Data Mining Methods Research Articles

Exploring Risk Factors of Type 2 Diabetes Mellitus Using Decision Tree and Random Forest Models: Baseline Data From Kharameh Cohort Study

Background and aims: Identifying subjects that are at risk of type 2 diabetes mellitus (T2DM) and predicting the associated risk factors are highly important. Thus, this study aimed to explore the risk factors and find the prediction model for T2DM using decision trees (DTs) and random forest (RF) models. Methods: This cross-sectional study is a part of the Kharameh Cohort Study. Kharameh Cohort is a part of the Fars Cohort, which started in 2014 with 10663 people aged 40–70. In this study, the risk factors of T2DM were explored using two data mining methods. Accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were applied to evaluate the models. The data were statistically analyzed using R software. Results: The DT modeling showed that age, triglycerides (TG), physical activity, systolic blood pressure, low-density lipoproteins (LDL), and body mass index (BMI) were the most associated factors in D2MT, while applying RF revealed that fasting blood sugar, cholesterol, creatinine, TG, gamma-glutamyl transferase physical activity, BMI, and LDL were the most effective on T2DM. The RF model was superior to the DT based on the applied criteria. Sensitivity, specificity, accuracy, and AUC for the RF were 73.4, 70.10, 73.5, and 79.1. These findings for the DT were 63.8, 69.7, 62.8, and 66.8, respectively. Conclusion: Based on the inferences, a strong association was found between several risk factors and the risk of T2DM. Therefore, predictive analytics using the RF model can be applied to identify the risk factors of other chronic diseases.

Radiomics-Based Diagnosis in Dentomaxillofacial Radiology: A Systematic Review.

Radiomics is a quantitative tool for digital image analysis. This systematic review aims to investigate the scientific articles to evaluate the potential implications of Radiomics analysis in Dentomaxillofacial Radiology (DMFR). Studies regarding Radiomics applications in DMFR and human samples, in vivo study, a case reports/series if ≧5 samples were included, while case reports/series if < 5 samples, articles other than in English, abstracts without full text, and studies published before 2015 were excluded. Fifty-one articles were selected from 3789 literatures. The QUADAS-2 tool was used for risk of bias assessment. The accuracy of predicting dentomaxillofacial pathologies was considered as the primary outcome, and the modeling type of Radiomics was considered as the secondary outcome. A meta-analysis could not be performed due to the lack of information and standardization among the reported accuracies. The reported accuracies were found between 0.66 and 99.65%. Logistic regression (n = 6) was found to be the most common Radiomics modeling type, followed by Support Vector Machine and Decision Tree (n = 5). Second-order statistics (n = 38) was the most common type of Radiomics application, followed by first-order (n = 26), higher-order (n = 20), and shape-based (n = 15) statistics. Further work is needed to increase standardization in the Radiomics workflow. Quantitative image analysis is an alternative tool for conventional visual radiographic evaluation. Radiomics systems depend on elements such as imaging modality, feature type, data mining, or statistical method. Radiomics applications do not justify digital transformation on their own, but the potential of its integration into the digital workflow is considerable.

Predicting Student Dropout Rates in Higher Education: A Comparative Study of Machine Learning Algorithms

Recent years have seen a significant increase in study interest in the areas of predicting student performance, avoiding failure, and identifying the variables affecting student dropout. One important indicator in online and open distance learning courses is the student dropout rate. We purpose the naive bayes classification method to construct the student dropout prediction using naive bayes. This work examines the critical topic of forecasting student dropout rates in higher education using machine learning approaches, with a particular emphasis on the random forest algorithm and the naive bayes algorithm. The study's goal is to properly anticipate dropout rates using data mining methods and machine learning algorithms after conducting a thorough evaluation of existing literature and approaches. The systematic method consists of data collection from a Kaggle dataset, data preparation to solve class imbalance via SMOTE oversampling, and algorithm selection. Random forest and naive Bayes approaches outperform other machine learning algorithms in terms of accuracy, sensitivity, specificity, and precision. The study underscores the importance of considering diverse factors such as demographic data, socioeconomic factors, and academic performance in dropout prediction models. The implications of this research extend beyond academia, with the potential to inform proactive interventions and support systems, ultimately leading to improved student outcomes and institutional effectiveness. According to this paper, the paper outputs that for the binary classification on the data set used in this project has best performed with Naive Bayes and Random Forest Algorithm with SMOTE oversampling. Keywords- SMOTE oversampling, machine learning, Random forest, naive bayes.

