Knowledge Discovery Techniques Research Articles

Fuzzy multi-neighborhood entropy-based interactive feature selection for unsupervised outlier detection

Unsupervised feature selection is one of the important techniques for unsupervised knowledge discovery, which aims to reduce the dimensionality of conditional feature sets as much as possible to improve the efficiency and accuracy of the algorithm. However, existing methods have the following two challenges: (1) They are mainly applicable to select numerical or nominal features and cannot effectively select heterogeneous features; (2) The relevance and redundancy are primarily considered to construct feature evaluation indexes, ignoring the interaction information of heterogeneous features. To solve the challenges mentioned above, this paper proposes an unsupervised heterogeneous feature selection method based on fuzzy multi-neighborhood entropy, which also considers the multi-correlation of features to select heterogeneous features. First, the fuzzy multi-neighborhood granule is constructed by considering the distribution characteristics of the data. Then, the concept of fuzzy entropy is introduced to define the fuzzy multi-neighborhood entropy and its associated uncertainty measures, and the relationship between them is discussed. Next, the relevance, redundancy, and interactivity among attributes are defined, and the idea of maximum relevance-minimum redundancy-maximum interactivity is used to construct the evaluation indexes of heterogeneous features. Finally, experiments are conducted on several publicly unbalanced datasets, and the results are in comparison with existing algorithms. The experimental results show that the proposed algorithm can select fewer heterogeneous features to improve the efficiency of outlier detection tasks. The code is publicly available online at https://github.com/BELLoney/MNIFS.

Anti-Hypertension Drugs Classes of Prevention and Side Effects Diseases Efficiency Analysis in Association Rule Mining Techniques

Purpose: The Healthcare department, pharmaceutical department, Hospital and Clinical sector find out and explore the diseases of status, where it has been spreaded the communicable and non-communicable diseases among the society. The healthcare department conducting different awareness of programming about the different diseases how is affected people and how is prevented the diseases in society. Accordingly, all the healthcare awareness of information passed through the different media channel, even though high and low blood pressure is pressing public health challenges and it is recognized as the biggest contributor to the global burden of diseases. Presently people health is silently affected by blood pressure low and high level and they unable to recognize that something is amiss, high blood pressure is known as the "silent killer” and If blood pressure is excessively elevated, it may have an impact on organ damage or health issues like coronary arteries, heart valve dysfunction, diabetes, kidney diseases, heart attack and stroke this all are risk factors of blood pressure abnormal status. Hypertension diseases affected the patients need to the best prevention and feature safety. The Doctors, Pharmacist and Nurse are using Anti - hypertension drugs classes of medicines for patients. Which anti-hypertension drug classes of medicine good efficacy for patients and anti-hypertension drugs classes of medicine prevention diseases, side effects diseases knowledge is important for healthcare professional. Machine learning and Data mining knowledge discovery techniques need to understand how different classes of anti-hypertensive drugs might interact with the patient’s and medications. If the healthcare professional has access to a predictive data mining technique model, they could use this to anticipate how the patient’s condition might change over time and adjust the treatment plan proactively. This kind of analytical data mining knowledge can lead to more effective treatment and better patient outcomes. Design/Methodology/Approach: Developing machine learning concept for different anti-hypertension drugs classes of medicine efficiency analysis in hypertension prevention diseases and side effects diseases and Healthcare professional to take right decision for future adjusts treatment plan to the hypertension affected patients. Findings/Result: Orange data mining analytical tool to identify the anti-hypertension drugs classes of medicines efficacy and Doctors can take right decision to better treatment for the patients. Originality/Value: Data mining association rules of support, confidence and lift correlation analysis system helps to identified about the drug of new knowledge efficiency. Paper Type: Analytical research methods applied for analysis the different types of anti-hypertension drugs classes of association correlation efficacy in data mining machine learning system.

