Data Mining Approach Research Articles

A Multi-dimensional Data Mining-based Study on the Prescriptions Developed by Professor Xu Zhiyin in Treating Thyroid Nodules.

In this study, we employed a multi-dimensional data mining approach to examine the clinical instances where Professor Xu Zhiyin treated thyroid nodules. Our aim is to understand the patterns of symptoms, underlying causes, and treatment approaches used for thyroid nodules. By doing so, the intention is to distill the essential aspects, compile Professor Xu Zhiyin's clinical insights, and investigate his scholarly perspectives. Professor Xu Zhiyin's clinical diagnoses and treatments spanning from 2009 to 2019 were entered into Microsoft Excel. Subsequently, the collected data was imported into the Medcase V5.2 system to facilitate data mining. Various techniques, such as frequency-based method, association rule analysis, and clustering, including a decentralized system clustering approach, were employed on a set of 346 cases involving patients with thyroid nodules that conformed to the specified criteria. The primary focus was on extracting insights regarding symptoms and the underlying causes from the medical records. By integrating these findings with Professor Xu Zhiyin's clinical expertise, we examined and summarized the outcomes of the data mining process. The fundamental prescriptions were successfully extracted using the techniques for mining across multiple dimensions. Utilizing the scattered grouping of these prescriptions and with reference to the cluster analysis of the frequency-linked system, the fundamental prescriptions proposed by Professor Xu Zhiyin for addressing thyroid nodules encompass the following ingredients: Glycyrrhiza uralensis Fisch, Cortex Moutan, Paeoniae radix rubra, Curcuma longa L., Radix Curcumae, persica seed, Citri Reticulatae Viride Pericarpium, Pinellia ternata, Spica Prunellae, Ostreae concha, Gleditsia sinensis spine, Tuckahoe and Radix Codonopsis. The fundamental prescriptions were acquired using the frequency approach, association rule technique, k-means clustering approach, and systematic clustering approach. The research findings corroborate one another, demonstrating that Professor Xu Zhiyin's approach to distinguishing and treating thyroid nodules is embodied in distinct prescriptions tailored to specific diseases.

Determining consumer demand patterns for production planning using a data mining approach

Determining consumer demand patterns for production planning using a data mining approach

Modeling performance evaluation in badminton sports: a fuzzy logic approach

Spectators and many young students have flocked to badminton matches in recent years. Badminton practice has received a lot of media coverage. The current state of badminton evaluation methods is lacking in reliability. This article's overarching goal is to examine the many applications of fuzzy logic in badminton performance evaluation and improvement. Data on the badminton technique's flexion and extension phases are mapped into the suggested model using a fuzzy inference system (FIS). This study suggests a fuzzy logic-based badminton-specific objective fuzzy inference system (Bmt-FIS) to evaluate team sports. Despite the gravity of the situation, decisions involving performance reviews often use subjective data. These common decision-making problems may be realistically addressed by fuzzy logic models. Fuzzy logic has the potential to be an effective tool in situations where both quantitative and qualitative data interpretation are allowed. To do this, it accounts for the inherent variability in athletic performance by taking into consideration the 'hazy' or 'uncertain' limitations of data. By taking limitations into account, a rule-based approach makes performance evaluation more precise. Here, a fuzzy inference system (FIS) uses the input variables to evaluate the student's performance. While data mining approaches have been studied, the adaptive neural fuzzy method outperforms others because of its exceptional accuracy.This method eloquently and clearly conveys the many levels of integrity and ambiguity. Also, fuzzy logic may be a great tool for evaluating badminton skills. This foundational study connects the dynamic realm of sports with static measures.

Prediction Model of School Drop Out Factors Using Classification Techniques in Selangor

Malaysia has been struggling to sustain the number of students graduating from school, with an increasing number leaving early, which poses significant concerns for the future generation. Several factors contribute to this issue, such as economic constraints, geographical challenges, transportation problems, and sociocultural norms. This paper uses a data mining approach to identify the attributes that lead to school dropouts and to determine the predictive model with the highest accuracy for forecasting dropout rates. The application of data mining approaches has proven effective in predicting students at risk of dropping out in general education. Nevertheless, there is a shortage of data mining-related studies on student attrition in public schools in Malaysia. The study utilized student and school datasets with consent from the respective departments in the Ministry of Education. This data includes information from 2,482 students across various primary schools in Selangor with initially 22 attributes collected from the dataset. After the attributes undergone feature selection process by using InfoGainAttributeEval, there are 12 features left including class attribute Status_DO. The collected data encompasses student demographics, academic performance and socioeconomic background. The experiments for this study used Decision Trees (J48), Naïve Bayes and Random Forest. By using classification techniques that were made available in WEKA, all attributes from the dataset were tested. The results of the analysis shown that Random Forest with the highest accuracy of 79.5729% in term of predicting student drop out hence indicate the reliability of this research as a decision support tool.

