Data Mining Applications Research Articles

Imperial Study on Convolutional Layer in CNN for Data Mining

Convolutional Neural Networks (CNNs) have become a cornerstone in the field of data mining, particularly for tasks involving large-scale image and pattern recognition. This paper presents an imperial study on the efficacy of convolutional layers within CNN architectures in data mining applications. We analyze the impact of varying convolutional layer configurations, such as kernel size, stride, and depth, on performance metrics including accuracy, computational cost, and feature extraction quality. By conducting a series of experiments across different datasets, we explore how these configurations influence the model's ability to generalize and detect complex patterns in structured and unstructured data. Our findings indicate that optimizing the convolutional layers significantly improves the efficiency of CNNs for data mining tasks, offering insights into best practices for architectural design in such applications. This study provides valuable guidelines for leveraging CNNs in the realm of data mining, pushing the boundaries of automated data analysis and knowledge discovery

Hospitality in Healthcare: Leveraging Employee e-WOM

ABSTRACT Providing hospitality services to patients to enhance their overall healing process has gained importance in the healthcare industry. It is not only just about getting healed but also the experience involved. This entails quality hospitality services in addition to medical services. Given that a happy employee provides better services, the role of job satisfaction of employees providing these services thus becomes important. Drawing from service quality frameworks, this article aims to identify enablers and deterrents of job satisfaction and validate with practitioners pulling out latent aspects that need to be focused on integrating hospitality in healthcare. We explored these aspects using a three-pronged approach. Initially, we identified the key themes influencing job satisfaction by applying data mining techniques to 13,000 employee reviews from companies recognized as Great Place to Work, sourced from Glassdoor. 1 Second, we validated the identified theme from experts through 24 focused group discussions comprising HR practitioners, thereby gaining interesting insights about certain important overlooked aspects. Third, we include literature support for each of the identified themes, stating their importance to the healthcare sector. The study provides a rubric of enablers and deterrents to job satisfaction, which can help the sector enhance employee satisfaction.

Research on Anomaly Detection of Ship Behavior Based on Data Mining

With the increase in the number of ships and the intensifying maritime traffic, maritime accidents such as ship collisions and grounding occur frequently, posing serious threats to people's lives and property. Therefore, timely detection and correction of abnormal ship behaviors are of great significance for improving navigation safety. The Automatic Identification System (AIS) can provide real-time information on ships' positions, speeds, headings, and more, offering a rich data resource for data mining. Through the analysis of these data, it is possible to deeply explore the navigation patterns and characteristics of abnormal behaviors of ships. This paper applies data mining technology to the detection of abnormal ship behaviors, which not only improves the accuracy and efficiency of detection but also provides maritime regulatory authorities with richer decision-making support.

Revolutionizing Inventory Management: The Role of Data Mining in Industry 4.0

In the era of Industry 4.0, with the rapid development and application of technologies such as the Internet of Things (IoT), Big Data and Artificial Intelligence (AI), the manufacturing industry is undergoing unprecedented changes. In this context, data mining technology has become integral to inventory management practices in all industries. This study examines the profound impact of data mining on inventory management efficiency. By leveraging advanced analytics and machine learning algorithms, data mining enables organizations to accurately forecast demand, optimize inventory levels, and improve supply chain transparency. Through real-world case studies and comprehensive analysis, this research highlights how data mining techniques, such as central object-based clustering algorithms, can be successfully applied to optimize material classification, warehouse space utilization and operational efficiency. In addition, this study explores the broader applications of data mining beyond inventory management, including marketing, financial risk management, healthcare, transportation, social media, and cybersecurity. Overall, this study provides valuable insights into how data mining can reshape inventory management practices and drive business growth in the digital age.

Fully Dynamic Submodular Maximization over Matroids

Maximizing monotone submodular functions under a matroid constraint is a classic algorithmic problem with multiple applications in data mining and machine learning. We study this significant problem in the fully dynamic setting, where elements can be both inserted and deleted in real-time. Our main result is a randomized algorithm that maintains an efficient data structure with an \({\tilde{O}({k^{2}}{\varepsilon})}\) amortized update time (in the number of insertions and deletions) and yields a \({(4+O(\varepsilon))}\) -approximate solution with respect to the dynamic optimum, where \(k\) is the rank of the matroid.