Book Classification and Recommendation for University Libraries using Grey Correlation and Bayesian Probability

In today’s information era, collaborative filtering algorithms are widely used and their distinct knowledge discovery techniques can effectively address numerous issues. However, conventional collaborative filtering algorithms encounter cold-start and data sparsity issues, which restrict their performance and accuracy. The study selected the multi-feature method to improve the traditional collaborative filtering algorithm, and introduced gray correlation calculation and Bayesian probability for user preference analysis. A learning resource recommendation model based on collaborative filtering was developed by comparing the target user’s characteristics with those of other users, calculating their similarity, selecting users with high similarity to the target user and forming a neighbor set. Using Bayesian probability and grey correlation to analyze user preferences in library systems can be well applied in book classification and recommendation problems in university libraries. The computing layer, which includes the collaborative filtering calculation stage and the group recommendation calculation stage, is the model’s main functional component. The smaller the value of mean absolute error, the higher the prediction accuracy of the model. The mean absolute error value of the multi-feature collaborative filtering algorithm was inferior to the traditional collaborative filtering algorithm, indicating that the classification accuracy of the former is higher than that of the latter. When the training set to test set ratio steadily became bigger, the mean absolute error value reached the lowest and smoothest point at 80%. In dataset A, the minimum mean absolute error values of multi-feature collaborative filtering and collaborative filtering were 0.765 and 0.809. Compared with traditional filtering algorithms, the mean absolute error value has decreased by 0.044. In dataset B, the mean absolute error values of multi-feature collaborative filtering and collaborative filtering were 0.796 and 0.836. Compared with traditional filtering algorithms, the mean absolute error value has decreased by 0.040. In dataset C, the minimum mean absolute error values of multi-feature collaborative filtering and collaborative filtering were 0.815 and 0.848. Compared with traditional filtering algorithms, the mean absolute error value has decreased by 0.033. When the accuracy was the highest; the mean absolute error value was the smallest at the grey correlation, which means that the technique improves the reliability of the recommendations compared with other methods. This means that the method has a positive impact on the accuracy of the recommendations compared to other methods. Grey correlation degree can comprehensively consider the interrelationships between multiple factors, handle uncertain and incomplete information, and explore potential user needs and behavior patterns. The implementation of the grey correlation degree has transformed the collaborative filtering algorithm into a group filtering algorithm, thereby enhancing its precision. The research on book classification and recommendation in university libraries, which enhances the group filtering algorithm, can address a range of issues such as improving classification accuracy, augmenting recommendation diversity, enhancing library management efficiency among others. This, in turn, enables more precise book recommendations to users.

Rule Mining Trends from 1987 to 2022: A Bibliometric Analysis and Visualization

Abstract Rule mining has emerged as a crucial technique in data mining and knowledge discovery, enabling the extraction of valuable insights and patterns from vast datasets. This has garnered significant attention from both academic and industrial communities. However, there is a lack of bibliometric and visualization research on rule mining, leading to an unclear delineation of research topics and trends in the field. To fill this gap, this paper provides a comprehensive and up-to-date bibliometric analysis of rule mining, covering 4524 publications published between 1987 and 2022. Using various metrics and visualization techniques, we examine the patterns, trends, and evolution of rule mining. The results show a sustained growth in rule mining research, with a significant increase in publication output in recent years, and its rapid expansion into new areas such as explainable artificial intelligence and privacy protection. While the majority of publications come from Asia, the National Natural Science Foundation of China emerges as the top funding agency in the field. We also identify highly productive authors and significant members of co-authorship networks, as well as the most influential publications and citation bursts. The need for international collaboration and the integration of diverse research perspectives is highlighted. Despite the progress in rule mining, several challenges still require further research, including scalability and efficiency, explainability, network security and privacy protection, and personalized and user-centered design. Overall, this paper provides a valuable roadmap for researchers, policymakers, and practitioners interested in rule-mining research.