Voice or text? The role of physician media choice on patient experience in online medical communities

AbstractOnline medical communities (OMCs) are a type of online healthcare, in which physician‐patient interaction can be comprised of a variety of media options such as pictures, text, and voice. These media formats are often used to create a personalized patient experience in AI‐driven conversational healthcare platforms. To explore how physician media usage affects patient experience, we propose a counterfactual causal inference model using AI‐driven data mining methods on 131,083 online consultation records and 7,666,111 messages sent by physicians from one of the largest OMCs in China. Our study reveals the negative impact of physician use of voice on patient experience, compared to text. Drawing upon social support theory, we identify the mechanism by which physician media usage for voice produces a negative effect. The findings indicate that the negative effect of physicians' voice‐media usage occurs mainly in low‐risk disease conditions, by weakening the role of professional and emotional support. In contrast, in high‐risk disease conditions, voice‐media usage strengthens the role of professional and emotional support in improving the patient's experience. Our study is one of the first to focus on the social support attributes of the different media formats used in OMCs. We use advanced AI text‐analysis algorithms to extract features related to social support in physician‐patient conversations, and thus contribute to the use of AI in feature extraction for research.

Investigation of the Factors Affecting Teachers’ Attitudes towards E-Portfolio Using Data Mining Methods

This current study aims to determine the factors affecting teachers' attitudes towards e-portfolios by using data mining methods such as the Classification and Regression Tree (CART) algorithm and Random Forest (RF) algorithm. The study group consists of 449 participants willing to participate on a volunteer basis. In the study, the data were collected using the "Scale of Attitude towards E-Portfolio", the "Reflective Thinking Tendency Scale", and the "Scale of Attitude towards Technology". The survey research design, one of the quantitative research approaches, was used in the study. The collected data were analyzed by using the CART and RF analysis methods, two of the data mining methods. It was concluded with the CART method that the variable with the highest predictor importance in predicting the level of (low-high) attitude towards electronic portfolio is the variable of attitude towards technology, while the variable of how teachers see themselves in using technology is with the lowest predictor importance. It was concluded that with the RF model the variable with the highest predictor importance in predicting the level of attitude (low-high) towards electronic portfolio is the variable of attitude towards technology, followed by the variable of reflective thinking tendency, while the variable with the lowest predictor importance is the variable of the type of the graduated university (state-foundation). The results show that the RF method better determines the predictor variables.

Research on BP neural network model-based data mining technique in KOL identification

Abstract. Accompanied by the development of artificial intelligence industry, to play the data value of web crawler technology has been the focus of research in the field of computer network and data science, and the data mining technology based on artificial neural network intelligent algorithm is widely used. In view of this, this paper takes the BP neural network data mining technology, which has excellent nonlinear mapping capability, parallel processing capability and fault tolerance and is widely used, as the basis, and integrates the methods and ideas of data mining into the mining of data laws in the field of KOL identification of key opinion leaders, with a view to finding valuable intrinsic laws and relationships between the mining of web crawler technology and the identification of KOL features. The research content of this paper mainly includes two aspects of research work, the design of high-performance data mining technology and the actual work in the field of KOL recognition. On the one hand, this paper comprehensively describes the basic theory and methods of data mining, and focuses on the in-depth analysis and elaboration of BP neural network-based data mining technology on the basis of understanding and analyzing a variety of data mining technologies. On the other hand, this paper aims to solve the problem of bias in the prediction of traditional KOL model and design the experimental method of KOL recognition by BP neural network algorithm.