Patterns and factors associated with dental service utilization among insured people: a data mining approach

BackgroundInsurance databases contain valuable information related to the use of dental services. This data is instrumental in decision-making processes, enhancing risk assessment, and predicting outcomes. The objective of this study was to identify patterns and factors influencing the utilization of dental services among complementary insured individuals, employing a data mining methodology.MethodsA secondary data analysis was conducted using a dental insurance dataset from Iran in 2022. The Cross-Industry Standard Process for Data Mining (CRISP-DM) was employed as a data mining approach for knowledge extraction from the database. The utilization of dental services was the outcome of interest, and independent variables were chosen based on the available information in the insurance dataset. Dental services were categorized into nine groups: diagnostic, preventive, periodontal, restorative, endodontic, prosthetic, implant, extraction/surgical, and orthodontic procedures. The independent variables included age, gender, family size, insurance history, franchise, insurance limit, and policyholder. A multinomial logistic regression model was utilized to investigate the factors associated with dental care utilization. All analyses were conducted using RapidMiner Version 2020.ResultsThe analysis encompassed a total of 654,418 records, corresponding to 118,268 insured individuals. Predominantly, restorative treatments were the most utilized services, accounting for approximately 38% of all services, followed by diagnostic (18.35%) and endodontic (13.3%) care. Individuals aged between 36 and 60 years had the highest rate of utilization for any dental services. Additionally, families comprising three to four members, individuals with a one-year insurance history, people contracted with a 20% franchise, individuals with a high insurance limit, and insured individuals with a small policyholder, exhibited the highest rate of service usage compared to their counterparts. The regression model revealed that all independent variables were significantly associated with the use of dental services. However, the patterns of association varied among different service categories.ConclusionsRestorative treatments emerged as the most frequently used dental services among insured individuals, followed by diagnostic and endodontic procedures. The pattern of service utilization was influenced by the characteristics of the insured individuals and attributes related to their insurance.

A Scalable and Extensible Logical Data Model of Electronic Health Record Audit Logs for Temporal Data Mining (RNteract): Model Conceptualization and Formulation.

Increased workload, including workload related to electronic health record (EHR) documentation, is reported as a main contributor to nurse burnout and adversely affects patient safety and nurse satisfaction. Traditional methods for workload analysis are either administrative measures (such as the nurse-patient ratio) that do not represent actual nursing care or are subjective and limited to snapshots of care (eg, time-motion studies). Observing care and testing workflow changes in real time can be obstructive to clinical care. An examination of EHR interactions using EHR audit logs could provide a scalable, unobtrusive way to quantify the nursing workload, at least to the extent that nursing work is represented in EHR documentation. EHR audit logs are extremely complex; however, simple analytical methods cannot discover complex temporal patterns, requiring use of state-of-the-art temporal data-mining approaches. To effectively use these approaches, it is necessary to structure the raw audit logs into a consistent and scalable logical data model that can be consumed by machine learning (ML) algorithms. We aimed to conceptualize a logical data model for nurse-EHR interactions that would support the future development of temporal ML models based on EHR audit log data. We conducted a preliminary review of EHR audit logs to understand the types of nursing-specific data captured. Using concepts derived from the literature and our previous experience studying temporal patterns in biomedical data, we formulated a logical data model that can describe nurse-EHR interactions, the nurse-intrinsic and situational characteristics that may influence those interactions, and outcomes of relevance to the nursing workload in a scalable and extensible manner. We describe the data structure and concepts from EHR audit log data associated with nursing workload as a logical data model named RNteract. We conceptually demonstrate how using this logical data model could support temporal unsupervised ML and state-of-the-art artificial intelligence (AI) methods for predictive modeling. The RNteract logical data model appears capable of supporting a variety of AI-based systems and should be generalizable to any type of EHR system or health care setting. Quantitatively identifying and analyzing temporal patterns of nurse-EHR interactions is foundational for developing interventions that support the nursing documentation workload and address nurse burnout.