Additive autoencoder for dimension estimation

Dimension reduction is one of the key data transformation techniques in machine learning and knowledge discovery. It can be realized by using linear and nonlinear transformation techniques. An additive autoencoder for dimension reduction, which is composed of a serially performed bias estimation, linear trend estimation, and nonlinear residual estimation, is proposed and analyzed. Compared to the classical model, adding an explicit linear operator to the overall transformation and considering the nonlinear residual estimation in the original data dimension significantly improves the data reproduction capabilities of the proposed model. The computational experiments confirm that an autoencoder of this form, with only a shallow network to encapsulate the nonlinear behavior, is able to identify an intrinsic dimension of a dataset with low autoencoding error. This observation leads to an investigation in which shallow and deep network structures, and how they are trained, are compared. We conclude that the deeper network structures obtain lower autoencoding errors during the identification of the intrinsic dimension. However, the detected dimension does not change compared to a shallow network. As far as we know, this is the first experimental result concluding no benefit from a deep architecture compared to its shallow counterpart.

Analysis of the Character of the Insured by Policy Holders of Private Insurance Companies in Indonesia with the Rough Set Method

Insurance provides services in the form of services in facilitating a disaster that may occur based on a history of problems that have occurred. One of the private insurance companies in Indonesia has provided an insurance service program to guarantee employee welfare for policy-holding companies. The purpose of this research is to find out the pattern of association rules from the character of the insured by the policyholder so that it is expected to be able to make valuable information as input or consideration for the policymakers of private insurance companies in Indonesia. This study uses the rough set method as one of the efficient data analysis methods or techniques in database mining or knowledge discovery in relational databases. Rough sets provide algorithms to quickly and easily find hidden patterns in data. The results of the pattern of association rules for the character of the insured by the policyholder have formed as many as 14 rules. The certainty value is intended as a proportionate evaluation amount in order to find out the record of insurance claims that can be chosen by the insured. At the same time, the coverage value is intended as an evaluation amount to produce a decision for the insured to submit a claim record. The probable percentage of all events that are most recommended is seen from the highest coverage value related to record indicators, namely for administration by 45.9%, inspection by 44.1%, others by 40.8%, operation by 48.8%, financing by 46.3%, maintenance by 42%, and compensation of 65.9%.

Machine Learning Techniques for Prognosis Estimation and Knowledge Discovery From Lab Test Results With Application to the COVID-19 Emergency

Machine Learning Techniques for Prognosis Estimation and Knowledge Discovery From Lab Test Results With Application to the COVID-19 Emergency

Some Insights Concerning the Halal Tourism Research. A Bibliometric Analysis

Halal tourism has become one of the topic issues that have a level of popularity among researchers in the scientific articles published in journals indexed in the Scopus database. This research study uses a bibliometric design analysis to map some insights related to the halal tourism by evaluating articles that have been published in journals indexed in Scopus database. Furthermore, data mining (knowledge discovery) techniques are used to analyze the titles and abstracts of 127 Scopus indexed journals that have been saved in the form of a CSV file. Finally, three analyzes are produced in this study: network visualization analysis, overlay visualization, and density visualization. 
 This study's results indicate a significantly increased of articles on halal tourism, especially in the Islamic marketing journals. Also, the main outputs obtained though this bibliometric research related to the halal tourism, reveal which is the most popular authors, which is the most cited articles, which is the country with most publications, which is the research domain most approached. In the same manner is underline which is less approached practices, i.e. , halal tourism from the perspective of Islamic law, etc.

A gradient boosting machine-based framework for electricity energy knowledge discovery

