports commercial bank examination and approval work of SMB loans. Show more

Keywords: Credit scoring, small and micro businesses, soft information, back-propagation neural network, comparative analysis

DOI: 10.3233/JIFS-200866

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4257-4274, 2021

Authors: Khoh, Wee How | Pang, Ying Han | Ooi, Shih Yin | Yap, Hui Yen

Article Type: Research Article

Abstract: Dynamic signature recognition emerges to perfectly solve the hygiene concern due to its no-contact characteristic. Nevertheless, the recognition of dynamic texture is challenging compared with the static signature image due to their unknown spatial and temporal nature. In this work, we present a multi-view spatiotemporal approach based on spectral histogramming for hand gesture signature recognition. A Microsoft Kinect sensor is adopted to capture the motion of signing in a sequence of depth frames. The depth frame sequence is viewed from three directional sights to retrieve rich information, such as temporal changes at each spatial location, the signing motion flow of …each vertical and horizontal spatial space in a temporal manner. Furthermore, the proposed approach performs feature description on different levels of locality. This function enables a multi-resolution analysis on this dynamic signature. The robustness of the proposed approach is reflected with the promising result by striking the state-of-the-art performance, as substantiated in the empirical results. Show more

Keywords: Hand gesture signature, dynamic signature, biometrics, spatiotemporal, gesture recognition

DOI: 10.3233/JIFS-200908

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4275-4286, 2021

Authors: Pei, Cong | Jiang, Feng | Li, Mao

Article Type: Research Article

Abstract: With the advent of cost-efficient depth cameras, many effective feature descriptors have been proposed for action recognition from depth sequences. However, most of them are based on single feature and thus unable to extract the action information comprehensively, e.g., some kinds of feature descriptors can represent the area where the motion occurs while they lack the ability of describing the order in which the action is performed. In this paper, a new feature representation scheme combining different feature descriptors is proposed to capture various aspects of action cues simultaneously. First of all, a depth sequence is divided into a series …of sub-sequences using motion energy based spatial-temporal pyramid. For each sub-sequence, on the one hand, the depth motion maps (DMMs) based completed local binary pattern (CLBP) descriptors are calculated through a patch-based strategy. On the other hand, each sub-sequence is partitioned into spatial grids and the polynormals descriptors are obtained for each of the grid sequences. Then, the sparse representation vectors of the DMMs based CLBP and the polynormals are calculated separately. After pooling, the ultimate representation vector of the sample is generated as the input of the classifier. Finally, two different fusion strategies are applied to conduct fusion. Through extensive experiments on two benchmark datasets, the performance of the proposed method is proved better than that of each single feature based recognition method. Show more

Keywords: Action recognition, feature fusion, depth motion maps, completed local binary pattern, polynormal

DOI: 10.3233/JIFS-200954

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4287-4299, 2021

Authors: Zhang, Zhaojun | Li, Xuanyu | Luan, Shengyang | Xu, Zhaoxiong

Article Type: Research Article

Abstract: Particle swarm optimization (PSO) as a successful optimization algorithm is widely used in many practical applications due to its advantages in fast convergence speed and convenient implementation. As a population optimization algorithm, the quality of initial population plays an important role in the performance of PSO. However, random initialization is used in population initialization for PSO. Using the solution of the solved problem as prior knowledge will help to improve the quality of the initial population solution. In this paper, we use homotopy analysis method (HAM) to build a bridge between the solved problems and the problems to be solved. …Therefore, an improved PSO framework based on HAM, called HAM-PSO, is proposed. The framework of HAM-PSO includes four main processes. It contains obtaining the prior knowledge, constructing homotopy function, generating initial solution and solving the to be solved by PSO. In fact, the framework does not change the PSO, but replaces the random population initialization. The basic PSO algorithm and three others typical PSO algorithms are used to verify the feasibility and effectiveness of this framework. The experimental results show that the four PSO using this framework are better than those without this framework. Show more

Keywords: Particle swarm optimization, homotopy analysis method, initial population, t-test