Design of Variation Tolerant Near Threshold Processor Using Artificial Ecosystem Optimizer with Hybrid Deep Learning

Recently, several applications of data mining and pattern recognition require statistical signal processing (SP) to be used and machine learning (ML) techniques for processing massive data volumes in energy-constrained contexts. It is developing interest in executing difficult ML techniques like convolutional neural network (CNN) on lesser power embedding environments to allow on-device learning and inference. Several of these platforms are that utilized as lower power sensor nodes with lower to medium throughput conditions. Near threshold processor (NT) proposals are appropriate for these applications where as affected by a vital enhancement in variants. This research offers an Artificial Ecosystem Optimizer with Hybrid Deep Learning for Variation-Tolerant Near-Threshold Processor (AEOHDL-VTNT). The inference of embedded systems at the network edge serves as the foundation for the AEOHDL-VTNT approach that is being discussed. In the described AEOHDL-VTNT approach involves two primary processes: HDL-based VTNT design and hyper parameter tweaking. Initially, the HDL model is used to develop the VTNT. Next, in the second step, the AEO method is used for hyper parameter tweaking of the HDL model, which improves the HDL's overall performance. A number of simulations were carried out to show how the AEOHDL-VTNT approach improved performance. The simulation results showed that the AEOHDL-VTNT approach outperformed other models.

Menyusun Display Barang dengan Data Mining Assosiation Rule Menggunakan Rapid Miner Algoritma FP-GROWTH

A cosmetic cheerful shop located in the Marelan area of Medan city is a shop that sells women's cosmetic equipment. The owner of a cheerful cosmetics shop is very disciplined in recording sales and purchases of goods. Every transaction is always recorded and stored in the Microsoft Excel application. Meanwhile, the Ceria Cosmetics store has been running since 2015 and has recorded lots of transactions, both sales and purchases. From 2015 until now the data has been increasing, and so far this data has only been a record and stored in a folder on the store's computer. From the piles of transaction data at the cheerful cosmetics store, the authors took the idea of research to utilize this pile of data. One of the applications of information technology that can be used to utilize the pile of data is the data mining technique. By using data mining, a pile of data from a cheerful cosmetic store can be used to find association patterns. And the application of data mining with association rules can use the FP-Growth algorithm. The FP-Growth algorithm can be used as an alternative to determine the most frequently occurring data set (frequent itemset) in a data set. Later, the combination patterns obtained in this study can be used to arrange the layout of goods (cosmetics) at the Ceria Cosmetics store in Medan Marelan.

ТЕОРИЯ НЕЧЕТКИХ МНОЖЕСТВ В ИНТЕЛЛЕКТУАЛЬНОМ АНАЛИЗЕ ДАННЫХ

The article is devoted to the application of data mining (KDD) and fuzzy sets to work with large amounts of data. The digital revolution has led to a significant increase in the volume and speed of data, which makes traditional analysis methods ineffective. Data mining covers the stages of purification, integration, transformation and extraction of knowledge from data. Fuzzy logic, which extends Boolean logic, allows for uncertainty and approximate reasoning, which makes it especially useful for processing complex and noisy data. The article discusses in detail the applications of fuzzy sets in tasks such as clustering, associative rule detection, data generalization, time series analysis, and web applications. Fuzzy logic helps solve problems related to incomplete or inaccurate data by providing more flexible and understandable solutions. The author also notes that the use of fuzzy sets improves the adaptation of systems to specific data requirements and features, which contributes to more accurate analysis and decision-making in areas such as healthcare, finance and telecommunications. These advantages make fuzzy sets an important tool for improving the efficiency of data mining and improving user interaction.