Enhancing the accuracy of blood-glucose tests by upgrading FTIR with multiple-reflections, quantum cascade laser, two-dimensional correlation spectroscopy and machine learning

The accuracy of screening diabetes from non-diabetes is drastically enhanced by strategically upgrading the bench-marking infrared spectroscopy technique for non-invasive tests of blood-glucose, both with state-of-the-art instrumentation-retrofits and with intelligent spectral-datamining tools. First, the signal-to-noise performance of FTIR in measuring the spectral features of a glucose solution containing bovine serum albumin is improved by 2–3 times with the common single-pass attenuated total-reflection setup replaced by a multi-passes-reflections setup. Second, replacing the ordinary infrared lamp with a quantum cascade laser further improves the signal-to-noise by 3 times. The performance of the upgraded spectrometer in screening hyperglycemia is gauged by the accuracy of such screens derived from 100 repetitive spectral-measurements per glucose concentration, for 24 glucose concentrations spanning the range of 70–300 mg/dL, a range which covers the blood-glucose contents of all non-diabetic and diabetic human-subjects. Third, intelligent datamining methods are exploited to implement decision trees for screening hyperglycemia. Their decisions are mapped into a confusion matrix and the matrix-elements are used to calculate the accuracy merits of each method. Evidently, the accuracy of the multi-passes-FTIR with the standard principal-components datamining method is 80 %. The adoptions of the quantum cascade laser and two-dimensional correlation spectroscopy datamining technique raises this to 96.3 %. Finally, a novel machine learning method, which comprises three different decision-tree tools to generate trial screening decisions and a “majority-voting” datamining tool to reach a final screening decision, yields the best accuracy of 98.8 % ever reported in the literature.

Research and performance analysis of random forest-based feature selection algorithm in sports effectiveness evaluation

The rapid progress in fields such as data mining and machine learning, as well as the explosive growth of sports big data, have posed new challenges to the research of sports big data. Most of the available sports data mining techniques concentrates on extracting and constructing effective features for basic sports data, which cannot be achieved simply by using data statistics. Especially in the targeted mining of sports data, traditional mining techniques still have shortcomings such as low classification accuracy and insufficient refinement. In order to solve the problem of low accuracy in traditional mining methods, the study combines the random forest algorithm with the artificial raindrop algorithm, and adopts a sports data mining method based on feature selection to achieve effective analysis of sports big data. This study is based on the evaluation method of motion effects using random forests, and uses feature extraction algorithms to study the motion effect impacts. It uses the information gain index to rank the importance of features and accurately gain the degree of influence of exercise on various indicators of the human body. Through simulation verification, the algorithm proposed by the research institute performs the best in accuracy and FI scores on the training and testing sets, with accuracies of 0.849 ± 0.021 and 0.819 ± 0.022, respectively, and F1 scores of 0.837 ± 0.020 and 0.864 ± 0.021, respectively. This indicates that the algorithm proposed by the research institute has high classification accuracy and performance proves that the Random Forest-based feature selection algorithm established in this study is superior to the existing traditional feature extraction and extraction methods in terms of both performance and accuracy. The proposal of this data analysis method has achieved accurate and efficient utilization of sports big data, which is of great significance for the development of the sports education industry.

Lane Change Behavior Patterns and Risk Analysis in Expressway Weaving Areas: Unsupervised Data-Mining Method

Lane Change Behavior Patterns and Risk Analysis in Expressway Weaving Areas: Unsupervised Data-Mining Method

Rapid Convolutional Algorithm for the Discovery of Blueberry Honey Authenticity Markers via Nontargeted LC-MS Analysis.

