Association Rule Mining Research Articles

Educational data mining in a university using LMS Moodle

Educational data mining is a technology for processing big data collected through educational portals and used for assessing the quality of educational content and predicting educational outcomes. Various data analysis methods are applied in the implementation of this technology, including anomaly detection, association rule mining, classification, cluster analysis, regression analysis, and factor analysis. Each of these methods uses special tools, such as spreadsheets, high­level programming languages (Python or R), low­code platforms (Loginom, Orange), or specialized analytics modules integrated into educational portals. The article reveals the features of the “Analytics” module embedded in LMS Moodle for tracking and optimizing the educational process. The authors provide recommendations for configuring and using this module in the educational process. The module “Analytics” allows to create forecasts of the success rate of mastering a separate training course without programming skills , does not require data preprocessing for analysis, and allows to quickly take measures to improve the educational process based on the results of the forecast. However, the module has a number of disadvantages (limited attributes and forecasting models, high requirements for computing power to run model training and build forecasts), so often administrators of educational portals disable this module. In this regard, an experiment was conducted to train a classification model for predicting students’ examination scores based on students’ activities on the educational portal of Nosov Magnitogorsk State Technical University. In the course of the experiment, data on students’ activity on the university’s educational portal for one academic year were taken, and machine learning models were built to predict examination scores for an individual student and subject. The results of the experiment showed F1 74 % in predicting the score that the students would receive on the intermediate assessment. This confirms the effectiveness of using educational data mining based on user data from educational platforms.

A Heuristic Approach to Analysis of the Genetic Susceptibility Profile in Patients Affected by Airway Allergies.

Allergic respiratory diseases such as asthma might be considered multifactorial diseases, having a complex pathogenesis that involves environmental factors and the activation of a large set of immune response pathways and mechanisms. In addition, variations in genetic background seem to play a central role. The method developed for the analysis of the complexities, as association rule mining, nowadays may be applied to different research areas including genetic and biological complexities such as atopic airway diseases to identify complex genetic or biological markers and enlighten new diagnostic and therapeutic targets. A total of 308 allergic patients and 205 controls were typed for 13 single nucleotide polymorphisms (SNPs) of cytokine and receptors genes involved in type 1 and type 2 inflammatory response (IL-4 rs2243250 C/T, IL-4R rs1801275A/G, IL-6 rs1800795 G/C, IL-10 rs1800872 A/C and rs1800896 A/G, IL-10RB rs2834167A/G, IL-13 rs1800925 C/T, IL-18 rs187238G/C, IFNγ rs 24030561A/T and IFNγR2 rs2834213G/A), the rs2228137C/T of CD23 receptor gene and rs577912C/T and rs564481C/T of Klotho genes, using KASPar SNP genotyping method. Clinical and laboratory data of patients were analyzed by formal statistic tools and by a data-mining technique-market basket analysis-selecting a minimum threshold of 90% of rule confidence. Formal statistical analyses show that IL-6 rs1800795GG, IL-10RB rs2834167G positive genotypes, IL-13 rs1800925CC, CD23 rs2228137TT Klotho rs564481TT, might be risk factors for allergy. Applying the association rule methodology, we identify 10 genotype combination patterns associated with susceptibility to allergies. Together these data necessitate being confirmed in further studies, indicating that the heuristic approach might be a straightforward and useful tool to find predictive and diagnostic molecular patterns that might be also considered potential therapeutic targets in allergy.

Personalized Book Recommendations: A Hybrid Approach Leveraging Collaborative Filtering, Association Rule Mining, and Content-Based Filtering

Personalized Book Recommendations: A Hybrid Approach Leveraging Collaborative Filtering, Association Rule Mining, and Content-Based Filtering

Data Mining and in Silico Analysis of Ethiopian Traditional Medicine: Unveiling the Therapeutic Potential of Rumex abyssinicus Jacq.

