Multiple Data Mining Techniques Research Articles

Binary Classification of Diabetes in Pima Indian Dataset: A Deep Learning Perspective

Abstract: In this study, we develop into the application of deep learning methodologies for diabetes prediction utilizing the Pima Indian dataset. Employing Keras with Theano as the backend, we establish a binary classification model to effectively forecast the presence or absence of diabetes in individuals. Our research aims to enhance the precision and reliability of diabetes diagnosis, ultimately contributing to improved healthcare decision-making. Our investigation leverages Keras, a high-level neural networks API, in conjunction with Theano, to conduct binary classification on the Pima Indian diabetes dataset. Our study provides valuable insights into the field of medical data analysis, showcasing the effectiveness of deep learning techniques in advancing diagnostic tools for proactive healthcare management. Diabetes mellitus, a prevalent chronic disease globally, necessitates the development of a system for early type 2 diabetes mellitus (T2DM) diagnosis. Multiple machine learning and data mining techniques, including ANN, SVM, KNN, decision trees, and Extreme Learning Machines, have emerged and been employed as aids in diabetes detection. Consequently, we introduce Deep Learning, a subfield of machine learning, which can effectively handle smaller datasets through efficient data processing techniques. This paper presents an in-depth review of Diabetic Retinopathy, covering its features, causes, various ML models, DL models, challenges, comparisons, and future directions for early DR detection. Diabetes mellitus is a global health concern with a rapidly increasing prevalence. In this context, machine learning technologies prove invaluable for early disease identification and diagnosis. The focus of this study is to identify the most effective ML algorithm for diabetes prediction.

An integrated strategy for rapid discovery and identification of the potential effective fragments of polysaccharides from Saposhnikoviae Radix

Ethnopharmacological relevanceSaposhnikoviae Radix (SR) is a traditional Chinese medicine, known as “Fangfeng”. As one of the main active components, Saposhnikoviae Radix polysaccharides (SP) demonstrated a range of biological activities, especially immunity regulation activity. Aims of the studyThis study aimed at exploring whether polysaccharides have activity after degradation, then discovering the potential effective fragments of SP. Materials and methodsHere we establish the chromatographic fingerprints method for 32 batches of 1-phenyl-3-methyl-5-pyrazolone (PMP) derivatives of oligosaccharides by HPLC, meanwhile evaluating its immunomodulatory activity in vivo. Then, the potential effective fragments of SP were screened out based on the spectrum-effect relationship analysis between fingerprints and the pharmacological results. Besides, liquid chromatography ion trap-time of flight mass spectrometry (LC-IT-TOF MS) coupled with multiple data-mining techniques was used to identify the potential effective oligosaccharides. ResultsThese findings showed that the hydrolysate of SP have significant immunomodulatory, and the immunity regulation activity varies under different hydrolysis conditions. The 4 potential effective peaks of the hydrolysate of SP were mined by spectrum-effect relationship. Finally, the chemical structure of 4 potential effective oligosaccharide fragments of SP was elucidated based on LC-IT-TOF MS. F10 was inferred tentatively to be Hex1→6Hex1→6Hex1→6Hex1→6Hex1→6Gal; F18 was confirmed to be Rhamnose; F14 was inferred tentatively to be Hex1→4Hex1→ 4Hex1→4Gal and F25 was tentatively inferred to be Ara1→6Gal. ConclusionsThis study may provide a sound experimental foundation in the exploration of the active fragments from macromolecular components with relatively complex structures such as polysaccharides.

Locating terrane boundaries in South China with big geochemical data mining

The basis for understanding continental tectonics to a large degree relies on terrane recognition. However, discrepant identifications and interpretations of the same terrane as well as its boundary are still commonly seen in South China. For years, the central debate has been revolving around the Yangtze-Cathaysia boundary and West-East Cathaysia boundary. Here we characterized the geochemical patterns (or differences) of the proposed terranes and located their boundaries by using a big geochemical dataset of South China, which consists of compositions of 75 elements in 5231 sediment samples. To extract valuable information from this big geochemical dataset, we applied multiple data mining techniques, including principal component analysis, sequential binary partition, Sobel operator, and support vector machine. Our results show that the geochemical patterns vary distinctly at regional scale across both the subblocks (i.e., Yangtze and Cathaysia) and terranes (i.e., Jiuling, Huaiyu, and Youjiang). Accordingly, we conclude that 1) the Northeast Jiangxi fault marks not only the Jiuling-Huaiyu boundary but also the north-eastern Yangtze-Cathaysia boundary; 2) the Yangtze-Cathaysia boundary extends westward along the Jiangshan-Pingxiang-Chaling fault; and 3) a fault zone comprising the Northwest Fujian fault and Zhenghe-Dapu fault possibly demarcates the West and East Cathaysia. The boundaries defined by the contrasting geochemical patterns in this study are also supported by geophysical, Nd isotopic, and geological evidence. Our observations will allow better explanation of where the proposed terranes originated and how South China was assembled. Our work demonstrates that data mining techniques are effective in deriving and integrating information for solving geoscience problems.