Clustering and Trend Analysis of Priority Commodities in the Archipelago Capital Region (IKN) using a Data Mining Approach

The policy of moving the capital from Jakarta to East Kalimantan planned by the President of the Republic of Indonesia Joko Widodo has caused a lot of polemic among the public. There are quite a few positive and negative comments on social media regarding the policy of moving the capital. The process of moving the capital requires careful preparation. One thing that needs to be considered is food security in IKN. This research provides recommendations for the main food commodities in IKN by applying data mining. We collect food productivity data available on the official website for East Kalimantan province. These data are processed and grouped into two groups, namely horticulture and livestock products using the K-Means method. After grouping, we predict the increase in productivity of each group using the ARIMA method. This research produces output in the form of grouping commodities into horticulture and livestock products. Productivity results for each type of commodity are displayed from 2016 to 2020 based on data on the official East Kalimantan Province website. Based on this data, predictions are made using the ARIMA method to predict productivity results from 2021 to 2025. Commodities with total productivity are grouped into high-priority commodities. Grouping the amount of productivity is carried out using the clustering method by comparing the amount of productivity for each commodity and producing commodities that are low priority, middle priority, priority and top priority based on the highest to lowest productivity numbers. The cluster quality for grouping horticultural commodities is 99.1%, while the cluster quality for grouping livestock commodities is 87.5%. Hasil prediksi terbaik yaitu ketika memprediksi produksi salak dan slaughter cattle dengan model ARIMA (0, 1, 0) dan ARIMA (2, 2, 2).

A Smart IoT Solution for Monitoring and Predicting Grocery Freshness and Quality

According to the statistics,40% of the food produced is wasted before it reaches the consumers. The core problem lies in logistics and infrastructure. Firstly weather varies widely and it makes generalization difficult. This makes it impossible for perishables to naturally survive long-haul transport. Secondly, cold storage facility is not available and affordable to all farmers. Today food spoilage is a crucial problem. The food we consume can affect in any form of contamination that may occur due to storage or chemical changes within the food. This makes it necessary to develop a system that can help people to identify and predict the freshness of food or quality of food items. A smart grocery management system helps users manage groceries in large warehouses and long transportations. Users no longer have to worry about continuous food monitoring. This project aims to ensure the freshness of grocery by predicting the spoilage of grocery in prior and to notify the user about the same so that he can take necessary precautions to prevent grocery spoilage. Continuous analysis of temperature and humidity of the grocery is carried out by using DHT11 Temperature and Humidity sensor to monitor varying parameters of the grocery. ESP8266 Node MCU is used as the main controller of the system which is used to control the sensing action of the sensor. Data collected from the sensor is used to analyse the variations in the values of temperature and humidity when the grocery is fresh and when it is completely spoilt. The humidity and temperature data are sent to the cloud storage by using the Wi-Fi module which is inbuilt to the Node MCU. ThingSpeak cloud platform is set up for the storage purpose which is retrieved back to use it for predicting the spoilage of the grocery which is the main goal of the system. A prediction algorithm is used for this purpose which is basically a combination of machine learning and data mining approach for forecasting or predicting the future values based on some past set of values. The values of fresh grocery has been given as the input to prediction algorithm which is used to predict the future values of temperature and humidity. Predicted values obtained is compared with the spoilt values measured earlier to detect the spoilage of grocery. Predicted status of the grocery has been notified to the user through SMS. Key Words: food, grocery, freshness, prediction

A model based on PDCA and data mining approach for the prevention of occupational accidents in the plumbing activity in the construction sector.

Occupational accidents in the plumbing activity in the construction sector in developing countries have high rates of work absenteeism. The productivity of enterprises is heavily influenced by it. To propose a model based on the Plan, Do, Check, and Act cycle and data mining for the prevention of occupational accidents in the plumbing activity in the construction sector. This cross-sectional study was administered on a total of 200 male technical workers in plumbing. It considers biological, biomechanical, chemical, and, physical risk factors. Three data mining algorithms were compared: Logistic Regression, Naive Bayes, and Decision Trees, classifying the occurrences occupational accident. The model was validated considering 20% of the data collected, maintaining the same proportion between accidents and non-accidents. The model was applied to data collected from the last 17 years of occupational accidents in the plumbing activity in a Colombian construction company. The results showed that, in 90.5% of the cases, the decision tree classifier (J48) correctly identified the possible cases of occupational accidents with the biological, chemical, and, biomechanical, risk factors training variables applied in the model. The results of this study are promising in that the model is efficient in predicting the occurrence of an occupational accident in the plumbing activity in the construction sector. For the accidents identified and the associated causes, a plan of measures to mitigate the risk of occupational accidents is proposed.

