Data Mining Techniques Research Articles

Forecasting and uncertainty analysis of tailings dam system safety based on data mining techniques

Tailings dams, as critical infrastructure, play a vital role in ensuring the safety and reliability of tailings pond systems. Predicting the trend of tailings dam monitoring parameters helps monitor its operational status and support safety management and decision-making. The traditional time series prediction model focuses on point prediction while neglecting the nonlinear feature of data and the uncertainty of prediction results, which is far from meeting the requirements for risk monitoring and safety management of tailings dam systems. To address this, Deep Auto-Encoder is used to extract and optimize features of multi-dimensional time series; Whale Optimization Algorithm, Long Short Term Memory, and Kernel Density Estimation are used to build a monitoring parameters trend interval prediction model. It can effectively predict and analyze the uncertainty of tailings dam monitoring parameters. Time series data of monitoring parameters are collected from a tailings dam in Anhui, China and modeled to verify the effectiveness of the model. The results show that the hybrid prediction model performs the best in monitoring parameter prediction accuracy and trend. The Root Mean Square Error values of the Whale Optimization Algorithm and Long Short Term Memory prediction for reservoir water level, surface displacement, internal displacement, phreatic line, and beach width are 0.0160, 1.0405, 0.0389, 0.0176, and 0.7525, with errors reduced by 0.37 % to 17.92 % compared to other models. The uncertainty analysis confirms the high reliability of the model. The Kernel Density Estimation effectively forecasts the fluctuation range of predicted values, significantly improving the prediction effectiveness of the monitoring parameters.

Innovative approaches in forensic accounting: The role of data analytics

Abstract The integration of data analytics into forensic accounting has revolutionized the detection and prevention of financial fraud. This paper conducts a comprehensive analysis of recent advancements in this field, highlighting the application of machine learning, data mining, and big data techniques in identifying fraudulent activities. By reviewing the latest research and examining case studies, we demonstrate the enhanced accuracy and efficiency these technologies offer over traditional methods. The findings suggest that financial institutions should adopt these advanced tools to mitigate fraud risks and improve overall financial security. The paper also explores future research directions, emphasizing the need for developing hybrid models and real-time detection systems to further enhance fraud detection capabilities.

Association rules forecasting for the foreign exchange market

Several association rule mining algorithms exist, and among them, Apriori is one of the most commonly used methods for extracting frequent item sets from vast databases and generating association rules to gain insights. In this research, we have applied a data mining technique to implement association rules and explore frequent item sets. Our study introduced a model that employs association rules to uncover associations between the foreign exchange market, the gold commodity, and the National Association of Securities Dealers automated quotations (NASDAQ). We suggested a method that used data mining to identify the good points of buying and selling in the foreign exchange market by utilizing technical indicators such as moving average convergence divergence (MACD) and the stochastic indicator to create association rules. The experimental findings indicate that the proposed model successfully generates strong association rules.

Bibliometric study of association rule-market basket analysis

Association rule-market basket analysis (AR-MBA) is a data mining technique for finding distinguished relationship patterns from a collection of items. The application of AR-MBA is also increasingly widespread, starting from retail and hotels to hospitals. So, bibliometrics related to AR-MBA needs to be done to reveal what research opportunities can be later carried out by reviewing and analyzing publications about AR-MBA. 91 bibliographies in 1 decade from 2012-2022 were collected using Harzing's Publish or Perish (PoP). VOSviewer is also employed to map authorship and publication topic trends. This paper is innovative because it identifies trends and future research directions in data mining, specifically in association with AR-MBA. The findings show publication productivity, top authors, types of publications, annual topic trends within a decade, term distribution, most cited and most influential articles, and research gaps that can be opportunities for further research.

Identification of fraudulent financial statements through a multi‐label classification approach

SummaryAlthough the financial audit controls in companies have advanced over the years, the number of corporate fraud instances is growing, thus raising the need for investigating the factors that can be used as early warning signals and developing effective systems for identifying financial fraud. In this paper, financial statements from 133 Greek companies listed in the Athens Stock Exchange over the period 2014 to 2019 are investigated, based on the fraud diamond theory. Financial data and corporate governance variables are used as inputs to data mining techniques to develop models that can identify patterns of irregularities in a company's financial reports. To this end, popular machine learning classification algorithms are employed in a novel multi‐label classification setting that not only identifies fraudulent cases but also considers the nature of the auditors' comments. The results indicate that the proposed multi‐label approach provides enhanced results compared to binary classification algorithms, avoiding inconsistent outputs with respect to the existence of different forms of manipulation of financial statements.

