Data Mining Practices Research Articles

Data Mining Best Practices and Efficiency in the Large-Scale Data Mining Using Artificial Intelligence and Generative AI

In the awakening era of the technological landscape, the emergence of key technologies such as generative AI and artificial intelligence (AI) has been instrumental in exercising an effective data mining process. The proliferated demand for best data mining practices continues to express how critical it is for growth in many sectors across different landscapes of human engagements and businesses.

Improvement of data mining technology in the context of big data

As big data continues to permeate every facet of modern life, a growing number of academics are doing studies on the topic. If the potential of data mining is fully explored and researched, there is no doubt that the sector will grow economically. In other words, it is believed that the data information management work in these industries will go more smoothly if the data mining technology and its algorithm can be better applied to various industries and fields related to big data at this stage, handle these data well, and get the desired information. In order to accomplish the objective of increasing the sector's worth. Utilizing a literature review method, this research examines how different businesses have improved their data mining practices, demonstrating how this practice has advanced in the era of big data.

An interval convexity-based framework for multilevel clustering with applications to single-linkage clustering

Hierarchies of sets, and most multilevel clustering models, have been characterized as convexities induced by interval functions satisfying specific properties, thus giving rise to a unifying framework for characterizing multilevel clustering models. Here, we show that this unifying framework can be relevant to data mining practice. First, we provide a flexible characterization of hierarchies and weak hierarchies as interval convexities. Second, we investigate the Apresjan hierarchy and the Bandelt and Dress weak hierarchy, and characterize them as interval convexities. Third, we propose a method for computing recursively a sequence of path-based dissimilarities which decreases from an arbitrary dissimilarity downto its subdominant ultrametric. We prove that these path-based dissimilarities define two sequences of nested families of interval convexities. One sequence increases from the Apresjan hierarchy to the single-link hierarchy, and the other from a subset of the single-link hierarchy to the Bandelt and Dress weak hierarchy. Applications to the simplification and validation of the single-link hierarchy of an arbitrary dissimilarity are discussed.

ADHE: A Tool to Characterize Higher Education Dropout Phenomenon

The field of academic analytics emerged in higher education institutions (HEI) because of developments in database technologies and the generalization of data-mining practices and business intelligence tools. We have designed and implemented a dashboard called ADHE (Academic Analytical Dashboard in Higher Education) for a Colombian higher education institution. The purpose of ADHE is to help administrators of academic programs in their decision-making process regarding the analysis of the phenomenon of student dropout. We used the pipeline methodology for processing large volumes of data was used, which is based on five phases: data acquisition, integration, cleaning, transformation, and visualization. All phases were carried out in the R programming language using academic information sources from the Faculty of Engineering of the Universidad de Antioquia and the Colombian Institute for the Evaluation of Education. The dashboard ADHE is open for free and can be consulted at: https://fhernanb.shinyapps.io/AppPermanencia/. The main findings were that social stratum, gender, and type of high school are associated with the student dropout phenomenon. Furthermore, in social stratum 1, male students and public high schools tend to have a higher student dropout proportion. Additionally, we conclude that admission to engineering programs requires a balance of qualitative and quantitative abilities. The dashboard ADHE should be used to help students, parents, teachers, and administrators understand student dropout dynamics.

Understanding data-driven decision making approach in Chinese higher education through the lens of Bakers model

Making decisions based on data is evolving in higher education worldwide. Higher education institutions (HEIs) in China are also adopting this trend of data-driven decision making to improve the quality standards in higher education. However, the data-driven decision making (DDDM) approach has recently evolved in China, and the data based decision-making models are still immature. Therefore, this study investigates the DDDM approach of Chinese HEIs through an established theoretical data model lens to highlight some of its immaturities and weaknesses for improvement. The study analyzes the DDDM approach in Chinese higher education by exploring the data-mining practices of a case study university. A leading Chinese university has been taken as a sample case study. The multiple interviews have been taken at both faculty and university levels to explore the DDDM practices. The results suggest that the Chinese university DDDM approach is consistent with the established Bakers model. However, the scope of data mining is limited in some areas when compared with the theoretical model. The university is also behind in using advanced applications and software required for DDDM. The results of this study highlight the DDDM approach in Chinese higher education and some of its immaturity. The results can be used to improve the DDDM approach in Chinese HEIs further to ensure more effective and efficient decision-making.

Feature Selection and Dwarf Mongoose Optimization Enabled Deep Learning for Heart Disease Detection.