A data driven performance assessment strategy for centralized chiller systems using data mining techniques and domain knowledge

Chillers are among the major energy consumers in building heating, ventilation and air conditioning systems and appropriate performance assessment of chiller systems is essential to ensuring their operational optimality while delivering satisfactory indoor thermal comfort. This paper presents a data driven performance assessment strategy for centralized chiller systems using multiple data mining and advanced visualization techniques. The energy consumption patterns of the chiller system were quantitatively and qualitatively analyzed by using the Conditional Inference Tree (CIT) and Agglomerative Hierarchical Clustering (AHC), and Association Rule Mining (ARM), respectively. A performance indicator of Coefficient of Performance (COP) Destruction (%) was introduced to represent the quality of the achieved COP. The performance of this strategy was evaluated using one-year operating data of a centralized chiller system installed in a commercial building. The results showed that the data mining techniques can be effectively used for performance assessment of chiller systems. The results from the quantitative and qualitative analysis showed that the chiller performance was strongly influenced by the temperature difference across the evaporator. The system studied generally showed good performance when the part load ratio was above 45% and the chiller power ratio was above 50%, and it showed relatively poor performance when the temperature difference across the evaporator was below 3.1 °C.

A data mining-based framework for the identification of daily electricity usage patterns and anomaly detection in building electricity consumption data

With the development of advanced information techniques, smart energy meters have made a considerable amount of real-time electricity consumption data available. These data provide a promising way to understand energy usage patterns and improve building energy management. However, previous studies have paid more attention to methodologies for the identification of energy usage patterns and are limited in the interpretability and applications of the patterns. In this context, this paper proposes a general data mining-based framework that can extract typical electricity load patterns (TELPs) and discover insightful information hidden in the patterns. The framework integrates multiple data mining techniques and mainly consists of three phases: data preparation, identification of TELPs and knowledge discovery in the patterns. A new clustering method with a two-step clustering analysis is proposed to identify the TELPs at the individual building level. Before clustering, five statistical features that represent the shapes of electricity load profiles are first defined to reduce the dimensions of daily electricity load profiles. The first clustering step aims at detecting outliers of daily electricity load profiles (DELPs) by using the density-based spatial clustering application with noise (DBSCAN) algorithm clustering technique, which addresses the data quality issues for electricity consumption data derived from energy consumption monitoring platforms (ECMPs). The second clustering step aims at grouping similar DELPs by means of the k-means algorithm to extract TELPs. The effectiveness of the proposed clustering method is demonstrated by a comparison with two single-step clustering techniques. Furthermore, a classification and regression tree (CART) algorithm is employed to discover insightful knowledge on TELPs and improve the interpretability of clustering results, namely, to explain the relations between dynamic influencing factors related to electricity consumption and TELPs. The proposed framework is applied to analyze the time-series electricity consumption data of three practical office buildings in Chongqing, and its effectiveness has been confirmed. A potential application of discovered knowledge is presented: early fault detection of anomalous electricity load profiles. The proposed framework can provide building managers with an efficient way to understand the characteristics of building electricity usage patterns and detect anomalies therein.

Analyzing Spatial Heterogeneity of Housing Prices Using Large Datasets

As an obstacle to the hedonic model’s reliability, housing submarkets have drawn plenty of scholarly attention because they lack an integrated and standardized classification framework and validation methods. By incorporating multiple spatial statistics and data mining techniques into a hybrid spatial data mining method, this study develops an innovative classification methodology that replaces spatial continuity with spatial connectivity. Employing Salt Lake County as the case, we identify 43 housing submarkets based on differentiation among structural differences, the complexity of urban space, and neighborhood characteristics. With the introduction of urban amenities into the validation framework, the comparison between the submarket-based model and non-submarket regression shows our classification not only enhances prediction accuracy but also achieves better theoretical comprehension of local housing markets. Besides contributing to an understanding of urban spatial heterogeneity, our study also provides a feasible spatial modeling method which is capable of processing a large dataset with more than 200,000 observations.