Market Basket Analysis

Abstract: Market Basket analysis is a technique applied by retailers to understand customer’s shopping behaviour from their stores. The result of the effective analysis may improve supplier’s profitability, quality of service and customer satisfaction. The purpose of this project is to make use of anonymized data on customers’ transactional orders to focus on descriptive analysis on the customer purchase patterns, items which are bought together and units that are highly purchased from the store to facilitate reordering and maintaining adequate product stock. Market Basket Analysis is an important aspect of a retail organization's analytical framework for deciding where products should be placed and developing sales promotions for various segments of consumers to increase customer loyalty and, as a result, benefit. Market Basket Analysis is a data mining technique that can be used in various fields, such as marketing and etc. The frequent itemsets are mined from the database using the Apriori algorithm and then the association rules are generated. The project will assist supermarket managers in determining the relationship between the items that their customers purchase.

Deep Learning Empowered Decision Support Systems for Thyroid Cancer Detection and Management

Exploiting the capability of deep learning-based techniques, this research addresses an important and relevant problem on how to cost effectively detect thyroid cancer. In recent decades, there has been a significant increase in the incidence of thyroid cancer, prompting the need to understand its epidemiology. Existing studies have primarily employed qualitative techniques to investigate a single risk factor correlated with the disease development at a time. However, such an approach is inefficient and tends to ignore the interwoven connections among factors, resulting in a considerable disagreement with the identified risk factors among scholars. Additionally, the use of deep learning techniques in conjunction with medical imaging for computer-aided diagnostic (CAD) systems design has shown promises in detecting the disease, while there are research gaps regarding the detection of subtype and their co-existence. More importantly, existing CAD systems have shortcomings in adapting to different sample groups. To address these challenges, this research aims to shed light on the pathogenesis of thyroid cancer, enhance diagnostic performance, and improve generalisation of deep learning-based decision support systems. By harnessing the power of machine learning, specifically data mining and deep learning techniques, we seek to improve our understanding of the underlying mechanisms of the disease and develop innovative, robust, accurate, and efficient diagnostic tools. Extensive experiments indicate superior performance of the proposed methods than existing works. The systems proposed in this study have great impact to the wider society and contribute to the advancement of thyroid cancer research while enhancing clinical practice in related detection and subsequent management.

Comparison of Methods for Applying the Data Mining Clustering Concept to Banten Provincial Government Poverty Data: Systematic Review

An in-depth understanding of the phenomenon of poverty in Banten Province is the key to formulating effective strategies in overcoming this problem. In general, poverty in this area is caused by several interrelated and complex factors. This research provides a comparative analysis of the application of clustering data mining methods in calculating population poverty data in certain regions or provinces, especially Banten. This research aims to assess the consistency and effectiveness of various clustering methods used in previous research. Using a qualitative approach with a literature review, secondary data from relevant research was collected through systematic searching, reading, and note-taking. The data collected is then carried out through a process of data collection, filtering, presentation and drawing conclusions. The results show that most of the literature reviews show similar results regarding the effectiveness of various clustering methods in analyzing poverty data in Banten. This shows that there is a consensus among previous studies regarding the efficacy of clustering data mining techniques in overcoming problems related to poverty in the region. These findings contribute to a deeper understanding of the methodology used in poverty analysis and provide insights for policy makers and researchers to develop more effective strategies in poverty alleviation efforts. The conclusion of this research is that clustering data mining in analyzing poverty data in Banten Province has shown quite high consistency and effectiveness. Various literature studies that have been reviewed show that various clustering methods that have been applied previously provide similar results in identifying poverty patterns, understanding socio-economic structures, and providing valuable insights for developing more targeted policies.

