Journal of Intelligent & Fuzzy Systems - Volume 44, issue 5

Authors: Ma, Hua | Tang, Yuqi | Zhang, Xuxiang | Zhu, Haibin | Huang, Peiji | Zhang, Hongyu

Article Type: Research Article

Abstract: An e-learning system should recommend learners appropriate learning resources according to their actual needs and cognitive status for improving their learning performance. To overcome the deficiencies of existing approaches (e.g., poor interpretability, limited efficiency and accuracy of recommendation), we propose a new recommendation approach to learning resources via knowledge graphs and learning style clustering. In this approach, the knowledge graphs of an online learning environment are constructed based on a generic ontology model, and the graph embedding algorithm and graph matching process are applied to optimize the efficiency of graph computation for identifying similar learning resources. By introducing learning style …theory, learners are clustered based on their learning styles. Based on the clustering results, the learners’ degrees of interest in similar learning resources are measured, and the recommendation results are obtained according to the degrees of interest. Finally, the experiments demonstrate that the proposed approach significantly enhances the computational efficiency and the quality of learning resource recommendation compared with the existing approaches in large-scale graph data scenarios. Show more

Keywords: Knowledge graphs, learning resource recommendation, learning style clustering, personalized learning

DOI: 10.3233/JIFS-222627

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8053-8069, 2023

Authors: Alnssyan, Badr | Hussein, Ekramy A. | Alizadeh, Morad | Afify, Ahmed Z. | Abdellatif, Ashraf D.

Article Type: Research Article

Abstract: We propose a new wider family called the weighted Lindley-G family. We derive some mathematical properties and special sub-models of the new family. We address the estimation of the model parameters by eight approaches of estimation. The estimation approaches are ranked and compared by using detailed simulations to develop a guideline for choosing the best approach for estimating the distribution parameters. The potentiality of the new family is illustrated via two applications to real-life data. It is shown that the proposed WLi-G family is more flexible as compared to some of the most cited families in the distribution theory literature …such as the exponentiated-G, beta-G, transmuted-G, and alpha-power-G families under the same baseline model. Show more

Keywords: Rényi entropy, exponential distribution, data analysis, Anderson–Darling estimation, maximum likelihood

DOI: 10.3233/JIFS-222758

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8071-8089, 2023

Authors: Duan, Chen | Liu, Yongli

Article Type: Research Article

Abstract: In fuzzy clustering algorithms, the possibilistic fuzzy clustering algorithm has been widely used in many fields. However, the traditional Euclidean distance cannot measure the similarity between samples well in high-dimensional data. Moreover, if there is an overlap between clusters or a strong correlation between features, clustering accuracy will be easily affected. To overcome the above problems, a collaborative possibilistic fuzzy clustering algorithm based on information bottleneck is proposed in this paper. This algorithm retains the advantages of the original algorithm, on the one hand, using mutual information loss as the similarity measure instead of Euclidean distance, which is conducive to …reducing subjective errors caused by arbitrary choices of similarity measures and improving the clustering accuracy; on the other hand, the collaborative idea is introduced into the possibilistic fuzzy clustering based on information bottleneck, which can form an accurate and complete representation of the data organization structure based on make full use of the correlation between different feature subsets for collaborative clustering. To examine the clustering performance of this algorithm, five algorithms were selected for comparison experiments on several datasets. Experimental results show that the proposed algorithm outperforms the comparison algorithms in terms of clustering accuracy and collaborative validity. Show more

Keywords: Possibilistic fuzzy clustering, collaborative clustering, information bottleneck, similarity measure

DOI: 10.3233/JIFS-223854

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8091-8102, 2023

Authors: Fu, Xingyu | Chen, Yingyue | Yan, Jingru | Chen, Yumin | Xu, Feng

Article Type: Research Article

Abstract:  The random forest is a combined classification method belonging to ensemble learning. The random forest is also an important machine learning algorithm. The random forest is universally applicable to most data sets. However, the random forest is difficult to deal with uncertain data, resulting in poor classification results. To overcome these shortcomings, a broad granular random forest algorithm is proposed by studying the theory of granular computing and the idea of breadth. First, we granulate the breadth of the relationship between the features of the data sets samples and then form a broad granular vector. In addition, the operation rules …of the granular vector are defined, and the granular decision tree model is proposed. Finally, the multiple granular decision tree voting method is adopted to obtain the result of the granular random forest. Some experiments are carried out on several UCI data sets, and the results show that the classification performance of the broad granular random forest algorithm is better than that of the traditional random forest algorithm. Show more