Metabolites identification of (+)-usnic acid in vivo by ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry

(+)-usnic acid (UA) is an active natural phenolic acid ingredient originating from Chinese traditional Tibetan herb. Usnea acid is expected to become a new agent for anticancer and remarkable antitumor. To reveal its metabolic profile, metabolites identification of UA in vivo was studied using ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UHPLC-Q-TOF-MS/MS) in this present study. The chromatographic separation was performed on a C18 column with a mobile phase consisted of methanol and water with a flow rate of 0.4 ml/min. The mass spectral analysis conducted in a negative electrospray ionization mode combined with information-dependent acquirement technology (IDA) was used to trace all the potential UA metabolites. Several sensitive and specific multiple data-mining techniques especially key product ions (KPIs) filter were applied to hunt and identify metabolites rapidly. As a result, a total of 36 metabolites were detected after oral administration of UA, including 33, 8 and 16 in rat urine, plasma and bile, respectively. These results showed that the probable metabolite pathways of UA were oxidation, reduction, dihydroxylation, glycine conjugation, glucuronide conjugation, N-acetylcysteine conjugation and methylation. It is the first time to elucidate the profile of UA in vivo. These results not only provided the basis of UA pharmacological properties, but also gave the guidance in clinical medication. Moreover, the analysis strategy and methodology proposed in this paper could be widely used in characterization of other phenolic acids metabolites.

An Effective Financial Statements Fraud Detection Model for the Sustainable Development of Financial Markets: Evidence from Taiwan

This study aims to establish a rigorous and effective model to detect enterprises’ financial statements fraud for the sustainable development of enterprises and financial markets. The research period is 2004–2014 and the sample is companies listed on either the Taiwan Stock Exchange or the Taipei Exchange, with a total of 160 companies (including 40 companies reporting financial statements fraud). This study adopts multiple data mining techniques. In the first stage, an artificial neural network (ANN) and a support vector machine (SVM) are deployed to screen out important variables. In the second stage, four types of decision trees (classification and regression tree (CART), chi-square automatic interaction detector (CHAID), C5.0, and quick unbiased efficient statistical tree (QUEST)) are constructed for classification. Both financial and non-financial variables are selected, in order to build a highly accurate model to detect fraudulent financial reporting. The empirical findings show that the variables screened with ANN and processed by CART (the ANN + CART model) yields the best classification results, with an accuracy of 90.83% in the detection of financial statements fraud.

Female breast shape categorization based on analysis of CAESAR 3D body scan data

In this study we explore the variation in female breast shape across the younger (age: 18–45), non-obese (BMI < 30) North American Caucasian population, a population that has not previously been well-represented in studies of breast shape. A method of classifying breast shape was developed based on multiple data-mining techniques. Forty-one relative measurements (i.e., ratios and angles) were constructed from 66 raw measurements (circumferences, depths, widths, etc.), extracted from 478 CAESAR (Civilian American and European Surface Anthropometry Resource) scans, using self-developed Matlab® programs. Seventy subjects were regarded as outliers and were removed. The remaining data were transformed and standardized to ensure robust analysis. To judge results, an algorithm was developed to visualize clustering outcomes in the form of side profiles of breasts. The results of three clustering methods, namely hierarchical, K-means, and K-medoids clustering, were compared. Finally, breast shapes were categorized into three and five groups by two different cluster number selection criteria proposed by the study: (1) based on misclassification rate; (2) based on the goodness-of-fit of the model. Several of the relative body measurements were identified to be critical in defining breast shape. The findings and the proposed methods of this study can contribute to the development of improved shape and sizing systems of bra products that work for both manufacturers and consumers. The new methodology developed in this study can also be applied to other types of intimate apparel products where an understanding of body shape plays a key role in body support, comfort, and fit.

Development of diagnostic model of lung cancer based on multiple tumor markers and data mining.