A Framework to Optimize Student Performance using Machine Learning

For scholars, mining data and extracting information from huge databases has emerged as an intriguing field of study. Since a few decades ago, the concept of using data mining techniques to extract information has been around. The dataset was originally intended to be partitioned and the inherent features examined using classification and clustering algorithms. They base their predictions on these characteristics. These forecasts have been made in the area of educational data mining for a variety of reasons, including to predict student success based on personal characteristics and help students find the right professors and courses. These goals have been drawn from the attrition and retention of students. These objectives are the focus of our research on student attrition and retention. Additionally, we have found exciting variables that aid in predicting students' success, suggesting the most qualified instructors, and assisting them in course selection.

Principal Component Analysis of a Real-World Cohort of Descemet Stripping Automated Endothelial Keratoplasty and Descemet Membrane Endothelial Keratoplasty Cases: Demonstration of a Powerful Data-Mining Technique for Identifying Areas of Research.

Principal component analysis (PCA) is a descriptive exploratory statistical technique that is widely used in complex fields for data mining. However, it is rarely used in ophthalmology. We explored its research potential with a large series of eyes that underwent 3 keratoplasty techniques: Descemet membrane endothelial keratoplasty (DMEK), conventional Descemet stripping automated endothelial keratoplasty (ConDSAEK), or ultrathin-DSAEK (UT-DSAEK). All consecutive DMEK/DSAEK cases conducted in 2016 to 2022 that had ≥24 months of follow-up were included. ConDSAEK and UT-DSAEK were defined as preoperative central graft thickness ≥130 and <130 μm, respectively. Seventy-six patient, disease, surgical practice, and temporal outcome variables were subjected to PCA, including preoperative anterior keratometry, the use of sulfur hexafluoride gas (SF6) versus air for primary tamponade, and postoperative best corrected visual acuity and endothelial cell density. Associations of interest that were revealed by PCA were assessed with the Welch t test or Pearson test. A total of 331 eyes were treated with DMEK (n = 165), ConDSAEK (n = 95), or UT-DSAEK (n = 71). PCA showed that ConDSAEK and UT-DSAEK clustered closely, including regarding postoperative best corrected visual acuity, and were clearly distinct from DMEK. PCA and follow-up univariate analyses suggested that in DMEK, 1) flatter preoperative anterior keratometry (average, K1, and K2) associated with more rebubbling (P = 0.004-0.089) and graft detachment (P = 0.007-0.022); 2) graft marking did not affect postoperative ECD; and 3) lower postoperative endothelial cell density associated with SF6 use (all P > 0.001) and longer surgery (P = 0.005-0.091). All associations are currently under additional investigation in our hospital. PCA is a powerful technique that can rapidly reveal clinically relevant associations in complex ophthalmological datasets.

Customer satisfaction antecedents in uncertain hospitality conditions: an exploratory data mining approach

PurposeInformed by trait and self-determination theories, the present study aims to extend the knowledge regarding the link between customer satisfaction (CS) and its antecedents, including job autonomy (JA), conscientiousness, customer uncertainty (CU) and extra-role customer service (E-RCS) in the hospitality industry.Design/methodology/approachA total of 306 frontline employees were selected from the hotels in North Cyprus, Turkey. Psychometric properties, including the validity and reliability of study variables, were assessed in the first step using confirmatory factor analysis. Then, the data were analyzed utilizing machine learning methods, mainly three exploratory data mining techniques, including lasso regression, decision trees and random forest, as well as partial dependence plots to visualize the role of suggested predictors on the outcome variable.FindingsData mining analysis shows that employees who can modify their job objectives are better equipped to satisfy customers in uncertain situations (JA8). In addition, the findings reveal that employees who believe they work hard to accomplish their personal and organizational goals (CON7) while also having the freedom to decide how to approach their job (JA1) and choose the procedures to utilize (JA2) are more likely to contribute to CS. In general, CS peaked when JA was high, but conscientiousness was moderate, while CU was low.Practical implicationsThis study bridges the gap among various factors at the employee and customer individual, corporate and macro-environmental levels. Hospitality organizations can cultivate a culture of autonomy and independence by promoting open communication and offering growth and development opportunities. This approach enhances conscientious employees’ engagement, leading to exceptional customer service performance, particularly, in uncertain situations.Originality/valueFrom the methodology perspective, this work proposes an opportunity for prospective scientists to broaden the trait and self-determination theories research model by relying on the riches of exploratory techniques without the limits imposed by traditional analytical techniques. Further, this study advances the current knowledge about service agility under uncertainty by extending organizational and service management research to consumer behavior literature.