DOI: 10.3233/JIFS-200979

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4301-4315, 2021

Authors: Mao, Hua | Cheng, Yilin

Article Type: Research Article

Abstract: Three-way decisions, as a better way than two-way decisions, has played an important role in many fields. As an extension of formal concept, rough semiconcept constitutes a new approach for data analysis. By now, three-way concept, which combines three-way decisions with formal concept, has been an efficient tool for knowledge representation problems. Hence, we want to further apply three-way decisions to rough semiconcept. In this work, we introduce three-way rough semiconcept by an example, which combines rough semiconcept with the assistant of three-way decisions. After that, we attain the structure of all three-way rough semiconcepts from an algebraic perspective. Furthermore, …we give two kinds of approximation operators, which can characterize three-way rough semiconcepts. Finally, we present algorithms for searching three-way rough semiconcepts. An example is to demonstrate the correct and effective of algorithms in this paper. Show more

Keywords: Three-way decisions, semiconcept, rough set, lattice theory

DOI: 10.3233/JIFS-200981

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4317-4330, 2021

Authors: Gao, Weiqi | Huang, Hao

Article Type: Research Article

Abstract: Graph convolutional networks (GCNs), which are capable of effectively processing graph-structural data, have been successfully applied in text classification task. Existing studies on GCN based text classification model largely concerns with the utilization of word co-occurrence and Term Frequency-Inverse Document Frequency (TF–IDF) information for graph construction, which to some extent ignore the context information of the texts. To solve this problem, we propose a gating context-aware text classification model with Bidirectional Encoder Representations from Transformers (BERT) and graph convolutional network, named as Gating Context GCN (GC-GCN). More specifically, we integrate the graph embedding with BERT embedding by using a GCN …with gating mechanism to enable the acquisition of context coding. We carry out text classification experiments to show the effectiveness of the proposed model. Experimental results shown our model has respectively obtained 0.19%, 0.57%, 1.05% and 1.17% improvements over the Text-GCN baseline on the 20NG, R8, R52, and Ohsumed benchmark datasets. Furthermore, to overcome the problem that word co-occurrence and TF–IDF are not suitable for graph construction for short texts, Euclidean distance is used to combine with word co-occurrence and TF–IDF information. We obtain an improvement by 1.38% on the MR dataset compared to Text-GCN baseline. Show more

Keywords: Text classification, graph convolutional network, BERT, gating mechanism, Euclidean distance

DOI: 10.3233/JIFS-201051

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4331-4343, 2021

Authors: Zhou, Jun | Peng, Jinghong | Liang, Guangchuan | Chen, Chuan | Zhou, Xuan | Qin, Yixiong

Article Type: Research Article

Abstract: Natural gas transmission network is the major facility connecting the upstream gas sources and downstream consumers. In this paper, a multi-objective optimization model is built to find the optimum operation scheme of the natural gas transmission network. This model aims to balance two conflicting optimization objective named maximum a specified node delivery flow rate and minimum compressor station power consumption cost. The decision variables involve continuous and discrete variables, including node delivery flow rate, number of running compressors and their rotational speed. Besides, a series of equality and inequality constraints for nodes, pipelines and compressor stations are introduced to control …the optimization results. Then, the developed optimization model is applied to a practical large tree-topology gas transmission network, which is 2,229 km in length with 7 compressor stations, 2 gas injection nodes and 20 gas delivery nodes. The ɛ -constraint method and GAMS/DICOPT solver are adopted to solve the bi-objective optimization model. The optimization result obtained is a set of Pareto optimal solutions. To verify the validity of the proposed method, the optimization results are compared with the actual operation scheme. Through the comparison of different Pareto optimal solutions, the variation law of objective functions and decision variables between different optimal solutions are clarified. Finally, sensitivity analyses are also performed to determine the influence of operating parameter changes on the optimization results. Show more

Keywords: Natural gas transmission network, operation optimization, compressor station, multi-objective optimization, ɛ-constraint method