A risk assessment of a gas pressure reduction station system with confidence for dealing with imprecisions and unknowns

Process systems are sensitive and vital industrial facilities. Disturbances in their performance may cause harm to the environment,humans,or significant economic damage. In risk assessment of chemical process industries, the available data, information, and knowledge are typically rare, limited, and often unrealistic. This issue poses a challenge to conducting a credible quantitative risk assessment and effects the robustness of the results. To address these challenges, this work proposes a methodology based on the Dempster-Shafer theory of evidence as the reasoning framework. It incorporates risk identification, analysis, and mitigation phases to ensure a thorough analysis of risks and the integration of proactive risk reduction strategies. The approach aims to model the worst-case hazard scenario and assess associated risks using various methods such as FMECA, Bow-Tie, Credal Network, and Dempster-Shafer theory. The proposed approach models imprecision and data ambiguity using intervals and associated belief mass. This extension provides a basis for addressing the fundamental problem of prior ignorance about the distribution of the observed data, which is prevalent in data mining applications. A new approach is proposed that utilizes Belief and Plausibility curves, similar to a Cumulative Distribution Function, to propagate uncertainty, enhance criticality discrimination, and determine cumulated belief measures. This approach is applied in analyzing the failure modes identified in FMECA and is further extended through the credal network for comprehensive risk assessment. Results show how to express irrelevant and independent judgments, and how to work out with inferences in credal networks. This issue is often overlooked, but if properly addressed it represents the key to ultimately drawing reliable conclusions and fully utilizing the system's available data. A case study of the City Gate Station system was used to verify the application potential of the proposed approach.

Penerapan Data Mining untuk Menentukan Ketersediaan Stok Barang Berdasarkan Permintaan Konsumen di PT Indonesia Thai Summit Plastech Menggunakan K-Means Clustering

Inventory is one of the important things for companies because most of the company's activities are embedded in inventory, with various kinds of problems that exist internally and externally, between employees and superiors and between the company and customers, one of which is inventory problems. So, by having a company's inventory that is stable and in line with customer needs, it will be able to prevent problems between the company and its customers and maintain the company's survival. From the results of research carried out in checking stock availability at PT Indonesia Thai Summit Plastech, basically it is still manual, there is no system that can provide and make things easier for the company or warehouse department. With the problems mentioned above, a system is needed that is able to make it easier and provide information to the company or warehouse quickly. The results of this research produce an output in the form of stock availability patterns, stock accuracy or inventory by applying data mining using the k means clustering method.

Machine Learning, Data Mining, and IoT Applications in Smart and Sustainable Networks

The smart and sustainable networks require highly connected systems that can improve their operational performance, reduce environmental impact, and increase functional efficiency [...]

An Overview of Characteristics and Application of Big Data

Big data is huge volume, heterogeneous, appropriated data. Big data applications where data gathering has developed ceaselessly, it is costly to oversee, catch or concentrate and procedure data utilizing existing programming apparatuses. Clustering Technique for unstructured for big data is a primary errand of exploratory data investigation and data mining applications. This phenomenon, commonly referred to as Big Data, presents both opportunities and challenges across various domains. This paper provides a comprehensive overview of the characteristics and applications of Big Data. it delves into the characteristics of Big Data, including its volume, velocity, variety, veracity, and value. Understanding these characteristics is crucial for harnessing the full potential of Big Data analytics. the paper explores the diverse applications of Big Data across industries such as healthcare, finance, retail, manufacturing, and transportation.

Seismic trend analysis: a data mining approach for pattern prediction

<span lang="EN-US">In the global context, seismic movements represent a constant for the population due to geophysical variability and other factors that make them possible, carrying with them the risk of losing innocent lives. The main purpose of our research is to apply data mining techniques to prevent seismic events of any magnitude to anticipate and mitigate future events. In the development of the research, we applied knowledge discovery database methodology. The clustering analysis results revealed the following: cluster 0 encompassed 45 items, with average magnitude of 0.230, representing 15.5% of the total events. Cluster 1 comprised 56 items with average magnitude of 0.156, equivalent to 19.2% of the total. Cluster 2, the largest, consisted of 94 items with average magnitude of 0.156, constituting 32.3% of the total seismic events. Cluster 3 was composed of 54 items, with average magnitude of 0.155, representing 18.3% of the total. Lastly, cluster 4 included 42 items, with average magnitude of 0.155, representing 14.5% of the total. In conclusion, cluster 3 emerged as the most significant, with 94 events and average magnitude of 0.141, equivalent to 32.3% of the total seismic events. This discovery underscores the need to utilize data mining techniques for earthquake prediction, enabling proactive measures against potential events, which are frequent in various geographic areas.</span>

