Search

Fuzzy C-Means is a well-known fuzzy clustering technique. Although FCM can cover the uncertainty problem by forming overlapping clusters, it involves issues such as sensitivity to noise and outliers, the fuzzification parameter m, and the initial guess of cluster centers. There are various improvements made to FCM to tackle these limitations. One such extension, FCOM, has proved to be an efficient method for handling noise and outliers. However, it still suffers from the last two ones. In the present paper, inspired by FCOM, a new model is designed to solve these issues. Whereas varying the degree of fuzziness m leads to different clusters, determining the appropriate value optimizes solutions. However, manually tuning this parameter can be time-consuming, especially when dealing with large data sets. To mitigate the dependence on this parameter, the proposed model utilizes the entropy theory to control the uncertainty associated with the input data. Extensive evaluations are conducted on benchmark datasets to analyze the impact of m on cluster formation and clustering results. The competitive results confirm the effectiveness of the proposed model for handling fuzziness degree and its capability to accelerate convergence to optimal solutions. Moreover, the results show that the proposed model discovers vague boundaries precisely. [ABSTRACT FROM AUTHOR]

The exponential spread of Covid-19 is not only a serious concern for public health but has also severely affected the global economy. India is not an exception. The banking sector must plan innovatively in a wide range of scenarios focusing upon Covid-19 specific requirements. It becomes essential to examine the impact of Covid-19 on the performance of the Indian banking sector and take focused initiatives at both the tactical and the strategic levels. This paper offers the Covid-19 Impact on Banking Ontology (Covid19-IBO) that provides semantic information about the impact of Covid-19 on the banking sector of India. The developed ontology has been verified and validated and has been made available on the Linked Open Data cloud. It can be utilized to annotate the related data to provide meaningful insights. The Covid-19 ontologies already available have some overlapping information that causes redundancy. Unified integration of these ontologies is required to operate upon them unambiguously. It becomes reasonable to develop a matching approach to link all these ontologies semantically. We, therefore, also provide a schema matching approach with reasonable results to map the Covid-19 ontologies. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of efficiently finding an optimal Bayesian network structure for maximizing the posterior probability. In particular, we focus on the B& B strategy to save the computational effort associated with finding the largest score. To make the search more efficient, we need a tighter upper bound so that the current score can exceed it more easily. We find two upper bounds and prove that they are tighter than the existing one (Campos and Ji, J Mach Learn Res 12(3):663-689, 2011). Finally, we demonstrate that the proposed two bounds render the search to be much more efficient using the Alarm and Insurance data sets. For example, the search is twice to three times faster for $$n=100$$ and almost twice faster for $$n=500$$ . We also experimentally verify that the overhead due to replacing the existing pruning rule by the proposed one is negligible. [ABSTRACT FROM AUTHOR]

Social networking is prevalent everywhere. Technology has made the world seem smaller by bringing individuals together, reconnecting people, and providing other opportunities for connection. However, the risk of data fraud has also increased. The early detection of fraud needs a deep analysis of tweets, which has been done using Neo4J. The Twitter API receives requests from thousands of developers every single day. Limits are imposed on the number of requests that may be made to assist in managing the overwhelming amount of these requests. The most typical request limit period is 15 min. If an endpoint has a rate limit of 1000 requests/15 min, then 1000 requests are permitted over 15 min. Every time a real-time tweet analysis error is returned to the application when these limits are exceeded, online updation of the Neo4J database is difficult. In the proposed method, first and foremost, a new time-based cache (TmCache) is introduced between the database and Twitter API, which removes the complexities of Twitter API and reduces the analyzing time by 85.36%. Furthermore, a machine learning-based approach Multinomial Naive Bayes classifier is used to classify the spam and non-spam tweets with the lowest training time and 96.6% accuracy. The proposed model is helpful in the early detection of spam accurately and can be used in reducing the risk of propagation of agendas which are a major threat to society at large in the current era of social networking. [ABSTRACT FROM AUTHOR]