Knowledge discovery in databases (KDD) has an important effect on various fields with the development of information science. Electricity energy forecasting (EEF), a primary application of KDD, aims to explore the inner potential rule of electrical data for the purpose to serve electricity-related organizations or groups. Meanwhile, the advent of the information society attracts more and more scholars to pay attention to EEF. The existing methods for EEF focus on using high-techs to improve the experimental results but fail to construct an applicable electricity energy KDD framework. To complement the research gap, our study aims to propose a gradient boosting machine-based KDD framework for electricity energy prediction and enrich knowledge discovery applications. To be specific, we draw on the traditional knowledge discovery process and techniques to make the framework reliable and extensible. Additionally, we leverage Gradient Boosting Machine (GBM) to improve the efficiency and accuracy of our approach. We also devise three metrics for the evaluation of the proposed framework including R-square (R2), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). Besides, we collect the electricity energy consumption (EEC) as well as meteorological data from 2013 to 2016 in New York state and take the EEC prediction of New York State as an example. Finally, we conduct extensive experiments to verify the superior performance of our framework and the results show that our model achieves outstanding results for the three metrics (around 0.87 for R2, 60.15 for MAE, and 4.79 for MAPE). Compared with real value and the official prediction model, our approach also has a remarkable prediction ability. Therefore, we find that the proposed framework is feasible and reliable for EEF and could provide practical references for other types of energy KDD.

The Use of Data Science in Digital Marketing Techniques: Work Programs, Performance Sequences and Methods.

Over the past decade, there have been tremendous advances in the use of data science to facilitate decision making and extract and thus act upon insights from large data sets in digital business environments. However, despite these advances, there is still a lack of relevant evidence on actions to improve data science management in digital businesses. To fill this gap in the literature. The purpose of this study is to review (i) analytical methods, (ii) usage, and performance metrics based on (iii). Data science used in digital business techniques and strategies. To this end, a comprehensive literature search was carried out on the important scientific contributions made so far in this area of ​​research. The results provide an overview of the most important applications of data science in digital business. Generate insights related to the creation of innovative data mining and knowledge discovery techniques. Important theoretical implications are discussed and a list of topics for further research in this area is provided. This report aims to develop recommendations for enhancing digital business strategies for enterprises, marketing, and non-technical researchers, and identify directions for further research on innovative data mining and discovery applications in science.

The Use of Data Science in Digital Marketing Techniques: Work Programs, Performance Sequences and Methods.

Over the past decade, there have been tremendous advances in the use of data science to facilitate decision making and extract and thus act upon insights from large data sets in digital business environments. However, despite these advances, there is still a lack of relevant evidence on actions to improve data science management in digital businesses. To fill this gap in the literature. The purpose of this study is to review (i) analytical methods, (ii) usage, and performance metrics based on (iii). Data science used in digital business techniques and strategies. To this end, a comprehensive literature search was carried out on the important scientific contributions made so far in this area of ​​research. The results provide an overview of the most important applications of data science in digital business. Generate insights related to the creation of innovative data mining and knowledge discovery techniques. Important theoretical implications are discussed and a list of topics for further research in this area is provided. This report aims to develop recommendations for enhancing digital business strategies for enterprises, marketing, and non-technical researchers, and identify directions for further research on innovative data mining and discovery applications in science.

A Comparative and Performance Analysis of Similarity Metrics in Recommender System Based on Hadoop Framework

With exponential growth of information available on web, there is a need for a personal assistance to customers for finding the best item out of the large set of items. There is also need for finding the best item having highest popularity. The personal assistance which will assist the user is the Recommender System. It is a software tool that uses knowledge discovery techniques to produce personalized recommendations. The majority of the recommendations are based on the machine learning algorithm and techniques which uses a standard data set for producing predictions. With tremendous growth of customers and items in the recent years, there is a key challenge for the recommendation system to produce quality recommendations based on the similarity of the users. There also lies a great challenge for the recommendation system to produce quality output when data set is huge, reduce latency, group users having similar interest and perform recommendations in seconds for millions of customers and products. Thus there is a need for new recommendation technologies which will produce quality output to users having similar tastes in a best possible manner. In this paper we present a comparative analysis of various similarity metrics used in recommendation system for clustering the users based on hadoop framework. The experimental results show that spearman rank correlation performs the best among the other similarity metrics based on three evaluation metrics.