DOI: 10.3233/JIFS-201072

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4345-4366, 2021

Authors: Sha, Xiuyan | Yin, Chuancun | Xu, Zeshui | Zhang, Shen

Article Type: Research Article

Abstract: In order to fully consider the decision-maker’s limited rationality and attitude to risk, this paper constructs the probabilistic hesitant fuzzy TOPSIS emergency decision-making model based on the cumulative prospect theory under the probabilistic hesitant fuzzy environment. Aiming at the problem of missing probabilistic information in the probabilistic hesitant fuzzy element, a new complement scheme is proposed. In this scheme, the weighted average result of the original data information is used to complement, and the original data information is retained to a large extent. Then this paper proposes several probabilistic hesitant fuzzy distance measures based on Lance distance. The decision reference …point is constructed by the probabilistic hesitant fuzzy Lance distance, which overcomes the influence of the extreme value on the decision-making result, and defines the value function based on the probabilistic hesitant fuzzy Lance distance. In view of the fact that the attribute weights are completely unknown, the probabilistic hesitant fuzzy exponential entropy is constructed by using the actual data, and the attribute weights of different prospect states are obtained. Aiming at the problem that attribute weights of different prospect states have different effects on the cumulative prospect value, the expression of the cumulative prospect value is improved. The improved closeness coefficient of the TOPSIS method is used to order the emergency schemes. Finally, the new method is applied to the emergency decision-making case of a sudden outbreak of epidemic respiratory disease. The results show that the contrast of the new method is obvious, which is conducive to distinguish different schemes. The new method is more suitable for the complex and changeable emergency decision-making field. Show more

Keywords: Probabilistic hesitant fuzzy Lance distance, probabilistic hesitant fuzzy exponential entropy, cumulative prospect theory, TOPSIS

DOI: 10.3233/JIFS-201119

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4367-4383, 2021

Authors: Elaskily, Mohamed A. | Alkinani, Monagi H. | Sedik, Ahmed | Dessouky, Mohamed M.

Article Type: Research Article

Abstract: Protecting information from manipulation is important challenge in current days. Digital images are one of the most popular information representation. Images could be used in several fields such as military, social media, security purposes, intelligence fields, evidences in courts, and newspapers. Digital image forgeries mean adding unusual patterns to the original images that cause a heterogeneity manner in form of image properties. Copy move forgery is one of the hardest types of image forgeries to be detected. It is happened by duplicating part or section of the image then adding again in the image itself but in another location. Forgery …detection algorithms are used in image security when the original content is not available. This paper illustrates a new approach for Copy Move Forgery Detection (CMFD) built basically on deep learning. The proposed model is depending on applying (Convolution Neural Network) CNN in addition to Convolutional Long Short-Term Memory (CovLSTM) networks. This method extracts image features by a sequence number of Convolutions (CNVs) layers, ConvLSTM layers, and pooling layers then matching features and detecting copy move forgery. This model had been applied to four aboveboard available databases: MICC-F220, MICC-F2000, MICC-F600, and SATs-130. Moreover, datasets have been combined to build new datasets for all purposes of generalization testing and coping with an over-fitting problem. In addition, the results of applying ConvLSTM model only have been added to show the differences in performance between using hybrid ConvLSTM and CNN compared with using CNN only. The proposed algorithm, when using number of epoch’s equal 100, gives high accuracy reached to 100% for some datasets with lowest Testing Time (TT) time nearly 1 second for some datasets when compared with the different previous algorithms. Show more

Keywords: Convolutional Long Short-Term Memory (CovLSTM), copy-move forgery detection, image authentication, tampered images, deep learning, and convolutional neural networks

DOI: 10.3233/JIFS-201192

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4385-4405, 2021

Authors: Chen, Chuanming | Zhang, Shuanggui | Yu, Qingying | Ye, Zitong | Ye, Zhen | Hu, Fan

Article Type: Research Article

Abstract: The analysis of user trajectory information and social relationships in social media, combined with the personalization of travel needs, allows users to better plan their travel routes. However, existing methods take only local factors into account, which results in a lack of pertinence and accuracy for the recommended route. In this study, we propose a method by which user clustering, improved genetic, and rectangular region path planning algorithms are combined to design personalized travel routes for users. First, the social relationships of users are analyzed, and close friends are clustered into categories to obtain several friend clusters. Next, the historical …trajectory data of users in the cluster are analyzed to obtain joint points in the trajectory map, these are matched according to the keywords entered by users. Finally, the search area is narrowed and the recommended travel route is obtained through improved genetic and rectangular region path planning algorithms. Theoretical analyses and experimental evaluations show that the proposed method is more accurate at path prediction and regional coverage than other methods. In particular, the average area coverage rate of the proposed method is better than that of the existing algorithm, with a maximum increasement ratio of 31.80%. Show more

Keywords: Tourism route, genetic algorithm, personalized recommendation, route planning

DOI: 10.3233/JIFS-201218

Citation: Journal of Intelligent & Fuzzy Systems, vol. 40, no. 3, pp. 4407-4423, 2021