Heart disease causes major death across the entire globe. Hence, heart disease prediction is a vital part of medical data analysis. Recently, various data mining and machine learning practices have been utilized to detect heart disease. However, these techniques are inadequate for effectual heart disease prediction due to the deficient test data. In order to progress the efficacy of detection performance, this research introduces the hybrid feature selection method for selecting the best features. Moreover, the missed value from the input data is filled with the quantile normalization and missing data imputation method. In addition, the best features relevant to disease detection are selected through the proposed hybrid Congruence coefficient Kumar-Hassebrook similarity. In addition, heart disease is predicted using SqueezeNet, which is tuned by the dwarf mongoose optimization algorithm (DMOA) that adapts the feeding aspects of dwarf mongoose. Moreover, the experimental result reveals that the DMOA-SqueezeNet method attained a maximum accuracy of 0.925, sensitivity of 0.926, and specificity of 0.918.

A Review on Data Mining Concepts and Tools

Data mining is a powerful new technology with great potential to help companies focus on the most important information in their data warehouses. It is well-known that data mining is a significant sub-field of knowledge management and that it is one of the most important information exploration phases in the database cycle. Over the ensuing decades, the practice of data mining will increase in market and learning companies. An introduction to the fundamentals of data mining is provided in this paper.

Data Mining Applications in Banking Sector While Preserving Customer Privacy

In real-life data mining applications, organizations cooperate by using each other’s data on the same data mining task for more accurate results, although they may have different security and privacy concerns. Privacy-preserving data mining (PPDM) practices involve rules and techniques that allow parties to collaborate on data mining applications while keeping their data private. The objective of this paper is to present a number of PPDM protocols and show how PPDM can be used in data mining applications in the banking sector. For this purpose, the paper discusses homomorphic cryptosystems and secure multiparty computing. Supported by experimental analysis, the paper demonstrates that data mining tasks such as clustering and Bayesian networks (association rules) that are commonly used in the banking sector can be efficiently and securely performed. This is the first study that combines PPDM protocols with applications for banking data mining. Doi: 10.28991/ESJ-2022-06-06-014 Full Text: PDF

Tactical Disinformation: Interrogation of Data Politics through Noise in Interactive Media Installations

Abstract This article examines the potential for contemporary interactive media art to interrogate and subvert the data mining practices that encode and exploit user data in the big data economy. Specifically the author argues that this can be accomplished through the exacerbation of data mining operations and the provision of tactical disinformation—useless, fake, and alternative information, an amplification of the “noise” (the unsanctioned elements) that highlights the nonneutrality fallibility, and inadequacy of data. The author explores this through a Brechtian methodology and an examination of three of his media installations to show how they estrange us from normalized information systems to highlight their limits.

KNOWLEDGE DISCOVERY AND DATA MINING OF STRUCTURED AND UNSTRUCTURED BUSINESS DATA: PROBLEMS AND PROSPECTS OF IMPLEMENTATION AND ADAPTATION IN CRISIS CONDITIONS

In modern conditions of the development of the global economy and in connection with the emergence of new branches of economic activity in the field of IT, the phenomenon of Structured and Unstructured Big Data - the use of Data Science for advanced in-depth analysis of data and knowledge in all possible modes - leads to competitive advantages for corporations and institutions, both at the regional and interstate levels, which is especially relevant in the context of the current macroeconomic and military crisis [1].The following topical issues are systematically investigated in the article: current status and prospects for further development of Knowledge Discovery in Data Base (KDD), problems and critical issues of theory and practice of Data Mining, the specifics of effective use of Knowledge Discovery in DB (Data Base) in the current crisis in Ukraine.The above trends and features of the KDD market should be taken into account in further theoretical research and practical implementation or reengineering of KDD systems in Ukraine. The obtained results are relevant and applicable not only for local companies and organizations, but also for international applications in the context of global, regional macroeconomic and current national crisis phenomena.

IoT data visualization for business intelligence in corporate finance

Business intelligence (BI) incorporates business research, data mining, data visualization, data tools,infrastructure, and best practices to help businesses make more data-driven choices.Business intelligence's challenging characteristics include data breaches, difficulty in analyzing different data sources, and poor data quality is consideredessential factors. In this paper, IoT-based Efficient Data Visualization Framework (IoT- EDVF) has been proposed to strengthen leaks' risk, analyze multiple data sources, and data quality management for business intelligence in corporate finance.Corporate analytics management is introduced to enhance the data analysis system's risk, and the complexity of different sources can allow accessing Business Intelligence. Financial risk analysis is implemented to improve data quality management initiative helps use main metrics of success, which are essential to the individual needs and objectives. The statistical outcomes of the simulation analysis show the increasedperformance with a lower delay response of 5ms and improved revenue analysis with the improvement of 29.42% over existing models proving the proposed framework's reliability.