Combining unsupervised and supervised learning in microscopy enables defect analysis of a full 4H-SiC wafer

Detecting and analyzing various defect types in semiconductor materials is an important prerequisite for understanding the underlying mechanisms and tailoring the production processes. Analysis of microscopy images that reveal defects typically requires image analysis tasks such as segmentation and object detection. With the permanently increasing amount of data from experiments, handling these tasks manually becomes more and more impossible. In this work, we combine various image analysis and data mining techniques to create a robust and accurate, automated image analysis pipeline for extracting the type and position of all defects in a microscopy image of a KOH-etched 4H-SiC wafer.Graphical abstract

Research of the mechanism of spatholobi caulis for immune-inflammatory of osteoarthritis based on clinical data mining,network pharmacology and cell experiment verification

BackgroundTraditional Chinese medicine is widely used in the treatment of osteoarthritis. This study explores the therapeutic effect of Traditional Chinese Medicine (TCM) on osteoarthritis (OA), as well as the mechanism by which Spatholobi Caulis (SPC) improves the immune-Inflammatory response of OA. MethodsThe clinical data mining techniques to evaluate the efficacy of Caulis spatholobi in treating OA, and elucidated its mechanism of action using network pharmacology, molecular docking, and cell experiments. The random walk model was applied to systematically evaluate the therapeutic efficacy of Chinese herbal medicine on OA. ResultsThe evaluation based on the clinical data mining showed that TCM is effective in improving ESR, CRP, C3, C4, and PLT. Data mining results showed significant OA treatment. Network analysis results showed that it may fight OA by targeting JUN, ESR1 and PPARG; molecular docking results showed good JUN, ESR1, PPARG and active component formononetin. Finally, in vitro experiments conducted the preliminary validation of formononetin for the core target of FLS. Cell proliferation was measured by CCK-8, and JUN, ESR1, PPARG protein, and mRNA expression was measured by quantitative real-time RCR at 48 h after drug intervention. ConclusionThe present study demonstrated that the SPC effectively relieved immune inflammation indicators in OA patients. Formononetin, the main active component of the chicken blood vine, acts to reduce the immune validation response in OA patients by reducing the expression of ESR1, JUN, and PPARG.

Higher dimensional sports statistics and real-time game prediction

The rapid expansion of comprehensive sports datasets and the successful application of data mining techniques in various domains have given rise to the emergence of sports data prediction techniques. These techniques enable the extraction of hidden knowledge that can significantly impact the sports industry, as more and more clubs are using Machine Learning (ML) and Deep Learning (DL) methods to manage athletes and training. In this research, the focusing and intriguing aspects is predicting the outcomes of a specific basketball athletes, which has garnered significant attention for research. The paper was motivated by a dual interest in college and NBA basketball matches, alongside a keen observation of the evolving strategies employed by coaches in athlete management. Additionally, the interest was further reinforced by firsthand observations of such evolving methods during a baseball game at City Field in New York. These factors collectively underpin the relevance and significance of this research endeavor, highlighting the intersection of personal interest and the evolving landscape of sports management as compelling reasons for its pursuit. In the process of data selection, we acquired data from previously published essays as well as from Kaggle, a reputable online platform. Following this, we proceeded to evaluate several prominent machine learning models, namely Linear Regression, KNN, Gradient Boosting, Elastic Net, and Lasso, to ascertain their effectiveness in predicting the performance of specific players. Through rigorous analysis and comparison, we concluded that Linear Regression and Gradient Boosting exhibited superior predictive capabilities compared to the other models considered. These two models demonstrated a higher degree of accuracy and reliability in forecasting player performance, thus establishing them as the most suitable choices for our predictive modeling purposes. This meticulous selection process, involving both data acquisition and model evaluation, forms the foundation of our research methodology and underscores the rigor and precision with which our conclusions are drawn.