Bees produce honey through the collection and transformation of nectar, whose botanical origin impacts the taste, nutritional value, and, therefore, the market price of the resulting honey. This phenomenon has led some to mislabel their honey so that it can be sold at a higher price. Metabolomics has been gaining popularity in food authentication, but rapid data mining algorithms are needed to facilitate the discovery of new authenticity markers. A nontargeted high-resolution liquid chromatography-mass spectrometry (HR/LC-MS) analysis of 262 monofloral honey samples, of which 50 were blueberry honey, was performed. Data mining methods were demonstrated for the discovery of binary single-markers (compound was only detected in blueberry honey), threshold single-markers (compound had the highest concentration in blueberry honey), and interval ratio-markers (the ratio of two compounds was within a unique interval in blueberry honey). A novel convolutional algorithm was developed for the discovery of interval ratio-markers, which trained 14× faster and achieved a 0.2 Matthews correlation coefficient (MCC) units higher classification score than existing open-source implementations. The convolutional algorithm also had classification performance similar to that of a brute-force search but trained 1521× faster. A pipeline for shortlisting candidate authenticity markers from the LC-MS spectra that may be suitable for chemical structure identification was also demonstrated and led to the identification of niacin as a blueberry honey threshold single-marker. This work demonstrates an end-to-end approach to mine the honey metabolome for novel authenticity markers and can readily be applied to other types of food and analytical chemistry instruments.

Optimizing Test Data Management Strategies in Banking Domain Projects

The steady rise in processing power over the past 20 years has resulted in an enormous volume of data. Furthermore, anybody may now easily create and consume material in any format thanks to recent advancements in Web technology. Large volumes of data are regularly gathered by banking systems, including trade finance data, SWIFT and telex communications, client information, details about transactions, risk profiles, credit card details, limit and collateral details, the compliance or Anti Money Laundering (AML)-related data, and limit and collateral details. Every day, thousands of choices are made at banks. These choices pertain to credit, default, beginning a relationship, investments, AML, and illicit funding, among other things. To make these crucial choices, one must rely on a variety of data and drill down capabilities offered by the banking systems. We created a set of specifications for the kinds of data that should be included in a product catalogue. We ascertained what data the departments need by using a survey and questionnaire of the retailer's staff. We ensured that there was no one standard for the information organisation and then put out our own plan. This enormous amount of data may be mined for information and intriguing patterns, which can then be used to the decision-making process. This article examines and summarises a number of data mining methods that have applications in the banking industry. An overview of data mining methods and procedures is given. It also sheds light on how these methods may be applied in the banking industry to facilitate and enhance decision-making.

General Industrial Process Optimization Method to Leverage Machine Learning Applied to Injection Moulding

ABSTRACTThe development of machine learning (ML) technologies is driving changes across many sectors. In industrial settings, this is called the fourth industrial revolution and encompasses several technologies pushing the boundaries of industrial automation. In this study, a general industrial process optimization (GIPO) methodology is formulated in the context of Industry 4.0 and tested on an industrial injection moulding machine (IMM). GIPO aims to encourage the practical inclusion of industrial artificial intelligence at all levels of the manufacturing process, while enabling industrial equipment to adapt to a changing processing environment. Special attention is given to the generality of the methodology so that it can be extended to other applications. In the example case study presented here, GIPO combines K‐nearest neighbours classification and nearest neighbours optimization methods to optimize an injection moulding process effectively. Practical implementation conducted on the IMM demonstrates a novel methodology to leverage data mining and ML methods in a real‐world setting to improve production quality, production time and energy cost.