ObjectiveTo develop early intelligent discriminative model of lung cancer and evaluate the efficiency of diagnosis value.MethodsBased on the genetic polymorphism profile of CYP1A1-rs1048943, GSTM1, mEH-rs1051740, XRCC1-rs1799782 and XRCC1-rs25489 and the methylations of p16 and RASSF1A gene, and the length of telomere in the peripheral blood from 200 lung cancer patients and 200 health persons, the discriminative model was established through decision tree and ANN technique.ResultsACU of the discriminative model based on multiple tumour markers increased by about 10%; The accuracy rate of decision tree model and ANN model for testing set were 93.00% and 89.62% respectively. The ROC analysis showed the decision tree model’s AUC is 0.929 (0.894∼0.964), the ANN model’s AUC is 0.894 (0.853∼0.935). However, the classify accuracy rate and AUC of Fisher discriminatory analysis model are all about 0.7.ConclusionThe early intelligent discriminative model of lung cancer based on multiple tumor markers and data mining techniques has a higher accuracy rate and might be useful for early diagnosis of lung cancer.

The application of a novel high-resolution mass spectrometry-based analytical strategy to rapid metabolite profiling of a dual drug combination in humans

Metabolite profiling of combination drugs in complex matrix is a big challenge. Development of an effective data mining technique for simultaneously extracting metabolites of one parent drug from both background matrix and combined drug-related signals could be a solution. This study presented a novel high resolution mass spectrometry (HRMS)-based data-mining strategy to fast and comprehensive metabolite identification of combination drugs in human. The model drug combination was verapamil-irbesartan (VER-IRB), which is widely used in clinic to treat hypertension. First, mass defect filter (MDF), as a targeted data mining tool, worked effectively except for those metabolites with similar MDF values. Second, the accurate mass-based background subtraction (BS), as an untargeted data-mining tool, was able to recover all relevant metabolites of VER-IRB from the full-scan MS dataset except for trace metabolites buried in the background noise and/or combined drug-related signals. Third, the novel ring double bond (RDB; valence values of elements in structure) filter, could show rich structural information in more sensitive full-scan MS chromatograms; however, it had a low capability to remove background noise and was difficult to differentiate the metabolites with RDB coverage. Fourth, an integrated strategy, i.e., untargeted BS followed by RDB, was effective for metabolite identification of VER and IRB, which have different RDB values. Majority of matrix signals were firstly removed using BS. Metabolite ions for each parent drug were then isolated from remaining background matrix and combined drug-related signals by imposing of preset RDB values/ranges around the parent drug and selected core substructures. In parallel, MDF was used to recover potential metabolites with similar RDB. As a result, a total of 74 metabolites were found for VER-IRB in human plasma and urine, among which ten metabolites have not been previously reported in human. The results demonstrated that the combination of accurate mass-based multiple data-mining techniques, i.e., untargeted background subtraction followed by ring double bond filtering in parallel with targeted mass defect filtering, can be a valuable tool for rapid metabolite profiling of combination drug.

Forecasting bivalve landings with multiple regression and data mining techniques: The case of the Portuguese Artisanal Dredge Fleet

This paper develops a decision support tool that can help fishery authorities to forecast bivalve landings for the dredge fleet accounting for several contextual conditions. These include weather conditions, phytotoxins episodes, stock-biomass indicators per species and tourism levels. Vessel characteristics and fishing effort are also taken into account for the estimation of landings. The relationship between these factors and monthly quantities landed per vessel is explored using multiple linear regression models and data mining techniques (random forests, support vector machines and neural networks). The models are specified for different regions in the Portugal mainland (Northwest, Southwest and South) using six years of data 2010–2015). Results showed that the impact of the contextual factors varies between regions and also depends on the vessels target species. The data mining techniques, namely the random forests, proved to be a robust decision support tool in this context, outperforming the predictive performance of the most popular technique used in this context, i.e. linear regression.

A comparative analysis of classifiers in cancer prediction using multiple data mining techniques

A comparative analysis of classifiers in cancer prediction using multiple data mining techniques

A comparative analysis of classifiers in cancer prediction using multiple data mining techniques

In recent years, application of data mining methods in health industry has received increased attention from both health professionals and scholars. This paper presents a data mining framework for detecting breast cancer based on real data from one of the Iran hospitals by applying association rules and the most commonly used classifiers. The former were adopted for reducing the size of datasets, while the latter were chosen for cancer prediction. A k-fold cross-validation procedure was included for evaluating the performance of the proposed classifiers. Among the six classifiers used in this paper, support vector machine achieved the best results, with an accuracy of 93%. It is worth mentioning that the approach proposed can be applied for detecting other diseases as well.