A Review on Data Mining Issues, Solution & Techniques

Data mining, the process of discovering patterns and knowledge from large datasets, has become a cornerstone of decision-making across various domains including business, healthcare, and finance. This paper reviews the current landscape of data mining applications, exploring the diverse techniques employed and the challenges faced. Key issues include data quality, privacy concerns, and the scalability of mining algorithms in the face of increasingly large and complex datasets. We discuss solutions to these challenges, such as advancements in data cleaning methodologies, privacy-preserving techniques, and the development of scalable algorithms. Additionally, we provide a comprehensive overview of current data mining techniques, including clustering, classification, and association rule mining, and evaluate their effectiveness in addressing real-world problems. The paper aims to provide understanding of both the potential and limitations of data mining, offering insights into future research directions and technological advancements needed to overcome existing hurdles.

Research on Industry 4.0 smart grid monitoring and energy management based on data mining and Internet of Things technology

With the increasing global focus on sustainability and energy efficiency, the concept of Industry 4.0 is becoming an important driver of smart grid technology advancement. Traditional power grids are faced with the challenges of high energy consumption and low management efficiency, and it is urgent to realize intelligent transformation through emerging technologies. This study aims to explore the application of data mining and Internet of Things technology in smart grid monitoring and energy management, with a particular focus on thermal energy management, in order to improve energy utilization efficiency and achieve dynamic regulation. A smart grid monitoring platform integrating data mining and Internet of Things technology is constructed. The data of power and heat energy are collected by sensors, and the data mining algorithm is used to analyze the energy consumption behavior of users and the running state of equipment. A prediction model is also introduced to realize dynamic prediction and optimal scheduling of heat demand. The experimental results show that the thermal energy management system based on the platform can significantly improve the energy utilization rate, data mining technology effectively identifies the peak energy use and potential energy saving opportunities, and the Internet of Things technology realizes real-time monitoring and feedback, enhancing the flexibility and response ability of the system. Through accurate energy monitoring and effective management strategies, it can not only realize the optimal allocation of resources, but also promote the development of enterprises in the direction of low-carbon environmental protection.

Random walks with variable restarts for negative-example-informed label propagation

Label propagation is frequently encountered in machine learning and data mining applications on graphs, either as a standalone problem or as part of node classification. Many label propagation algorithms utilize random walks (or network propagation), which provide limited ability to take into account negatively-labeled nodes (i.e., nodes that are known to be not associated with the label of interest). Specialized algorithms to incorporate negatively-labeled nodes generally focus on learning or readjusting the edge weights to drive walks away from negatively-labeled nodes and toward positively-labeled nodes. This approach has several disadvantages, as it increases the number of parameters to be learned, and does not necessarily drive the walk away from regions of the network that are rich in negatively-labeled nodes. We reformulate random walk with restarts and network propagation to enable “variable restarts", that is the increased likelihood of restarting at a positively-labeled node when a negatively-labeled node is encountered. Based on this reformulation, we develop CusTaRd, an algorithm that effectively combines variable restart probabilities and edge re-weighting to avoid negatively-labeled nodes. To assess the performance of CusTaRd, we perform comprehensive experiments on network datasets commonly used in benchmarking label propagation and node classification algorithms. Our results show that CusTaRd consistently outperforms competing algorithms that learn edge weights or restart profiles, and that negatives close to positive examples are generally more informative than more distant negatives.