Application of probabilistic and stochastic models and data mining for forecasting the contingent of old age pension recipients in the context of systemic uncertainty

The object of the research is mathematical models for forecasting the contingent of pension recipients in conditions of uncertainty caused by both the reform of the pension system and the impact of armed aggression. Based on the study of statistical information on the structure and dynamics of the contingent of pension recipients, an approach to uncovering systemic uncertainty in the task of forecasting the contingent of pensioners is proposed. This work is part of a study of the application of data mining methods of intellectual data analysis and mathematical modeling in information technology intended for use in the pension system. The main focus of this work is on forecasting the dynamics of the contingent of pension recipients by age, in particular, forecasting the number of newly appointed pensions. The difficulty of forecasting the contingent of pension recipients, in particular by age, is connected with the fact that it is necessary to ensure the representativeness and variability of data sets. In addition, it should be taken into account that a significant number of factors must be included in the model in accordance with the requirements of regulatory documents. Another problematic issue is that the time series of the investigated indicators, such as data on the insurance experience of insured persons (based on the results of a sample survey), may contain significant (more than 40 %) gaps that can be filled only on the basis of primary (paper) documents. Therefore, the input data sets are formed with assumptions about the probability of the accumulation of insurance experience in various groups of insured persons. The paper proposes an analytical toolkit based on the use of probabilistic and statistical models in the form of Bayesian networks, intended for use in specialized decision-making support systems of the Ukrainian pension system. In the course of the study, a number of numerical experiments were carried out, in which the correctness of the proposed method was investigated. The proposals presented in the paper will improve the stability of the pension system of Ukraine, including through a more accurate definition of the dynamics of the contingent of pension recipients and, accordingly, the costs of pension payments. The proposed models and methods can be used as part of decision-making support systems of state and public administration bodies to analyze the results of reforming the pension system.

DEVELOPING A WEB BROWSER EXTENSION TO PREVENT THE SPREAD OF FAKE NEWS

Fake news causes huge social problems such as misdirecting and provoking the masses, creating an atmosphere of chaos by spreading fear. Detecting and stopping the dissemination of fake news has been an important and priority issue due to its rapid spread, difficulty to detect, and negative effects. In our study, a new Chrome extension that detects fake news has been developed in order to detect and prevent the spread of fake news. In the study; natural language processing, data mining methods, and various machine learning techniques for instance Passive Aggressive, Random Forest, Support Vector Machine, AdaBoost, XGBoost, and Long-Short-Term Memory (LSTM) algorithms are used. The accuracy rates of the algorithms, confusion matrices, AUC rates, and ROC curves were compared. The created machine learning model has been implemented in the online internet environment by using Flask and Rest API in the Python program. Finally, the Chrome extension interface was built using Javascript, HTML, and CSS. The LSTM algorithm gave the highest prediction result with a rate of 90.72% compared to other algorithms.

A systematic review of lithium biology and pharmacology and toxicological evaluation of new organic lithium salts

Objective: to systematize scientific data on biomedical studies investigating trace element lithium over the past 70 years; evaluate toxic properties of lithium ascorbate (LiAsc) as an important promising candidate molecule.Material and methods. An analysis of 49,959 publications on lithium biomedical research retrieved from PubMed/MEDLINE database was carried out using modern data mining methods developed within the framework of topological approach to recognizing (Yu.I. Zhuravlev scientific school). Publications found by experts and not indexed in PubMed/MEDLINE were used in discussing the results of a systematic analysis of publications array retrieved from PubMed/MEDLINE. An experimental study of chronic 180 day-long LiAsc (at doses of 5, 50 and 150 mg/kg) toxicity was performed on 36 “Soviet chinchilla” rabbits by assessing local irritant action. Intoxication clinical picture, body weight dynamics, water and food intake as well as physiological, hematological and biochemical parameters were analyzed.Results. Classification and systematization of all currently available publications on lithium biology and medicine were performed. It was shown that pharmacological applications of lithium salts in mental disorders as well as lithium effects on simple sugars metabolism, lipid metabolism, blood pressure regulation, hematopoiesis, inflammation and tumor growth inhibition, neurotransmitter homeostasis, neurotrophic and neuroprotective molecular mechanisms as well as homeostasis of other electrolytes comprised promising fields of lithium drug research. The prospects for using organic lithium salts, particularly LiAsc, for various therapeutic goals were also discussed. 180-day-long oral administration of LiAsc at doses of 5, 50, 150 mg/kg resulted in no macroscopic signs of local inflammatory reaction while examining its local irritant effect.Conclusion. The lithium-ion effect on neurotransmitters promotes neuroprotection and reduces a risk of addiction. The antihypertensive, antiatherosclerotic, antidiabetic, antitumor and neurotrophic effects related to organic lithium salts may be beneficial in various therapeutic applications.