Application of Data Mining Techniques in Healthcare: Identifying Inter-Disease Relationships through Association Rule Mining

This research focuses on the application of data mining techniques in a healthcare environment by utilizing patient visit data from Hospital X, coded with ICD-10 diagnoses. The purpose of this study is to explore the application of data mining techniques in a healthcare environment, specifically to identify the relationship between diseases using patient visit data from X Hospital. This research utilizes the FP-Growth algorithm method followed by Association Rule Mining to find frequent occurrences of diseases in the data set. The research process involved data pre-processing, transformation into binary format, and careful parameter setting (minimum support 0.95 and confidence 0.9). The results showed a strong association between chronic conditions such as hypertension and diabetes, which are prevalent in the patient population. This association provides insight into potential comorbidities and may assist healthcare providers in improving diagnosis accuracy and treatment effectiveness. This research has implications for the application of data mining techniques, demonstrating its potential in improving predictive analytics in healthcare and strategic planning. This approach not only aids in the efficient allocation of healthcare resources, but also aligns with the broader goal of improving personalized patient care.

CLUSTERING ANALYSIS FOR GROUPING SUB-DISTRICTS IN BOJONEGORO DISTRICT WITH THE K-MEANS METHOD WITH A VARIETY OF APPROACHES

Population data is an important piece of information that is useful for regional planning and development. Insight into the state of an area is more straightforward to observe if there are grouped sub-districts. In this case, data mining techniques can identify patterns and relationships in population data. The K-Means algorithm is a clustering technique that divides data into groups or clusters based on similar characteristics. This research aims to apply the K-Means method with various approaches to clustering sub-districts in the Bojonegoro district according to population data. The research method used is a quantitative method with an exploratory study in the application of the K-Means method with a variety of approaches, namely the use of the Kernel K-Means method by utilizing the mapping function to map data to a higher dimension before the clustering process. In addition, the Fast K-Means method is used, which reduces the model training time to improve the cluster-centered recalibration problem as the amount of data increases. The data source used in this research is secondary population data in the form of birth, death, migrant, and moving variables obtained from the Satu Data Bojonegoro website developed by the Bojonegoro Regency Government. It is found that the best K-Means approach is the Kernel K-Means method with a number of clusters of 5. The performance of the cluster method is evaluated by measuring the average distance within the cluster. The data coordinate pattern in the Kernel K-means method clustering shows a smooth initial trend when the value of the number of clusters is 5 so that the clusters formed are obtained clearly. The conclusion from this study's results is that the K-Means method's best approach in grouping sub-districts in Bojonegoro district is the Kernel K-Means approach.

Prescription Data Mining and Network Pharmacology Study of 1152 Patients with Rectal Prolapse Using Traditional Chinese Medicine.

In recent years, the incidence of rectal prolapse has increased significantly due to the sedentary lifestyle and irregular eating habits of modern life. However, there is a lack of clinical studies on the treatment of rectal prolapse with traditional Chinese medicine (TCM) with a large sample size. Therefore, this study investigated the characteristics of rectal prolapse treatment formulas and then studied the network pharmacology of their core therapeutic drugs, which can help to provide a reference for the treatment and postoperative care of rectal prolapse patients. This study aimed to explore the prescription characteristics and the mechanism of action of core drugs in the treatment of rectal prolapse in Chinese medicine through data mining and bioinformatics techniques. We collected the diagnosis and treatment information of patients with rectal prolapse from January 2014 to September 2021 in the electronic case database of Nanjing Hospital of TCM, mined the patient information and prescription features using R, screened the active ingredients of the core pairs of drugs and disease drug intersection targets using TCMSP and GnenCard databases, and constructed a Protein-protein interaction (PPI) network using STRING and Cytoscape, and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of the intersecting targets were performed using Metascape and R. We found that prolapse is easy to occur in people over 50 years old, preferably in autumn and winter. Commonly used therapeutic Chinese medicines include Glycyrrhiza glabra, Radix angelicae sinensis, Radix astragali, Atractylodes macrocephala, and Pericarpium citri reticulatae, which are mostly deficiency tonic medicines, warm in nature, and belong to spleen meridian. The core therapeutic medicinal pair was "Bupleuri radix-Cimicifugae rhizoma". There were 190 common targets of Bupleuri radix and Cimicifugae rhizoma, and 71 intersection targets of the drug pair and prolapse. The main components of the core drugs for the treatment of prolapse may be quercetin, kaempferol, Stigmasterol, etc, and the core targets may be CASP3, AKT1, HIF1A, etc. The total number of GO entries for the intersection targets of "Bupleuri radix-Cimicifugae rhizoma" and diseases was 3495, among which the molecular functions accounted for the largest proportion, mainly Pathways in cancer, IL-18 signaling pathway, etc. KEGG enriched pathway analysis yielded 168 results, and the major pathways were pathways in cancer, lipid and atherosclerosis, IL-17 signaling pathway, etc. Conclusion: This study adopted real-world research methodology and used data mining and bioinformatics technology to mine the medication law of rectal prolapse and its core drug action mechanism from the clinical information of Chinese medicine.