RNA-binding protein transcripts as potential biomarkers for detecting Primary Sclerosing Cholangitis and for predicting its progression to Cholangiocarcinoma.

Primary Sclerosing Cholangitis (PSC) is a persistent inflammatory liver condition that affects the bile ducts and is commonly diagnosed in young individuals. Despite efforts to incorporate various clinical, biochemical and molecular parameters for diagnosing PSC, it remains challenging, and no biomarkers characteristic of the disease have been identified hitherto. PSC is linked with an uncertain prognosis, and there is a pressing need to explore multiomics databases to establish a new biomarker panel for the early detection of PSC's gradual progression into Cholangiocarcinoma (CCA) and for the development of effective therapeutic interventions. Apart from non-coding RNAs, other components of the Ribonucleoprotein (RNP) complex, such as RNA-Binding Proteins (RBPs), also hold great promise as biomarkers due to their versatile expression in pathological conditions. In the present review, an update on the RBP transcripts that show dysregulated expression in PSC and CCA is provided. Moreover, by utilizing a bioinformatic data mining approach, we give insight into those RBP transcripts that also exhibit differential expression in liver and gall bladder, as well as in body fluids, and are promising as biomarkers for diagnosing and predicting the prognosis of PSC. Expression data were bioinformatically extracted from public repositories usingTCGA Bile Duct Cancer dataset for CCA and specific NCBI GEO datasets for both PSC and CCA; more specifically, RBPs annotations were obtained from RBP World database. Interestingly, our comprehensive analysis shows an elevated expression of the non-canonical RBPs, FANCD2, as well as the microtubule dynamics regulator, ASPM, transcripts in the body fluids of patients with PSC and CCA compared with their respective controls, with the same trend in expression being observed in gall bladder and liver cancer tissues. Consequently, the manipulation of tissue expression of RBP transcripts might be considered as a strategy to mitigate the onset of CCA in PSC patients, and warrants further experimental investigation. The analysis performed herein may be helpful in the identification of non-invasive biomarkers for the early detection of PSC and for predicting its progression into CCA. In conclusion, future clinical research should investigate in more depth the full potential of RBP transcripts as biomarkers for human pathologies.

Application of dominance-based rough set approach in vaginal dosage forms optimization

The objective of this study was to identify cause-effect relationships between various data related to vaginal dosage forms, such as composition, process parameters (condition attributes), and quality (decision attribute), by employing a mathematical data mining approach that utilizes Rough Set Theory. The analyzed data were organized in a tabular format known as an information system. Objects were labeled as rows, and attributes as columns, with attribute-values serving as entries. Each formulation was described by its condition attributes (composition, manufacturing process, quality parameters), while the decision attribute indicated the quality of the formulation (expressed as pH after vaginal tablet or pessary application) and enabled classification as correct/incorrect or good/bad. The decision rules generated by the study provide insights into relationships between the condition attributes and the decision attribute. They highlight the importance of LA content and LA:polymer ratios in maintaining physiological pH. Additionally, the findings confirm the favorable impact of chitosan on pH regulation while indicating that the type of formulation (pessaries or tablets) does not significantly affect the classification of objects. Decision rules with high confirmation coefficients of condition attributes can help developing optimal formulations. These decision rules contain essential information that describes the cause-and-effect relationships and are free of redundant information. They form the basis for understanding and optimizing processes, and establishing a design space.

Enhancing employees’ quality of work life and engagement to foster corporate social responsibility: a data mining approach

PurposeThis paper aims to present a dynamic model for strategic and personalized decision-making in human resources (HR), using data mining techniques to enhance corporate social sustainability (CSS). The focus is on the interconnectedness of employee engagement (EE), enablement and the quality of work life.Design/methodology/approachThe proposed model integrates various HR data, including demographic information, job specifications, payment and rewards, attendance and absence, alongside employees’ perceptions of their work-life quality, engagement and enablement. Data mining processes are applied to generate meaningful insights for senior and middle managers.FindingsThe study implemented the model within a production organization, revealing that factors influencing EE and enablement differ based on gender, marital status and occupational group. Performance-based rewards play a significant role in enhancing engagement, regardless of the reward amount. Factors such as “being recognized for competency” influence engagement for women, while payment has a greater impact on men. Engagement does not directly influence the quality of work life, but subcomponents like perceived transparency and the organization’s processes, particularly the “employee performance evaluation system,” improve work-life quality.Research limitations/implicationsThe findings are specific to the studied organization, limiting generalizability. Future research should explore the model’s effectiveness in different cultural and organizational settings.Practical implicationsThe proposed model provides practical implications for organizations that enhance CSS. Organizations can gain insights into factors influencing EE and enablement by using data mining techniques, enabling informed decision-making and tailored human resource management practices.Social implicationsThis research addresses the societal concern regarding the impact of business activities on sustainability. Organizations can contribute to a more socially responsible and sustainable business environment by focusing on work-life quality and EE.Originality/valueThis paper offers a dynamic model using data mining and machine learning techniques for sustainable human resource management. It emphasizes the importance of customization to align practices with the unique needs of the workforce.