Using K-NN Algorithm for Evaluating Feature Selection on High Dimensional Datasets

Data mining is the process of using statistics, mathematics, artificial intelligence and machine learning to identify problems that exist in data so as to produce useful information. Based on its function, data mining is grouped into description, estimation, classification, clustering, and association. K-NN is one of the best data mining methods and is widely used in research. K-NN algorithm was introduced by Fix and Hodges in 1951. K-NN algorithm is a simple algorithm and is often used to cluster supervised data. Feature selection attribute selection is a data mining technique used in the pre-processing stage. This technique works by reducing complex attributes that will be managed at the processing and analysis stage. In this study, the most effective feature selection to improve the accuracy of the K-NN algorithm by increasing accuracy by 95.12% on the breast cancer dataset and 88.75% on the prostate cancer dataset.

Multi-fidelity data mining-based design optimization framework for scramjet-based aircraft

Aircraft design faces complex multidisciplinary problems and often involves numerous design variables, making direct optimization a time-consuming task. With the development of big data and machine learning technologies, using data mining methods to assist optimization has become key to addressing the inefficiency in multidisciplinary design optimization (MDO). To address the high cost of data acquisition in current data mining, this paper proposes an aircraft design optimization framework based on multi-fidelity data mining, using a large amount of inexpensive cursory estimation data to execute the MDO process and form a multidisciplinary optimization database. Then, data mining methods are used to decouple disciplinary relationships, transforming multidisciplinary coupling problems into simpler multi-objective problems. Subsequently, under viscous CFD computation conditions, the EDGE-p-BFFD deformation method is used to fine-tune the obtained configurations, further enhancing multidisciplinary performance. Before both optimizations, data mining methods were used to analyze the impact of design variables on objective functions, forming design knowledge and completing variable selection. In the multidisciplinary optimization practice of airbreathing aircraft, it can be found that the proposed framework effectively achieves disciplinary decoupling and explores the relationship between shape and performance. In the multidisciplinary optimization of the waverider based on cursory estimations, particle swarm optimization increased the aircraft's cruising range by 8.74 %. In the multi-objective optimization of the lifting body based on viscous CFD calculations, data mining was used to transform the problem into several key objectives. The MOEA/D algorithm increased the range by 2.9 %, validating the effectiveness of the data mining-based MDO framework proposed in this paper.

A novel physics-guided spatial-temporal data mining method with external and internal causal attention for drilling risk evaluation

As drilling technology advances and operations extend into more complex geological environments, evaluating drilling risks has become increasingly complex, challenging the effectiveness of traditional methods. The novel physics-guided spatial-temporal data mining method that integrates external and internal causal attention mechanisms for drilling risk evaluation is proposed to address this issue. Firstly, a risk calibration method based on spatial-temporal sequence clustering is designed. This method dynamically calibrates drilling risks by mining subtle changes in sign data during drilling. Secondly, an expert experience extraction method based on the correlation of drilling risk signs and a fuzzy inference system is established. Kendall's tau is used to quantify the correlation between drilling risk signs. The fuzzy inference system is employed to convert fuzzy and difficult-to-quantify expert experience into computable and interpretable rules. In order to improve the flexibility and adaptability of the fuzzy inference system, an expert experience rules base is also constructed. Subsequently, a spatial-temporal data mining model integrating both external and internal causal attention mechanisms (STMIEICAM) is constructed. The external causal attention mechanism (ECAM) quantified the correlation between signs and risk. The internal causal attention mechanism (ICAM) improved the model's ability to capture and quantify the features of spatial-temporal sequences. Finally, the physical knowledge from the fuzzy inference system and well-site is embedded into the STMIEICAM model, forming a physics-guided spatial-temporal data mining model integrating both external and internal causal attention mechanisms (PG-STMIEICAM) that enables graded evaluation of drilling risks. The proposed method was applied to overflow risk evaluation in an oil field to validate its effectiveness. The results demonstrate that the method not only excels in uncovering hidden relationships within the data but also integrates expert knowledge, achieving accurate evaluation of drilling risks.