Keywords: Granular computing, Broad granular vector, Granular decision tree, Granular random forest, Classification

DOI: 10.3233/JIFS-223960

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8103-8117, 2023

Authors: Sun, Qiong | Jiang, Jingjing | Wang, Zhongsheng | Liao, Bin

Article Type: Research Article

Abstract: By using the interactive big data between enterprises and stakeholders in social media, this paper investigates the views of different stakeholders on the disclosure of enterprise digital transformation. In view of the social media platform brings together different stakeholders, this paper uses the organizational hypocrisy theory to explore the stakeholders’ Reflection on the hypocritical speech, decision-making and action strategies adopted in the disclosure of enterprise digital transformation. Through data mining and computer-aided emotion analysis, the posts of sina Weibo’s top 500 Chinese enterprises from December 31, 2020 to December 31, 2021 and the reactions of stakeholders are retrieved and analyzed. …It is found that stakeholders have different reactions to the hypocrisy strategies of enterprises. Although stakeholders pay more attention to information related to actions, and the disclosure of such actions will cause positive and negative reactions, the inconsistency of speech and decision-making will produce positive reactions and reduce negative impressions. Overall, research shows that the use of organizational hypocrisy strategies in social media can enable enterprises to manage the views and legitimacy of stakeholders. Show more

Keywords: Organizational hypocrisy, digital transformation, stakeholder, emotional analysis

DOI: 10.3233/JIFS-224092

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8119-8132, 2023

Authors: Geng, Xiaonan | Liu, Peng

Article Type: Research Article

Abstract: This article has been retracted. A retraction notice can be found at https://doi.org/10.3233/JIFS-219433 .

DOI: 10.3233/JIFS-224316

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8133-8145, 2023

Authors: Gong, Zengtai | Wang, Fangdi

Article Type: Research Article

Abstract: Complex fuzzy set, as an extension of classical fuzzy sets, could describe the fuzzy characters of things more detail and comprehensively and is very useful in dealing with vagueness and uncertainty of problems that include the periodic or recurring phenomena. Note that a complex fuzzy set is different from the fuzzy complex set introduced and discussed by many scholars, since the membership degree of a complex fuzzy set is a complex number with length less than or equal to 1 while a fuzzy complex set is a real number with membership degree less than or equal to 1, and the …universe is the complex plane. As the mathematical theoretical basis of fuzzy mathematics, fuzzy set and its mapping, corresponding fuzzy complex set and its mapping have been investigated in depth because they integrate and cross the methods and results of classical real analysis and complex analysis. However, there is no comprehensive investigation on complex fuzzy set and its corresponding mathematical theory, even include decomposition theorems, extension principles and the basic operations of the complex fuzzy set. As is well known, the cut set of fuzzy sets is the bridge between fuzzy sets and classical sets, which plays a significant role in fuzzy sets and fuzzy systems. In this paper, the concept of (r , θ)-cut sets of complex fuzzy sets is proposed and their properties are discussed. Meanwhile, the decomposition theorems and the extension principles of complex fuzzy set based on (r , θ)-cut sets are deduced and corresponding properties are investigated. All these conclusions not only deeply enrich the fundamental theory of complex fuzzy set, but also provide a powerful tool to investigate complex fuzzy set. Finally, an example application of signal detection demonstrates the utility of the (r , θ)-cut sets of complex fuzzy sets in practice. Show more

Keywords: Complex fuzzy set, (r, θ)-cut sets, decomposition theorems, extension principles