Privacy Tradeoffs: Who Should Make Them, and How?

Privacy debates are contentious in part because we have not reached a broadly recognized cultural consensus about whether interests in privacy are like most other interests that can be traded off in utilitarian, cost-benefit terms, or if instead privacy is different—fundamental to conceptions of dignity and personal liberty. Thus, at the heart of privacy debates is an unresolved question: is privacy just another interest that can and should be bartered, mined, and used in the economy, or is it different? This question identifies and isolates a wedge between those who hold essentially utilitarian views of ethics (and who would see many data practices as acceptable) and those who hold views of natural and fundamental rights (for whom common data mining practices are either never acceptable or, at the very least, never acceptable without significant participation and consent of the subject). This essay provides an intervention of a purely descriptive sort. First, I lay out several candidates for ethical guidelines that might legitimately undergird privacy law and policy. Only one of the ethical models (the natural right to sanctuary) can track the full scope and implications of fundamental rights-based privacy laws like the GDPR. Second, the project contributes to the field of descriptive ethics by using a vignette experiment to discover which of the various ethical models people actually do seem to hold and abide by. The vignette study uses a factorial design to help isolate the roles of various factors that may contribute to the respondents’ gauge of what an ethical firm should or should not do in the context of personal data use as well as two other non-privacy-related contexts. The results can shed light on whether privacy-related ethics are different and distinct from business ethics more generally. They also illuminate which version(s) of good and “bad” share broad support and deserve to be reflected in privacy law or business practice. The results of the vignette experiment show that on balance, Americans subscribe to some form of utilitarianism, although a substantial minority subscribe to a natural right to sanctuary approach. Thus, consent and prohibitions of data practices are appropriate where the likely risks to some groups (most importantly, data subjects, but also firms and third parties) outweigh the benefits.

Reversal, Disconnect, and Proposition

This paper examines the potential countertactics of contemporary interactive media art to interrogate the data-mining practices that encode the everyday and exploit user data in the big data economy. It argues that noise is the "other" of information, a way to counter the operation of turning the world into data commodities. Through a Brechtian methodology informed by philosophy and critical theories of media and technology, the paper suggests that amplifying the "noise" of the digital media assemblages deviates from their everyday normative functions and estranges our relationship to them, inviting critical ways of understanding, relating to, and engaging these ubiquitous systems. All three noted artworks destabilize the protocols of data-mining to examine data politics. Specifically, the paper looks at three different tactics that amplify the "noise" of digital culture in different ways: reversing roles, disconnecting, and proposing viable alternatives.

Emergency Infrastructure and Locational Extraction: Problematizing Computer Assisted Dispatch Systems as Public Good

This article analyzes the increasing articulation between third-party software and emergency infrastructures through a focus on the computer-assisted dispatch system RapidDeploy, which purports to help 9-1-1 responders more accurately and efficiently respond to emergency situations. We build from research that focuses on overlaps between surveillance and emergency response to demonstrate how large-scale data mining practices—in particular, location extraction—are repurposed as beneficial, if not life-saving, measures. In focusing on the capacity to extract and analyze location in the name of public good, we discursively analyze how RapidDeploy’s official company blog constructs the benevolence of data collection. We then demonstrate how these viewpoints are reproduced in the news reports in one city—Charleston, South Carolina—where RapidDeploy has formally partnered with emergency response services. This analysis is used to demonstrate how words like “data” and “cloud” continue to be used as vague buzzwords for companies situated at the intersections of surveillance and civic function, and to argue for greater attention to how the trend towards platformization continues to blur the relationships between surveillance and emergency response systems. This case study examines the public face of this company and, as such, analyzes the language used to gain public assent for the software and its function.