A hybrid load prediction method of office buildings based on physical simulation database and LightGBM algorithm

Building load prediction plays an important role in building energy savings and mechanical and electrical system optimization control. The dynamic energy consumption can be accurately calculated using the traditional physical energy simulation method that entails a complex setup and verification process due to the numerous input parameters required. It is also difficult to change the physical model once it has been determined. The data-mining method is fast in calculation and simple to use, but its prediction accuracy is limited by historical data quality. It is difficult to predict the load of new buildings without historical data. To solve these problems, this study proposes a hybrid building load prediction method for office buildings. The proposed method uses EnergyPlus to generate a building load database that includes than 25.14 million data cases, covering 35 types of building geometry and 2870 building examples. Based on the above database, the LightGBM algorithm was selected to extract feature variables that affect the load and build a load prediction model. The training results show that there are 24 key feature variables for office building load prediction. The hourly MAPE of the cooling load prediction model is 6.95 % and RMSE is 4.31 W/m2, and the hourly MAPE of the heating load prediction model is 7.09 % and RMSE is 11.64 W/m2 compared with EnergyPlus model. Two actual office buildings are selected as case studies to validate the model prediction accuracy. Results show that comparing predicted results with measured data, the hourly cooling load of the MAPE is 12.42 %. Coparing predicted results with actual heating load, daily MAPE is 7.97 %.

Analysis of medication rule of Traditional Chinese Medicine in treating depression based on data mining

ObjectiveDepression is a mental disorder characterized by persistent feelings of sadness, decreased interest or pleasure in activities and reduced energy. As a highly prevalent disorder, it seriously endangers the psychosocial functioning of patients. Many scholars have conducted clinical studies on the treatment of depression using different herbal remedies, but there are no studies that integrate these remedies to explore the general medication rule. This study aims to explore the medication pattern of Traditional Chinese Medicine (TCM) treatment for depression through data mining methods, so as to provide scientific theoretical basis and reference for clinical treatment and new prescription development. MethodsBased on the PRISMA principle, 121 articles involving 10810 patients with depression of TCM treatment were collected. We then performed frequency, association rule, and hierarchical clustering analysis of Chinese herbs using Microsoft Excel 2016, SPSS Modeler 18.0 and IBM SPSS Statistics 23. ResultsAmong the 270 herbs collected, the three most frequently occurring herbs are Gancao, Chaihu, and Shaoyao. The categories of high-frequency herbs are mainly deficiency-tonifying, Qi-regulating and blood-activating and stasis-eliminating herbs. Through the Apriori algorithm, we mined 21 herbal groups of association rules, and among which the combination of Chaihu-Shaoyao-Gancao has the highest level of support. Furthermore, five novel clustering combinations were identified, predominantly derived from Xiaoyao-San, Chaihu-Shugan-San, Sini powder, Kaixin-San and Chaihu-Jia-Longgu-Muli Decoction. ConclusionThe current study not only concluded the frequent combinations but also developed five new drug cluster combinations for depression, which can provide evidence-based references for the future clinical treatment and is helpful to understand the potential pharmaceutical mechanism from the properties, tastes, meridian tropisms and categories. The clinical effectiveness of these combinations needs to be verified by future study.