Multicomponent traditional medicine prescriptions are widely used in Ethiopia for disease treatment. However, inconsistencies across practitioners, cultures, and locations have hindered the development of reliable therapeutic medicines. Systematic analysis of traditional medicine data is crucial for identifying consistent and reliable medicinal materials. In this study, we compiled and analyzed a dataset of 505 prescriptions, encompassing 567 medicinal materials used for treating 106 diseases. Using association rule mining, we identified significant associations between diseases and medicinal materials. Notably, wound healing-the most frequently treated condition-was strongly associated with Rumex abyssinicus Jacq., showing a high support value. This association led to further in silico and network analysis of R. abyssinicus Jacq. compounds, revealing 756 therapeutic targets enriched in various KEGG pathways and biological processes. The Random-Walk with Restart (RWR) algorithm applied to the CODA PPI network identified these targets as linked to diseases such as cancer, inflammation, and metabolic, immune, respiratory, and neurological disorders. Many hub target genes from the PPI network were also directly associated with wound healing, supporting the traditional use of R. abyssinicus Jacq. for treating wounds. In conclusion, this study uncovers significant associations between diseases and medicinal materials in Ethiopian traditional medicine, emphasizing the therapeutic potential of R. abyssinicus Jacq. These findings provide a foundation for further research, including in vitro and in vivo studies, to explore and validate the efficacy of traditional and natural product-derived medicines.

Exploring the predictive factors of heart disease using rare association rule mining

Cardiovascular diseases continue to be the leading cause of mortality worldwide, claiming a significant number of lives each year. Despite the advancements in predictive models, including logistic regression, neural networks, and random forests, these techniques often lack transparency and interpretability, limiting their practical application in clinical settings. To address this challenge, this research introduces EPFHD-RARMING, an innovative approach designed to enhance the understanding and predictability of heart disease through the discovery of rare and meaningful patterns. EPFHD-RARMING utilizes rare association rule mining to uncover hidden and unexpected rules that identify critical factors contributing to heart disease. This method is particularly adept at identifying high-risk patterns in individuals who appear healthy but may develop heart disease under certain conditions, thus facilitating early intervention and preventive measures. By integrating these insights with established feature engineering techniques, EPFHD-RARMING enhances its practical utility, enabling medical professionals to proactively manage patient care and tailor interventions to individual risk profiles. This study demonstrates the effectiveness of EPFHD-RARMING in providing a deeper, actionable understanding of the complex dynamics of heart disease. The model’s ability to identify and interpret rare patterns holds significant promise for advancing medical analytics and improving patient outcomes. Moreover, the applicability of EPFHD-RARMING extends beyond the healthcare domain, offering valuable insights in various fields where the discovery of rare patterns is critical, such as finance, marketing, and cybersecurity. This study conducts a comprehensive evaluation, which demonstrates the superior performance of EPFHD-RARMING compared to traditional predictive models in identifying key factors contributing to heart disease, in terms of interestingness, explainability, and comprehensiveness of insights. The results underscore the potential of this innovative approach to revolutionize our understanding and prediction of heart disease, ultimately contributing to more effective and personalized healthcare solutions. This research emphasizes the importance of rare association rule mining in medical analytics and paves the way for future studies to explore and utilize these techniques across diverse domains.

SAFARM: simulated annealing based framework for association rule mining

SAFARM: simulated annealing based framework for association rule mining

Integration of Price Data in Wind Power Sales Systems Using Association Rule Mining Combined with Deep Learning Algorithms

With the growing attention of the global population towards renewable energy sources such as wind power derived from natural resources such as Solar, Geothermal, and Wind energy, there is a growing need for genuine estimation of wind energy. Wind energy is important in the supply of electricity in the global energy markets because of its enormous importance in the delivery of renewable energy. As such, it is crucial for accurate appreciation of the power of the wind to ably respond to issues that relate to the trading of power while at the same time addressing issues to do with planning, scheduling and strategic positioning of wind power generation. Therefore, the present work aims to develop a new model known as the Association Rule with DL-based Wind Power Generation and Price Prediction (ARDL-WPGPP). It utilizes two datasets: An Energy dataset includes various columns like tie, wind onshore forecast, and price actual, whereas the weather features dataset includes only wind speed. These datasets are combined to create a dataset where each transaction represents 2 hours of wind generation. A data mining approach is employed to uncover hidden patterns, rules, concepts, and correlations within these datasets, operating on various types of data including quantitative, textual, and multimedia formats. To efficiently extract rules from the dataset, an improved Apriori algorithm is introduced. Subsequently, the generated rules incorporating wind speed are passed into an improved LSTM model, which learns by comparing the label data. The price actual value serves as the label data, assigning labels to data points based on their actual price value. High price actual values indicate high wind power prices, while low values indicate low wind power prices.