Efficient Discovery of Partial Periodic Patterns in Large Temporal Databases

Periodic pattern mining is an emerging technique for knowledge discovery. Most previous approaches have aimed to find only those patterns that exhibit full (or perfect) periodic behavior in databases. Consequently, the existing approaches miss interesting patterns that exhibit partial periodic behavior in a database. With this motivation, this paper proposes a novel model for finding partial periodic patterns that may exist in temporal databases. An efficient pattern-growth algorithm, called Partial Periodic Pattern-growth (3P-growth), is also presented, which can effectively find all desired patterns within a database. Substantial experiments on both real-world and synthetic databases showed that our algorithm is not only efficient in terms of memory and runtime, but is also highly scalable. Finally, the effectiveness of our patterns is demonstrated using two case studies. In the first case study, our model was employed to identify the highly polluted areas in Japan. In the second case study, our model was employed to identify the road segments on which people regularly face traffic congestion.

A density estimation approach for detecting and explaining exceptional values in categorical data

In this work we deal with the problem of detecting and explaining anomalous values in categorical datasets. We take the perspective of perceiving an attribute value as anomalous if its frequency is exceptional within the overall distribution of frequencies. As a first main contribution, we provide the notion of frequency occurrence. This measure can be thought of as a form of Kernel Density Estimation applied to the domain of frequency values. As a second contribution, we define an outlierness measure for categorical values that leverages the cumulated frequency distribution of the frequency occurrence distribution. This measure is able to identify two kinds of anomalies, called lower outliers and upper outliers, corresponding to exceptionally low or high frequent values. Moreover, we provide interpretable explanations for anomalous data values. We point out that providing interpretable explanations for the knowledge mined is a desirable feature of any knowledge discovery technique, though most of the traditional outlier detection methods do not provide explanations. Considering that when dealing with explanations the user could be overwhelmed by a huge amount of redundant information, as a third main contribution, we define a mechanism that allows us to single out outstanding explanations. The proposed technique is knowledge-centric, since we focus on explanation-property pairs and anomalous objects are a by-product of the mined knowledge. This clearly differentiates the proposed approach from traditional outlier detection approaches which instead are object-centric. The experiments highlight that the method is scalable and also able to identify anomalies of a different nature from those detected by traditional techniques.

Efficient artificial intelligence techniques for inverse design and knowledge discovery in metamaterials

A new artificial-intelligence (AI) approach based on dimensionality reduction techniques for the design and knowledge discovery in metamaterial structures will be presented. It is shown that reducing the dimensionality of the response and design spaces by using a pseudoencoder can reduce the computation time by 2–3 orders of magnitude. In addition, using pruning techniques can simplify the trained algorithm to further reduce the computation time while providing simpler paths for relating input-output relationships for understanding the role of design parameters (i.e., knowledge discovery). It is also shown that this approach can enable an evolutionary design method in which the initial design can be evolved intelligently into an alternative with favorable specifications. The application of these AI approaches for both optical and acoustic metamaterials will be presented.

Generating Terrain Data for Geomorphological Analysis by Integrating Topographical Features and Conditional Generative Adversarial Networks

With the development of artificial intelligence techniques for geographical knowledge discovery, simulated terrain generation based on deep-learning algorithms has become one practical way to construct accurate terrain data. However, it is still necessary to discuss whether the simulated topographic data contain the characteristics of specific landforms and can support related geographical studies. Therefore, in this study, a deep learning-based model inspired by previous research is constructed to generate loess landform data. We analyzed the influence of inputting different topographic features on terrain generation and evaluated the similarity between the simulated and reference data. The results show that the deep learning-based model can generate simulated topographic data that include similar elevation and slope probability distributions to the reference data of the loess landform. In addition, the generated results may have inaccurate terrain details, which can be regarded as noise in some cases. This indicates that the selection of input features should be carefully considered. Finally, the simulated data can subsequently support landform and terrain research, especially with intelligence algorithms that require large sets of topographic data.