DTreeSim: A new approach to compute decision tree similarity using re-mining

A number of recent studies have used a decision tree approach as a data mining technique; some of them needed to evaluate the similarity of decision trees to compare the knowledge reflected in different trees or datasets. There have been multiple perspectives and multiple calculation techniques to measure the similarity of two decision trees, such as using a simple formula or an entropy measure. The main objective of this study is to compute the similarity of decision trees using data mining techniques. This study proposes DTreeSim, a new approach that applies multiple data mining techniques (classification, sequential pattern mining, and k-nearest neighbors) sequentially to identify similarities among decision trees. After the construction of decision trees from different data marts using a classification algorithm, sequential pattern mining was applied to the decision trees to obtain rules, and then the k-nearest neighbor algorithm was performed on these rules to compute similarities using two novel measures: general similarity and pieced similarity. Our experimental studies compared the results of these novel similarity measures and also compared our approach with existing approaches. Our comparisons indicate that our proposed approach performs better than existing approaches, because it takes into account the values of the branches in the trees through sequential pattern mining.

Metabolite identification of the antimalarial piperaquine in vivo using liquid chromatography-high-resolution mass spectrometry in combination with multiple data-mining tools in tandem.

Artemisinin-based combination therapy is widely used for the treatment of uncomplicated Plasmodium falciparum malaria, and piperaquine (PQ) is one of important partner drugs. The pharmacokinetics of PQ is characterized by a low clearance and a large volume of distribution; however, metabolism of PQ has not been thoroughly investigated. In this work, the metabolite profiling of PQ in human and rat was studied using liquid chromatography tandem high-resolution LTQ-Orbitrap mass spectrometry (HRMS). The biological samples were pretreated by solid-phase extraction. Data processes were carried out using multiple data-mining techniques in tandem, i.e., isotope pattern filter followed by mass defect filter. A total of six metabolites (M1-M6) were identified for PQ in human (plasma and urine) and rat (plasma, urine and bile). Three reported metabolites were also found in this study, which included N-oxidation (M1, M2) and carboxylic products (M3). The subsequent N-oxidation of M3 resulted in a new metabolite M4 detected in urine and bile samples. A new metabolic pathway N-dealkylation was found for PQ in human and rat, leading to two new metabolites (M5 and M6). This study demonstrated that LC-HRMS(n) in combination with multiple data-mining techniques in tandem can be a valuable analytical strategy for rapid metabolite profiling of drugs. Copyright © 2016 John Wiley & Sons, Ltd.

Identification of metabolites of oridonin in rats with a single run on UPLC-Triple-TOF-MS/MS system based on multiple mass defect filter data acquisition and multiple data processing techniques.

Oridonin (ORI) is an active natural ent-kaurane diterpenoid ingredient originating from well-known traditional Chinese herb medicine and is expected to be pursued as a new anticancer agent. In the present study, a novel and efficient approach was developed for in vivo screening and identification of ORI metabolites using ultra high performance liquid chromatography coupled with hybrid triple quadrupole time-of-flight mass spectrometry (UPLC-Triple-TOF-MS/MS). This analytical strategy was as follows: an effective on-line data acquisition method multiple mass defect filter (MMDF) combined with dynamic background subtraction (DBS), was developed to trace all of potential metabolites of ORI. The MMDF and DBS method could trigger an information dependent acquisition scan, which could give the information of low-level metabolites masked by background noise and endogenous components in complex matrix. Moreover, the sensitive and specific multiple data-mining techniques including extracted ion chromatography, mass defect filtering, product ion filtering and neutral loss filtering were employed to identify the metabolites of ORI. Then, structures for the metabolites were successfully assigned based on accurate masses, the mass fragmentation of ORI and metabolic knowledge. Finally, an important parameter Clog P was used to estimate the retention time of isomers. Based on the proposed strategy, 16 phase I and 2 phase II metabolites were detected in rats after oral administration of ORI. The main biotransformation route of ORI was identified as reduction, oxidation, dehydroxylation and glucuronic acid conjugation. This is the first study of ORI metabolism in vivo. This study not only proposed a practical strategy for rapidly screening and identifying metabolites, but also provided useful information for further study of the pharmacology and mechanism of ORI in vivo. At the same time this methodology can be widely applied for the structural characterization of the metabolites of other ent-kaurane diterpenoid.