Adaptive Minimum Support Threshold for Association Rule Mining

Adaptive Minimum Support Threshold for Association Rule Mining

Efficacy and Mechanism of Core Traditional Chinese Medicines for Treating Malignant Lymphoma based on Efficacy Studies: A Study Supported by Network Pharmacology and Molecular Docking.

The burden of malignant lymphoma in China is greater than the global equivalent. The randomized controlled trials provide medical evidence that TCM can improve the response and survival in patients with lymphoma. However, the mechanisms underlying remain undefined. Evidence-based data mining for traditional Chinese medicine (TCM) on improving response and survival in malignant lymphoma treatment was performed in this study. In addition, the mechanisms of TCM through network pharmacology and molecular docking were explored. The China national knowledge infrastructure, Wanfang Data, China Science and Technology Journal Database, PubMed, and Web of Science databases were searched to select TCM formulas with response and survival benefits in the treatment of malignant lymphomas. We then analyzed and visualized the tropism of taste, frequency of drug use, dosage, clustering, association rules mining (minimum support threshold as 0.20, the minimum confidence threshold as 0.80 and lift >1), and complex networks for potential core herb compositions using Excel, IBM SPSS Statistics 26, and IBM SPSS Modeler 18. TCM systems pharmacology, GeneCards, Online Mendelian Inheritance in Man, and other databases were used to screen potential core active ingredients and malignant lymphoma-related targets. The intersection targets were used to construct a protein interaction network using Cytoscape to obtain the key targets. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway enrichment were used to analyze the core target, and molecular docking of key components and targets was performed using CB-Dock2. Twenty-four Chinese herbal formulae were included, encompassing 107 herbs with mainly cold and warm properties and bitter and sweet flavors. They were associated with the yin meridians of the liver, spleen, and lungs. The TCMs underwent association rule analysis, identified 27 association rules, including 12 herb pairs and 13 angle medicine, and clustered into eight classes by clustering analysis. Combined with the results from mining analysis, Pinelliae (Ban-xia), Poria (Fu-ling), Atractylodis macrocephalae (Bai-zhu), Curcumae (E-zhu), and Sparganii (San-leng) were the potential core herbs According to network pharmacology and molecular docking, the main core components of the potential core drugs are hederagenin, cerevisterol, 14- acetyl-12-senecioyl-2E,8E,10E-atractylentriol, 12,13-epoxy-9-hydroxynonadeca-7,10-dienoic acid, cavidine, and baicalein. These core drugs are mainly involved in the pathways of EGFR tyrosine kinase inhibitor resistance, PD-1/L1, natural killer cell-mediated cytotoxicity, NF-κB, epithelial cell signaling in H. pylori infections, and Th17 cell differentiation. They aid in regulating the transmembrane receptor protein tyrosine kinase signaling pathway, ERBB signaling pathway, PI3K signaling pathway, and phosphorylation process. Ten key components and eight key targets, including baicalein and hederagenin, demonstrated strong binding activity. Collectively, some core herbs exerted anti-tumor effects through immune and inflammatory pathway modulation, inhibition of immune escape, and induction of cell apoptosis. These findings support future evidence-based research on malignant lymphoma treatment using TCM.

Big data technology for teaching quality monitoring and improvement in higher education - joint K-means clustering algorithm and Apriori algorithm