Partial random under/oversampling for multilabel problems

Many current data mining applications address problems with instances that belong to more than one class. The term multilabel classification has been introduced as a way of describing this task. Advantageously using the correlation among the labels can provide better performance than methods that manage each label separately. One of the major challenges in multilabel datasets is the class-imbalance problem. In most cases, several or many of the labels are sparsely populated, producing heavily imbalanced datasets. Standard methods used for single-label class-imbalanced datasets are not easily applicable due to the lack of a proper concept of minority instance in the multilabel case. In this paper, we propose a new approach based on partial undersampling and/or oversampling of the instances that is more suitable for the multilabel case. The method modifies the concept of undersampling and oversampling to implement partial undersampling/oversampling of instances. In a large set of 55 real-world multilabel problems, our approach improves the results of current methods for dealing with class-imbalanced datasets in multilabel problems.

Research and progress on the mechanism of lower urinary tract neuromodulation: a literature review.

The storage and periodic voiding of urine in the lower urinary tract are regulated by a complex neural control system that includes the brain, spinal cord, and peripheral autonomic ganglia. Investigating the neuromodulation mechanisms of the lower urinary tract helps to deepen our understanding of urine storage and voiding processes, reveal the mechanisms underlying lower urinary tract dysfunction, and provide new strategies and insights for the treatment and management of related diseases. However, the current understanding of the neuromodulation mechanisms of the lower urinary tract is still limited, and further research methods are needed to elucidate its mechanisms and potential pathological mechanisms. This article provides an overview of the research progress in the functional study of the lower urinary tract system, as well as the key neural regulatory mechanisms during the micturition process. In addition, the commonly used research methods for studying the regulatory mechanisms of the lower urinary tract and the methods for evaluating lower urinary tract function in rodents are discussed. Finally, the latest advances and prospects of artificial intelligence in the research of neuromodulation mechanisms of the lower urinary tract are discussed. This includes the potential roles of machine learning in the diagnosis of lower urinary tract diseases and intelligent-assisted surgical systems, as well as the application of data mining and pattern recognition techniques in advancing lower urinary tract research. Our aim is to provide researchers with novel strategies and insights for the treatment and management of lower urinary tract dysfunction by conducting in-depth research and gaining a comprehensive understanding of the latest advancements in the neural regulation mechanisms of the lower urinary tract.

SENTIMENT ANALYSIS OF ELECTRIC CAR PRODUCT TRENDS IN INDONESIA USING BM25 AND K-NEAREST NEIGHBOR

The global and Indonesian shift towards electric vehicles (EVs) is driven by efforts to reduce emissions and promote sustainable energy. Social media, especially Twitter, functions as an important measuring tool regarding public sentiment towards electric vehicles in Indonesia, so that it can influence policy making. This research uses the BM25 and K-Nearest Neighbor (KNN) methods to analyze sentiment, which aims to improve EV adoption strategies. Conducted in 2023, this research applies data mining, specifically Knowledge Discovery and Data Mining (KDD), analyzing primary and secondary data descriptively and quantitatively starting with data collection from Twitter, followed by data crawling and initial text processing. Next, labeling, term frequency (TF) and inverse document frequency (IDF) calculations were carried out using the BM25 and KNN methods, with an Evaluation and Validation Diagram that visualized the process. The findings show that negative sentiment dominates at 48% (4800 data), followed by 34% (3400 data) neutral sentiment and 18% (1800 data) positive sentiment. The balanced distribution of sentiment highlights the diverse perceptions of society. BM25 and KNN pre-processing methods effectively reduce overfitting and underfitting, especially in negative and neutral sentiments. Accuracy testing without BM25 resulted in 58.6% to 60.25%, while integrating BM25 with KNN increased accuracy by 12.5% to 71% to 72.75%. Understanding sentiment provides a basis for decision making and policy development, as well as providing insight into public perceptions of electric vehicles in Indonesia. Implications include leveraging positive sentiment for marketing, adjusting strategies, refining pricing, addressing infrastructure and reliability issues, and collaborating with governments to increase adoption of electric vehicles in society.

Research on the application of data mining in the field of healthcare

Research on the application of data mining in the field of healthcare