Secure and unifold mining model for pattern discovery from streaming data

The intimidating challenge is practice of data mining (DM) over the streams of data because of its continuous data streaming. On the data streams, the practices of mining should be performed on cluster of streamed records in specified interval of time. The representation of window is the buffered records set which might be dynamic or static in the size. When compared with other practices of mining, the 'frequent pattern mining' on the streams of data are crucial. This occurs because, for predicting the pattern frequency, many of the existing methods repeatedly scan entire buffered transactions. This denotes the intricacy of procedure and overhead of memory. This paper proposes novel DM algorithms in particular for identifying the frequent patterns from indefinite data streams which scans every window once, therefore windows buffered records is pruned that evades computational and memory overhead. 'Unifold mining model for pattern discovery from streaming data' is the contribution of this paper. The outperformance of UMM when compared with other contemporary models is represented by crucial assessment of algorithm and optimisation schemes.

Secure and unifold mining model for pattern discovery from streaming data

The intimidating challenge is practice of data mining (DM) over the streams of data because of its continuous data streaming. On the data streams, the practices of mining should be performed on cluster of streamed records in specified interval of time. The representation of window is the buffered records set which might be dynamic or static in the size. When compared with other practices of mining, the 'frequent pattern mining' on the streams of data are crucial. This occurs because, for predicting the pattern frequency, many of the existing methods repeatedly scan entire buffered transactions. This denotes the intricacy of procedure and overhead of memory. This paper proposes novel DM algorithms in particular for identifying the frequent patterns from indefinite data streams which scans every window once, therefore windows buffered records is pruned that evades computational and memory overhead. 'Unifold mining model for pattern discovery from streaming data' is the contribution of this paper. The outperformance of UMM when compared with other contemporary models is represented by crucial assessment of algorithm and optimisation schemes.

Systematic Survey of Data Mining Technique

Information mining is the disclosure of plans and models in huge and complex informational indexes. There are two sorts of information mining: Model construction and model area. Model design in data mining is actually similar to verifiable illustrating, but new issues arise by virtue of the huge sizes of the data records and the way that Data mining is habitually discretionary data assessment. Model disclosure searches for quirks or minimal close by plans in data, with the enormous mass of the data being insignificant. Without a doubt, one viewpoint on various large-scale data mining practices is that they essentially contain isolating and information decrease.

Large Dataset Classification Using Parallel Processing Concept

Much attention has been paid to large data technologies in the past few years mainly due to its capability to impact business analytics and data mining practices, as well as the possibility of influencing an ambit of a highly effective decision-making tools. With the current increase in the number of modern applications (including social media and other web-based and healthcare applications) which generates high data in different forms and volume, the processing of such huge data volume is becoming a challenge with the conventional data processing tools. This has resulted in the emergence of big data analytics which also comes with many challenges. This paper introduced the use of principal components analysis (PCA) for data size reduction, followed by SVM parallelization. The proposed scheme in this study was executed on the Spark platform and the experimental findings revealed the capability of the proposed scheme to reduce the classifiers’ classification time without much influence on the classification accuracy of the classifier.

Sentimental classification analysis of polarity multi-view textual data using data mining techniques

The data and information available in most community environments is complex in nature. Sentimental data resources may possibly consist of textual data collected from multiple information sources with different representations and usually handled by different analytical models. These types of data resource characteristics can form multi-view polarity textual data. However, knowledge creation from this type of sentimental textual data requires considerable analytical efforts and capabilities. In particular, data mining practices can provide exceptional results in handling textual data formats. Besides, in the case of the textual data exists as multi-view or unstructured data formats, the hybrid and integrated analysis efforts of text data mining algorithms are vital to get helpful results. The objective of this research is to enhance the knowledge discovery from sentimental multi-view textual data which can be considered as unstructured data format to classify the polarity information documents in the form of two different categories or types of useful information. A proposed framework with integrated data mining algorithms has been discussed in this paper, which is achieved through the application of X-means algorithm for clustering and HotSpot algorithm of association rules. The analysis results have shown improved accuracies of classifying the sentimental multi-view textual data into two categories through the application of the proposed framework on online polarity user-reviews dataset upon a given topics.

Instructors' understanding of and responses to student privacy in the datafied classroom

AbstractThe prevalence of educational technology has led to the creation of the “datafied classroom,” a sociotechnical environment captured in and treated as analyzable data for educational data mining and learning analytics practices. This “datafied classroom” presents risks, namely to student privacy. There is a need to better understand how faculty perceive student privacy issues and strategize to address them in practice. This poster reports emerging findings from an IMLS‐funded study that surveyed 502 instructors representing colleges and universities in the United States. The survey captured instructor knowledge of, attitudes and values about, and instructional design experiences with student privacy issues.