DOI: 10.3233/JIFS-221639

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8147-8162, 2023

Authors: Wang, Wenpu | Lin, Wei | Gao, Fengxiang | Chang, Shuli

Article Type: Research Article

Abstract: Business English teaching quality evaluation Business English is a new type of composite specialty, which is a discipline innovation made by China’s higher education to adapt to the new market demand and international standards since the reform and opening up. Over the past 20 years, it has cultivated a number of compound talents for the cause of China’s reform and opening up. However, the backwardness of business English theoretical research has greatly restricted the development of business English. At present, Business English has been officially approved as a new major for undergraduate enrollment by the Ministry of Education of the …People’s Republic of China. Its subject nature, specialty structure, training objectives, and specialty compound characteristics need to be qualitatively studied theoretically. The business English teaching quality evaluation is viewed as the multiple attribute decision making (MADM) issue. In this paper, we connect the geometric Heronian mean (GHM) operator and power geometric (PG) with 2-tuple linguistic neutrosophic numbers (2TLNNs) to propose the generalized 2-tuple linguistic neutrosophic power geometric HM (G2TLNPGHM) operator. Then, the G2TLNGHM operator is applied to deal with the MADM problems under 2TLNNs. Finally, an example for business English teaching quality evaluation is used to show the proposed methods. Some comparative analysis and parameter influence analysis are fully given. The results show that the built algorithms method is useful for business English teaching quality evaluation. Show more

Keywords: Multiple attribute decision making (MADM), 2-tuple linguistic neutrosophic numbers set (2TLNNSs), G2TLNPGHM operator, business English teaching quality

DOI: 10.3233/JIFS-223850

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8163-8175, 2023

Authors: Singh, Pardeep | Singh, Nitin Kumar | Monika, | Chand, Satish

Article Type: Research Article

Abstract: One major issue plaguing online social media is hate speech, a complex phenomenon whose identification and target categorization have been studied by the natural language processing community. In recent years, notable studies have been made towards hate speech detection using various mechanisms varying from traditional machine learning to complex deep neural network models. However, these studies mainly focus on high-resource English language. The multilingual societies such as the Indian subcontinent: English, Hindi and Hindi-English code-mixed languages are widespread and convenient for the users. The research works studying hate speech detection in these languages are still very limited. To fill this …gap, we propose an mBERT-GRU framework comprising of multilingual BERT embedding and bidirectional GRU layers to learn the cumulative features for hate speech detection and its target categorization. We evaluated our work on three datasets HASOC-2019, HS and HEOT to prove the competitive performance. Our results show that the proposed framework outperformed monolingual and state-of-the-art methods on English, Hindi and Hindi-English code-mixed datasets with Macro-F1 measure values of 0.87, 0.83 and 0.77, respectively. Show more

Keywords: Deep learning, GRU, hate speech detection, multilingual BERT, social media, text analysis

DOI: 10.3233/JIFS-222057

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8177-8192, 2023

Authors: Chughtai, Iqra Toheed | Naseer, Asma | Tamoor, Maria | Asif, Saara | Jabbar, Mamoona | Shahid, Rabia

Article Type: Research Article

Abstract: In the past few years, due to the increased usage of internet, smartphones, sensors and digital cameras, more than a million images are generated and uploaded daily on social media platforms. The massive generation of such multimedia contents has resulted in an exponential growth in the stored and shared data. Certain ever-growing image repositories, consisting of medical images, satellites images, surveillance footages, military reconnaissance, fingerprints and scientific data etc., has increased the motivation for developing robust and efficient search methods for image retrieval as per user requirements. Hence, it is need of the hour to search and retrieve relevant images …efficiently and with good accuracy. The current research focuses on Content-based Image Retrieval (CBIR) and explores well-known transfer learning-based classifiers such as VGG16, VGG19, EfficientNetB0, ResNet50 and their variants. These deep transfer leaners are trained on three benchmark image datasets i.e., CIFAR-10, CIFAR-100 and CINIC-10 containing 10, 100, and 10 classes respectively. In total 16 customized models are evaluated on these benchmark datasets and 96% accuracy is achieved for CIFAR-10 while 83% accuracy is achieved for CIFAR-100. Show more

Keywords: CBIR, transfer learning, CNN, VGG-16, VGG-19, ResNet-50, EfficientNet, deep learning

DOI: 10.3233/JIFS-223449

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 8193-8218, 2023