With the development of big data technology, the field of monitoring and improving teaching quality in universities has ushered in new opportunities and challenges. Big data technology enables the capture and analysis of massive amounts of data generated during the teaching process, providing the possibility for a deeper understanding of teaching activities. However, how to extract useful information from these vast amounts of data and transform it into strategies for teaching improvement is a challenge. The research aims to propose a teaching quality monitoring and improvement method based on big data technology, which combines K-means clustering algorithm and association rule mining algorithm to improve the accuracy of teaching monitoring and the effectiveness of teaching improvement. In order to cope with these challenges, the study proposes a research method of big data technology based on joint K-mean clustering algorithm and association rule mining algorithm. The study first analyzes the teaching quality monitoring and evaluation indexes using the K-mean algorithm. Then the association rule mining algorithm is utilized to mine the data in the teaching quality monitoring indicators with association rules on the basis of the obtained cluster analysis. Finally, on the basis of association rule mining, the study constructs the assessment model of teaching quality monitoring indicators by utilizing the fused method. The outcomes revealed that the average of data analysis accuracy and the average of recall rate of the modeling method were 93.79 % and 91.95 %, respectively. Meanwhile, the evaluation time of the modeling method in the process of teaching quality monitoring data processing was 17.3 s, and the evaluation precision was 93.15 % respectively. Additionally, the process's overall confidence and enhancement are 95.01 % and 86.73 %, respectively, and the modeling method's performance is compared to other approaches. This shown that the approach may significantly boost the precision and effectiveness of monitoring the quality of instruction, as well as offer strong backing for the enhancement of instruction in higher education institutions.

Predicting adverse birth outcome among childbearing women in Sub-Saharan Africa: employing innovative machine learning techniques

BackgroundAdverse birth outcomes, including preterm birth, low birth weight, and stillbirth, remain a major global health challenge, particularly in developing regions. Understanding the possible risk factors is crucial for designing effective interventions for birth outcomes. Accordingly, this study aimed to develop a predictive model for adverse birth outcomes among childbearing women in Sub-Saharan Africa using advanced machine learning techniques. Additionally, this study aimed to employ a novel data science interpretability techniques to identify the key risk factors and quantify the impact of each feature on the model prediction.MethodsThe study population involved women of childbearing age from 26 Sub-Saharan African countries who had given birth within five years before the data collection, totaling 139,659 participants. Our data source was a recent Demographic Health Survey (DHS). We utilized various data balancing techniques. Ten advanced machine learning algorithms were employed, with the dataset split into 80% training and 20% testing sets. Model evaluation was conducted using various performance metrics, along with hyperparameter optimization. Association rule mining and SHAP analysis were employed to enhance model interpretability.ResultsBased on our findings, about 28.59% (95% CI: 28.36, 28.83) of childbearing women in Sub-Saharan Africa experienced adverse birth outcomes. After repeated experimentation and evaluation, the random forest model emerged as the top-performing machine learning algorithm, with an AUC of 0.95 and an accuracy of 88.0%. The key risk factors identified were home deliveries, lack of prenatal iron supplementation, fewer than four antenatal care (ANC) visits, short and long delivery intervals, unwanted pregnancy, primiparous mothers, and geographic location in the West African region.ConclusionThe region continues to face persistent adverse birth outcomes, emphasizing the urgent need for increased attention and action. Encouragingly, advanced machine learning methods, particularly the random forest algorithm, have uncovered crucial insights that can guide targeted actions. Specifically, the analysis identifies risky groups, including first-time mothers, women with short or long birth intervals, and those with unwanted pregnancies. To address the needs of these high-risk women, the researchers recommend immediately providing iron supplements, scheduling comprehensive prenatal care, and strongly encouraging facility-based deliveries or skilled birth attendance.

The association of road, traffic, and environmental factors with wrong-way riding behavior on shared e-bikes

Wrong-way riding (WWR) behavior of electric bikes (e-bikes) is a hazardous practice that frequently threatens traffic safety. It is imperative to accurately detect instances of WWR and analyze the various factors contributing to this behavior to develop and implement effective management policies. The emergence of shared e-bikes and advancements in machine learning have opened up new avenues for accurately examining the determinants of WWR behavior. Therefore, this study utilizes a large-scale dataset of shared e-bike trajectory data to establish a framework for detecting WWR behavior. Subsequently, association rule mining investigates the correlation among road, traffic, environmental factors, and the likelihood of higher and lower WWR behavior. The results indicate that: (1) In the association rules for higher WWR frequency, the absence of central separation facilities and bike lane separation facilities on the tertiary road, as well as lower land use density and the lack of public transportation facilities, enhanced the likelihood of WWR occurrence; (2) In the association rules for lower WWR frequency, longer crossing facility distance, physical bike lane separation, along with higher land use mixing and the presence of public transportation facilities, reduce the probability of WWR occurrence; (3) Time conditions display spatial heterogeneity, and different land use and road condition modulate the impact of time factors on rider’ intention for WWR. Finally, based on the study findings, adequate policy recommendations are proposed for urban management agencies to reduce WWR behavior and enhance the safety of the e-bike traffic system.