Prediction of the operational performance of a vehicle seat thermal management system using statistical and machine learning techniques

To assess and forecast the operational performance of a modified car seat for thermal management using an air conditioning system, statistical and machine learning (ML) models were used. By extending evaporator/condenser coils beneath the back and cushion surfaces of the car seat and using operational data on the HVAC system, such as seat temperature readings, an interval of operation was gathered. Using a data mining approach, statistically relevant factors and varying the compressor speed from 500 to 1600 rpm under various scenarios to model the system were selected. Utilizing key feature variables, our data-driven approach yielded predictions with favorable accuracy for the Coefficient of Performance (COP) of the HVAC system. By using the Akaike Information Criterion (AIC) to improve the Linear Regression (LR) model, the Root Mean Square Error (RMSE) dropped to 0.20, the Mean Absolute Error (MAE) dropped to 0.16, and the Coefficient of Determination (R2) increased to 98 %. The Random Forest (RF) model, optimized with hyperparameters, demonstrated moderate predictive capability, with RMSE (0.52), MAE (0.37), and R2 (94 %). Furthermore, polynomial feature augmentation, individual and combined predictor analysis, and iterative predictor combinations all improved predictive accuracy. Detailed information on the algorithms was given for the sake of other researchers.

EPH9 Safety Assessment of Nirmatrelvir-Ritonavir in the FDA Adverse Event Reporting System: A Data Mining Approach

EPH9 Safety Assessment of Nirmatrelvir-Ritonavir in the FDA Adverse Event Reporting System: A Data Mining Approach

Data mining approach for prediction of academic success in open and distance education

Predicting and improving the academic achievement of university students is a multifactorial problem. Considering the low success rates and high dropout rates, particularly in open education programs characterized by mass enrollment, academic success is an important research area with its causes and consequences. This study aimed to solve a classification problem (successful or unsuccessful), predict students’ academic success, and identify those at risk. The primary objective was to predict the academic success status with 26,708 students enrolled in Istanbul University open and distance education programs between 2011 and 2017. Predictions were based demographic data and success grades in Turkish, Atatürk's Principles and History of Revolution, English, and Disaster Culture courses. The study utilized classification models from supervised learning algorithms and was conducted using the SPSS Modeler 18 program. Initially, the data was divided into 70% training and 30% test data. Then, models were constructed by using Random Forest, Tree-AS, C&RT, C5.0, CHAID, QUEST, Naive Bayes, Logistic Regression, NeuralNet, and SVM algorithms. Model performances were compared according to accuracy, sensitivity, specificity, F1 score, positive predictive value, negative predictive value, and Matthews Correlation Coefficient criteria. The C&RT model demonstrated the best performance, achieving the highest specificity value of 0.915.

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.

Repurposing of drugs against bacterial infections: A pharmacovigilance-based data mining approach.

The World Health Organization (WHO) has published a list of priority pathogens that urgently require research to develop new antibiotics. The main aim of the current study is to identify potential marketed drugs that can be repurposed against bacterial infections. A pharmacovigilance-based drug repurposing approach was used to identify potential drugs. OpenVigil 2.1 tool was used to query the FDA Adverse Event Reporting System database. The reporting odds ratio (ROR) < 1, ROR95CI upper bound <1, and no. of cases ≥30 were used for filtering and sorting of drugs. Sunburst plot was used to represent drugs in a hierarchical order using the Anatomical Therapeutic Chemical classification. Molecular docking and dynamics were performed using the Maestro and Desmond modules of Schrodinger 2023 software respectively. A total of 40 drugs with different classes were identified based on the pharmacovigilance approach which has antibacterial potential. The molecular docking results have shown energetically favored binding conformation of lisinopril against 3-deoxy-manno-octulosonate cytidylyltransferase, UDP-2,3-diacylglucosamine hydrolase, and penicillin-binding protein 3 (PBP3) of Pseudomonas aeruginosa; olmesartan, atorvastatin against lipoteichoic acids flippase LtaA and rosiglitazone and varenicline against d-alanine ligase of Staphylococcus aureus; valsartan against peptidoglycan deacetylase (SpPgdA) and atorvastatin against CDP-activated ribitol for teichoic acid precursors of Streptococcus pneumoniae. Further, molecular dynamic results have shown the stability of identified drugs in the active site of bacterial targets except lisinopril with PBP3. Lisinopril, olmesartan, atorvastatin, rosiglitazone, varenicline, and valsartan have been identified as potential drugs for repurposing against bacterial infection.