MR-DBIFOA: a parallel Density-based Clustering Algorithm by Using Improve Fruit Fly Optimization

<p>Clustering is an important technique for data analysis and knowledge discovery. In the context of big data, the density-based clustering algorithm faces three challenging problems: unreasonable division of data gridding, poor parameter optimization ability and low efficiency of parallelization. In this study, a density-based clustering algorithm by using improve fruit fly optimization based on MapReduce (MR-DBIFOA) is proposed to tackle these three problems. Firstly, based on KD-Tree, a division strategy (KDG) is proposed to divide the cell of grid adaptively. Secondly, an improve fruit fly optimization algorithm (IFOA) which use the step strategy based on knowledge learn (KLSS) and the clustering criterion function (CFF) is designed. In addition, based on IFOA algorithm, the optimal parameters of local clustering are dynamically selected, which can improve the clustering effect of local clustering. Meanwhile, in order to improve the parallel efficiency, the density-based clustering algorithm using IFOA (MR-QRMEC) are proposed to parallel compute the local clusters of clustering algorithm. Finally, based on QR-Tree and MapReduce, a cluster merging algorithm (MR-QRMEC) is proposed to get the result of clustering algorithm more quickly, which improve the core clusters merging efficiency of density-based clustering algorithm. The experimental results show that the MR-DBIFOA algorithm has better clustering results and performs better parallelization in big data.</p> <p> </p>

A Statistical-Mining Techniques’ Collaboration for Minimizing Dimensionality in Ovarian Cancer Data

A feature is a single measurable criterion to an observation of a process. While knowledge discovery techniques successfully contribute to many fields, however, the extensive required data processing could hinder the performance of these techniques. One of the main issues in processing data is the dimensionality of the data. Therefore, focusing on reducing the data dimensionality through eliminating the insignificant attributes could be considered one of the successful steps for raising the applied techniques’ performance. On the other hand, focusing on the applied field, ovarian cancer patients continuously suffer from the extensive analysis requirements for detecting the disease as well as monitoring the treatment progress. Therefore, identifying the most significant required analysis could be a positive step to reduce the emotional and financial suffering. This research aims to reduce the data dimensionality of the ovarian cancer disease and highlight the most significant analysis using the collaboration of clustering techniques and statistical techniques. The research succeeded to identify twelve significant analysis out of forty-four with a total of fourteen significant attributes for ovarian cancer data.

Smart Competence Management Using Business Analytics with Fuzzy Predicates

Organizations consider human capital as one of their most important assets. Experts in the field have focused on the research and development of human talent management skills. At present, companies are giving high importance to the management of this intangible resource. Management by competencies and skills is basic in the selection and development of the most valuable asset the organization has: its human capital. A conceptual framework of the intelligent management of human capital and some more advanced knowledge discovery techniques are presented in this paper. A methodology for smart detection of core competencies based on fuzzy logic predicates and business analytics is proposed. The proposed methodology allows: (1) the evaluation of the importance of competencies, (2) the identification of competencies achievement level of each employee, (3) the identification of competencies with difficulties, (4) the identification of competencies that have influence on others, and (5) a hierarchization of the competencies to select the most appropriated for the employee recruitment plan. Furthermore, an analysis is proposed using knowledge discovery, which allows one to identify which competences have influence on a specific one. All of the above is useful to build an ideal profile for a position. A case study was carried out in order to show the implementation and interpretation of our proposal.

MapReduce Based Crow Search Adopted Partitional Clustering Algorithms For Handling Large Scale Data

Cluster analysis is the prominent data mining technique in knowledge discovery and it discovers the hidden patterns from the data. The K-Means, K-Modes and K-Prototypes are partition based clustering algorithms and these algorithms select the initial centroids randomly. Because of its random selection of initial centroids, these algorithms provide the local optima in solutions. To solve these issues, the strategy of Crow Search algorithm is employed with these algorithms to obtain the global optimum solution. With the advances in information technology, the size of data increased in a drastic manner from terabytes to petabytes. To make proposed algorithms suitable to handle these voluminous data, the phenomena of parallel implementation of these clustering algorithms with Hadoop Mapreduce framework. The proposed algorithms are experimented with large scale data and the results are compared in terms of cluster evaluation measures and computation time with the number of nodes.