The application of high-resolution mass spectrometry-based data-mining tools in tandem to metabolite profiling of a triple drug combination in humans

Patients are usually exposed to multiple drugs, and metabolite profiling of each drug in complex biological matrices is a big challenge. This study presented a new application of an improved high resolution mass spectrometry (HRMS)-based data-mining tools in tandem to fast and comprehensive metabolite identification of combination drugs in human. The model drug combination was metronidazole-pantoprazole-clarithromycin (MET-PAN-CLAR), which is widely used in clinic to treat ulcers caused by Helicobacter pylori. First, mass defect filter (MDF), as a targeted data processing tool, was able to recover all relevant metabolites of MET-PAN-CLAR in human plasma and urine from the full-scan MS dataset when appropriate MDF templates for each drug were defined. Second, the accurate mass-based background subtraction (BS), as an untargeted data-mining tool, worked effectively except for several trace metabolites, which were buried in the remaining background signals. Third, an integrated strategy, i.e., untargeted BS followed by improved MDF, was effective for metabolite identification of MET-PAN-CLAR. Most metabolites except for trace ones were found in the first step of BS-processed datasets, and the results led to the setup of appropriate metabolite MDF template for the subsequent MDF data processing. Trace metabolites were further recovered by MDF, which used both common MDF templates and the novel metabolite-based MDF templates. As a result, a total of 44 metabolites or related components were found for MET-PAN-CLAR in human plasma and urine using the integrated strategy. New metabolic pathways such as N-glucuronidation of PAN and dehydrogenation of CLAR were found. This study demonstrated that the combination of accurate mass-based multiple data-mining techniques in tandem, i.e., untargeted background subtraction followed by targeted mass defect filtering, can be a valuable tool for rapid metabolite profiling of combination drugs in vivo.

A Novel Pattern Merger Algorithm for generating Actionable Rules for Multi-Source Combined Mining

The idea of combined mining is very useful and flexible for identifying in-depth patterns through combining sets of items from multiple datasets or using multiple data mining techniques. The identified combined patterns disclose in-depth business intelligence, which are more informative and actionable for business decision-making. The business data is scattered at various locations and to arrive at decisions it needs analysis of data by integrating entire data. This paper emphasizes on applying Association Rule Mining technique on various data sources located at different locations and patterns so obtained are transported to main site. Finally by applying the proposed novel Pattern Merger Algorithm on the aforesaid patterns, all the generated patterns are merged to obtain actionable rules. These actionable patterns assist in strategic business planning and also pin points various issues arising post application of the data integration technique. Domain Knowledge concept is also included in the Rule generation technique to obtain the final results which are in the form of actionable rules. General Terms Pattern Merger Algorithm , Multisource Combined Mining.

Study on medication regularity of traditional Chinese medicines in treating attention deficit hyperactivity disorder based on data mining

In this study, efforts were made to collect 88 traditional Chinese medicine (TCM) prescriptions for treating ADHD from public literatures, establish a medication database and conduct data mining on the efficacy, properties, tastes, meridian entries and compatibility regularity of TCMs by using frequency analysis, association rules analysis and cluster analysis. The results showed 33 herbs with a high frequency (≥ 10%). The top 12 frequently applied ADHD herbs were respectively Acori Tatarinowii Rhizome, Polygalae Radix, Prepared Rehmanniae Radix, Os Draconis, Glycyrrhizae Radix et Rhizome, Poria, Concha Ostreae, Testudinis Carapacis et Plastri, Paeoniae Alba Radix, Schisandrae Fructus Chinensis, Fructus Corni, Dioscoreae Rhizoma. The top seven types of frequently applied ADHD herbs by effect were respectively tonic herb, sedative herbs, livei-wind calming herbs, heat clearing herbs, astringent herbs, awaking herbs and damp clearing and diuresis promoting herbs. The top three properties were warm, cold and neutral, with warm as the dominance. The top three tastes were sweetness, bitterness and acridity, with sweetness as the dominance. The top three meridian entries were liver meridian, heart meridian and kidney meridian. The association rules analysis showed commonly used herbal pairs. Polygalae Radix-Acori Tatarinowii Rhizome, Concha Ostreae-Os Draconis, Prepared Rehmanniae Radix-Fructus Corni-Os Draconis. Eight highly frequent cluster prescriptions were summarized based on the cluster analysis. This study indicates that basic TCM prescriptions for treating ADHD were principally composed of tonics (nourishing liver and kidney, invigorating heart and spleen) and tranquilization, which were supplemented with herbs for calming liver, clearing heat, promoting qi-circulation, promoting blood-circulation, reducing phlegm and calming endogenous wind. The combination of multiple data mining techniques can reflect the clinical medication regularity of herbs for treating ADHD in an overall and objective manner.