Digital banking fortification: a real-time isolation forest architecture for detecting online transaction fraud

Since the use of the Internet has increased exponentially, numerous organizations, including the financial industry, offer services online. As a result, financial scams are expanding in quantity and complexity worldwide, resulting in massive revenue losses and making digital fraudulent transactions a severe issue. Abnormal attempts and illegal access are instances of these dangers that fraudulent activity detection systems must identify. Machine learning and data mining approaches have been extensively used to address this issue in recent years. However, these approaches must be enhanced regarding real-time detection speed, tackling enormous amounts of data, and finding undiscovered attack patterns. Consequently, the present study provides a real-time architecture for averting and identifying digital transaction fraud, which relies on the Isolation Forest (IForest) approach and big data analytic tools, including Spark Streaming, sparkling water, Kafka, and PostgreSQL. This architecture seeks to improve present detection strategies by increasing accuracy for detection when considering enormous amounts of data. Two real datasets of online transactional fraud are used to assess the proposed architecture, and the findings are compared to relevant studies. The investigation results showed that IForest performed flawlessly, achieving an accuracy of 0.99 in two datasets.

Technical–tactical differences between female and male elite football: A data mining approach through neural network analysis, binary logistic regression, and decision tree techniques

The technical−tactical performance of women’s football has improved markedly in recent years. Despite this improvement, there are still differences between men’s football and women’s football. The objectives of this study were to know the technical and tactical key performance indicators (KPIs) that differentiate elite men’s and women’s football teams as well as to determine which statistical techniques demonstrate superior classification ability and interpretability in football terms. For this purpose, 768 matches corresponding to the latest editions of the UEFA Champions League, UEFA Euro and FIFA World Cup for men and women were analyzed. First, the differences at the bivariate level were analyzed using student’s t-test for independent sample ( p &lt; 0.05) for the male and female teams. Secondly, three data mining classification algorithms were applied: (i) Artificial Neural Network (ANN), (ii) Binary Logistic Regression, and (iii) Decision Tree. Significant differences were found between men’s football and women’s football in variables related to technical elements such as lost balls (ES = 1.19), ball recoveries (ES = 1.00), and accurate passes (ES = 0.97), as well as regulatory aspects like fouls (ES = 0.59), successful tackles (ES = 0.46), and yellow cards (0.45). On the other hand, the classification models presented excellent or good predictive capability [Range AUC 0.774−0.982], with very small differences between the ANN’s and logistic regression models. This result justifies the use of simpler models as the linear regression model to understand the differences between men’s and women’s football. Moreover, the observed differences may offer insights for future efforts aimed at enhancing the performance of women’s football.

Predicting Public Personnel Selection Examination Achievement: A Data Mining Approach

This research investigates the predictive variables related to the Public Personnel Selection Examination (KPSS), utilized for recruitment in public institutions and organizations. The study explores predictor variables' importance levels by analysing longitudinal data, including examinees' high-stakes exams, demographic information, and educational backgrounds. It compares the prediction performances of machine learning algorithms such as artificial neural networks, random forest, support vector machine, and k-nearest neighbour. The findings reveal that the quantitative test of the graduate education exam is the most influential predictor, closely followed by the mathematics test of the university entrance exam. These results highlight the importance of quantitative reasoning skills in predicting KPSS achievement. Additionally, variables related to undergraduate programs and universities demonstrate significant importance in predicting KPSS achievement. Notably, the artificial neural networks model demonstrates superior predictive accuracy compared to other models, indicating its effectiveness in KPSS prediction. This research sheds light on important predictors of KPSS achievement and provides valuable insights into the effectiveness of different prediction models.