In today's digital world, information is growing along with the expansion of Internet usage worldwide. As a consequence, bulk of data is generated constantly which is known to be "Big Data". One of the most evolving technologies in twenty-first century is Big Data analytics, it is promising field for extracting knowledge from very large datasets and enhancing benefits while lowering costs. Due to the enormous success of big data analytics, the healthcare sector is increasingly shifting toward adopting these approaches to diagnose diseases. Due to the recent boom in medical big data and the development of computational methods, researchers and practitioners have gained the ability to mine and visualize medical big data on a larger scale. Thus, with the aid of integration of big data analytics in healthcare sectors, precise medical data analysis is now feasible with early sickness detection, health status monitoring, patient treatment, and community services is now achievable. With all these improvements, a deadly disease COVID is considered in this comprehensive review with the intention of offering remedies utilizing big data analytics. The use of big data applications is vital to managing pandemic conditions, such as predicting outbreaks of COVID-19 and identifying cases and patterns of spread of COVID-19. Research is still being done on leveraging big data analytics to forecast COVID-19. But precise and early identification of COVID disease is still lacking due to the volume of medical records like dissimilar medical imaging modalities. Meanwhile, Digital imaging has now become essential to COVID diagnosis, but the main challenge is the storage of massive volumes of data. Taking these limitations into account, a comprehensive analysis is presented in the systematic literature review (SLR) to provide a deeper understanding of big data in the field of COVID-19. [ABSTRACT FROM AUTHOR]

Probabilistic record linkage is a well established topic in the literature. Fellegi-Sunter probabilistic record linkage and its enhanced versions are commonly used methods, which calculate match and non-match weights for each pair of records. Bayesian network classifiers-naive Bayes classifier and TAN have also been successfully used here. Recently, an extended version of TAN (called ETAN) has been developed and proved superior in classification accuracy to conventional TAN. However, no previous work has applied ETAN to record linkage and investigated the benefits of using naturally existing hierarchical feature level information and parsed fields of the datasets. In this work, we extend the naive Bayes classifier with such hierarchical feature level information. Finally we illustrate the benefits of our method over previously proposed methods on four datasets in terms of the linkage performance ( $$F_1$$ score). We also show the results can be further improved by evaluating the benefit provided by additionally parsing the fields of these datasets. [ABSTRACT FROM AUTHOR]

Accuracy and interpretability are two perspectives that are difficult to balance; this is referred to as the accuracy-interpretability dilemma. If credit models gain interpretability, they lose accuracy, and vice versa. Researchers continue to develop an array of very complicated predictive models; however, the finance industry needs interpretable models that can be used in actual practice. Especially, advanced sequential ensembles are seldom considered in credit scoring. Therefore, it is worthwhile to explore new rule extraction methods capable of building sequential ensemble classifiers that are effective for credit scoring. To enhance the accuracy and interpretability of extracted rules, we extend continuous recursive-rule extraction (continuous Re-RX) to a high accuracy-priority rule extraction method referred to as continuous Re-RX with J48graft. Continuous Re-RX with J48graft uses a recursive approach called subdivision. This approach consists of a backpropagation neural network, pruning, and a J48graft decision tree for mixed datasets (those containing discrete and continuous attributes) to construct a high accuracy-priority rule extraction method. Compared with previous rule extraction methods for Australian- and German-based datasets, continuous Re-RX with J48graft achieved the highest accuracies, 88.4 and 79.0%, respectively, using tenfold cross validation (CV) and the Friedman and Bonferroni-Dunn tests, and 87.82 and 78.4%, respectively, using 10 runs of tenfold CV, with the best Friedman score. We also demonstrate how continuous Re-RX with J48graft overcomes the accuracy-interpretability dilemma based on its performance. We believe that continuous Re-RX with J48graft can help overcome the accuracy-interpretability dilemma for transparency of Big Data in financial situations and for industrial applications. [ABSTRACT FROM AUTHOR]