A fourfold-objective-based cloud privacy preservation model with proposed association rule hiding and deep learning assisted optimal key generation

ABSTRACT Numerous studies have been conducted in an attempt to preserve cloud privacy, yet the majority of cutting-edge solutions fall short when it comes to handling sensitive data. This research proposes a “privacy preservation model in the cloud environment”. The four stages of recommended security preservation methodology are “identification of sensitive data, generation of an optimal tuned key, suggested data sanitization, and data restoration”. Initially, owner’s data enters the Sensitive data identification process. The sensitive information in the input (owner’s data) is identified via Augmented Dynamic Itemset Counting (ADIC) based Associative Rule Mining Model. Subsequently, the identified sensitive data are sanitized via the newly created tuned key. The generated tuned key is formulated with new fourfold objective-hybrid optimization approach-based deep learning approach. The optimally tuned key is generated with LSTM on the basis of fourfold objectives and the new hybrid MUAOA. The created keys, as well as generated sensitive rules, are fed into the deep learning model. The MUAOA technique is a conceptual blend of standard AOA and CMBO, respectively. As a result, unauthorized people will be unable to access information. Finally, comparative evaluation is undergone and proposed LSTM+MUAOA has achieved higher values on privacy about 5.21 compared to other existing models.

Analyzing the Relationships of Psychological Symptoms in the SCL-90 Using Genetic Algorithms and Association Rule Mining

Analyzing the Relationships of Psychological Symptoms in the SCL-90 Using Genetic Algorithms and Association Rule Mining

Interactions between patterns of multimorbidity and functional status among hospitalized older patients: a novel approach using cluster analysis and association rule mining

BackgroundMultimorbidity (MM) is generally defined as the presence of 2 or more chronic diseases in the same patient and seems to be frequently associated with frailty and poor quality of life. However, the complex interplay between MM and functional status in hospitalized older patients has not been fully elucidated so far. Here, we implemented a 2-step approach, combining cluster analysis and association rule mining to explore how patterns of MM and disease associations change as a function of disability.MethodsThis retrospective cohort study included 3366 hospitalized older patients discharged from acute care units of Ancona and Cosenza sites of Italian National Institute on Aging (INRCA-IRCCS) between 2011 and 2017. Cluster analysis and association rule mining (ARM) were used to explore patterns of MM and disease associations in the whole population and after stratifying by dependency in activities of daily living (ADL) at discharge. Sensitivity analyses in men and women were conducted to test for robustness of study findings.ResultsOut of 3366 included patients, 78% were multimorbid. According to functional status, 22.2% of patients had no disability in ADL (functionally independent group), 22.7% had 1 ADL dependency (mildly dependent group), and 57.4% 2 or more ADL impaired (moderately-severely dependent group). Two main MM clusters were identified in the whole general population and in single ADL groups. ARM revealed interesting within-cluster disease associations, characterized by high lift and confidence. Specifically, in the functionally independent group, the most significant ones involved atrial fibrillation (AF)-anemia and chronic kidney disease (CKD) (lift = 2.32), followed by coronary artery disease (CAD)-AF and heart failure (HF) (lift = 2.29); in patients with moderate-severe ADL disability, the most significant ARM involved CAD-HF and AF (lift = 1.97), thyroid dysfunction and AF (lift = 1.75), cerebrovascular disease (CVD)-CAD and AF (lift = 1.55), and hypertension-anemia and CKD (lift = 1.43).ConclusionsHospitalized older patients have high rates of MM and functional impairment. Combining cluster analysis to ARM may assist physicians in discovering unexpected disease associations in patients with different ADL status. This could be relevant in the view of individuating personalized diagnostic and therapeutic approaches, according to the modern principles of precision medicine.