Predicting game-induced emotions using EEG, data mining and machine learning

BackgroundEmotion is a complex phenomenon that greatly affects human behavior and thinking in daily life. Electroencephalography (EEG), one of the human physiological signals, has been emphasized by most researchers in emotion recognition as its specific properties are closely associated with human emotion. However, the number of human emotion recognition studies using computer games as stimuli is still insufficient as there were no relevant publicly available datasets provided in the past decades. Most of the recent studies using the Gameemo public dataset have not clarified the relationship between the EEG signal’s changes and the emotion elicited using computer games. Thus, this paper is proposed to introduce the use of data mining techniques in investigating the relationships between the frequency changes of EEG signals and the human emotion elicited when playing different kinds of computer games. The data acquisition stage, data pre-processing, data annotation and feature extraction stage were designed and conducted in this paper to obtain and extract the EEG features from the Gameemo dataset. The cross-subject and subject-based experiments were conducted to evaluate the classifiers’ performance. The top 10 association rules generated by the RCAR classifier will be examined to determine the possible relationship between the EEG signal's frequency changes and game-induced emotions.ResultsThe RCAR classifier constructed for cross-subject experiment achieved highest accuracy, precision, recall and F1-score evaluated with over 90% in classifying the HAPV, HANV and LANV game-induced emotions. The 20 experiment cases’ results from subject-based experiments supported that the SVM classifier could accurately classify the 4 emotion states with a kappa value over 0.62, demonstrating the SVM-based algorithm’s capabilities in precisely determining the emotion label for each participant’s EEG features’ instance.ConclusionThe findings in this study fill the existing gap of game-induced emotion recognition field by providing an in-depth evaluation on the ruleset algorithm’s performance and feasibility of applying the generated rules on the game-induced EEG data for justifying the emotional state prediction result.

Augmentation of Soft Partition with a Granular Prototype Based Fuzzy C-Means

Clustering is a fundamental cornerstone in unsupervised learning, playing a pivotal role in various data mining techniques. The precise and efficient classification of data stands as a central focus for numerous researchers and practitioners alike. In this study, we design an effective soft partition classification method which refines and extends the prototype of the well-known Fuzzy C-Means clustering algorithm. Specifically, the developed scheme employs membership function to extend the prototypes into a series of granular prototypes, thus achieving a deeper revelation of the structure of the data. This process softly divides the data into core and extended parts. The core part can be succinctly encapsulated through several information granules, whereas the extended part lacks discernible geometry and requires formal descriptors (such as membership formulas). Our objective is to develop information granules that shape the core structure within the dataset, delineate their characteristics, and explore the interaction among these granules that result in their deformation. The granular prototypes become the main component of the information granules and provide an optimization space for traditional prototypes. Subsequently, we apply quantum-behaved particle swarm optimization to identify the optimal partition matrix for the data. This optimized matrix significantly enhances the partition performance of the data. Experimental results provide substantial evidence of the effectiveness of the proposed approach.

Improving CTR Prediction in Advertising with XGBoost

Click Through Rate (CTR) prediction is crucial in digital advertising for optimizing marketing strategies. This paper presents a review of significant contributions in this field, highlighting methodologies and findings from various studies. Pioneering research laid foundational groundwork for CTR estimation methods, while subsequent analyses explored the impact of ad types and design effects on user engagement. Utilization of data mining techniques and the proposal of advanced prediction models further enhanced CTR prediction accuracy. Additionally, this paper introduces our method utilizing XGBoost, a powerful ensemble learning algorithm, to address existing challenges and enhance CTR prediction accuracy. This review offers valuable insights for marketers aiming to optimize their advertising campaigns in the dynamic landscape of advertising.