Uncovering pedestrian midblock crash severity patterns using association rules mining

The current study investigated the contributing factors and temporal variation in pedestrian crashes at midblock, with a particular focus on the severity levels: fatal/severe, moderate injury, and minor/no injury. It used association rules mining to uncover patterns between crash-contributing factors. By generating, evaluating, and visualising association rules for each severity level within each cluster, significant findings were discovered. Significant associations are observed between fatal crashes on weekdays and factors such as alcohol or drug impairment and nighttime. Similarly, factors including one-way roadway type, summer, and 25 MPH posted speed limit have a strong association with moderate injury crashes during weekdays. On weekends, nighttime crashes with non-motorised vehicles have the strongest association with fatal/severe injury crashes. Moreover, the generated rules for nighttime pedestrian fatal/severe injury crashes highlighted physical or drug-impaired pedestrians as the predominant attribute. The findings can enhance pedestrian safety at midblocks through targeted interventions.

English Network Teaching Model and Design of Evaluation System Based on Association Rule Algorithm

This study innovates English network teaching by applying a refined Association Rule Mining (ARM) algorithm. It integrates an “interest” parameter into ARM, dynamically adapting content to individual learners' profiles, improving engagement and outcomes. Controlled experiments, spanning diverse online platforms, validate the ARM model's efficacy by correlating learning content with academic performance, specifically CET-4 and CET-6 scores. Comprehensive preprocessing ensures data quality and privacy, employing techniques like de-identification, data perturbation, and aggregation. Advanced data analysis, including cross-validation and multivariate techniques, bolsters findings' reliability. Results highlight the ARM model's capacity to generate personalized learning paths, transcending conventional methods, and its potential as a cornerstone for data-driven education reforms. Future research will explore machine learning refinements and cultural adaptability to broaden its impact, fostering equitable, high-quality digital English education worldwide.

Identifying core IoT technologies using ARM and FCM: A comprehensive data-driven method

The internet of things (IoT) technology has garnered significant attention in recent years due to its wide-ranging applications. IoT, with its high connectivity capabilities, integrates various industrial, domestic, and agricultural devices into a smart and remotely controllable software and hardware platform. The field of IoT technology is expansive and encompasses a multitude of sub-technologies. Identifying core technologies in this domain is crucial for guiding research and development efforts by companies. Given the interrelation of these core technologies and their combination with recent decision-making approaches, network-based strategies have recently received special attention. The developed methods are based on static conditions and the assumption of stability, while in emerging technologies like IoT, the pace of changes over time is high. This leads to changes in the importance of technologies under various scenarios.In this study, in order to analyze the extracted patent data, association rule mining (ARM) algorithms were applied to identify the relationships between technologies and social network analysis was used to analyze the relationships between technologies and estimate their initial weights. Finally, fuzzy cognitive map (FCM) were used to estimate the final weights of technologies and rank them. The fcm approach allows for simultaneous modeling of both static and dynamic states of the system and, on the other hand, by calculating under various scenarios, suggests a core technology that is sustainable.The research results show that digital information transmission technologies, digital or electrical data processing, and wireless communication networks are the most important sub-technologies of Internet of things.

Forecasting Customer Actions: Exploring Machine Learning Techniques for Behavior Analysis and Prediction

Modern modelling of today’s consumer behaviour is highly dependent on data mining which mines customers’ data in order to answer specific questions in time. This complexity and uncertainty relating to the consumer be-haviour makes it difficult for us to predict. Therefore, selection of appropri-ate methods is very important. Predictive models can be designed so as to identify specific group or individual actions and thus making them useful marketing tools. However, many models are overly simplistic for they leave out some of the essential factors hence leading to inaccurate forecasting out-comes. Consequently, instead of concentrating on customer behaviour-firm capital structure relationships only a strong association rule mining model should be created using online store data that effectively estimates customer response. Knowing their business clients is crucial when it comes to profitability by aligning the company’s earnings with their concerns. Artificial intelligence (AI) optimizes product location by clustering techniques targeting the right customers. It examines customer behaviour and buying trends utilizing Kaggle’s client membership card records such as Customer ID, age, gender, annual income, spending score among others. This includes basic market analytics based on demographic information like age or income split into groups, however big data superiority over traditional methods cannot be underestimated including machine learning applications such as Customer Seg-mentation Automated Methodology (CSAM), k-means clustering for user segmentation among others. This approach facilitates focused advertising campaigns and development and launch of new products aimed at particular consumer sections.

Extended Association Rule Mining and Its Application to Software Engineering Data Sets

Extended Association Rule Mining and Its Application to Software Engineering Data Sets