Effectiveness Of Data Mining Techniques Research Articles

An analysis of association rule mining for classification based on artificial immune system

One of the most important and effective data mining techniques is association rule mining. Associative classification uses the principle of rule finding and technique of classification to generate a classifier for prediction. The rule search method is also computationally expensive for small support threshold values, which are critical for designing an efficient classifier. The artificial immune system (AIS) employs the powerful informational capacities of the immune system. The population-based search model combined with evolutionary computation techniques allows the artificial immune system clonal selection methodology to manage a complex search space. This study calculated accuracy across a variety of clonal characteristics and generations to assess the efficacy of the artificial immune system-based categorization method. The output of these systems is shown on several benchmark datasets. Based on the accuracy of the different clonal factors (0.1 to 0.9) and generations (10, 20, 30, 40, 50, and 60), a comparison study is performed. The accuracy is computed using four standard datasets. It is observed that in every dataset for several generations, the approach provides the maximum accuracy with a clonal factor of 0.4.

Brand Switching Pattern Discovery by Data Mining Techniques for the Telecommunication Industry in Australia

There is more than one mobile-phone subscription per member of the Australian population. The number of complaints against the mobile-phone-service providers is also high. Therefore, the mobile service providers are facing a huge challenge in retaining their customers. There are a number of existing models to analyse customer behaviour and switching patterns. A number of switching models may also exist within a large market. These models are often not useful due to the heterogeneous nature of the market. Therefore, in this study we use data mining techniques to let the data talk to help us discover switching patterns without requiring us to use any models and domain knowledge. We use a variety of decision tree and decision forest techniques on a real mobile-phone-usage dataset in order to demonstrate the effectiveness of data mining techniques in knowledge discovery. We report many interesting patterns, and discuss them from a brand-switching and marketing perspective, through which they are found to be very sensible and interesting.

Classifying Educational Data Using Support Vector Machines: A Supervised Data Mining Technique

With increase in Educational Institutions there is increase in new trends which results in large data. The data is unstructured which need to be transformed into structured form and to find out meaningful information, effective Mining Tools are required. Educational Data Mining helps in facilitating utilization of resources related to student performance, predicting placement results and finding new educational trends. In this paper placement data of students has been taken and classification approach using SVM is followed on training data for predicting results which not only helps educational institutions to improve student placements from extracted knowledge as well enhances the competitive advantage and decision making by applying data mining techniques. SVM is a supervised learning and effective data mining technique to train the data for pattern recognition and predictions.

Applications of association rule mining in health informatics: a survey

Association rule mining is an effective data mining technique which has been used widely in health informatics research right from its introduction. Since health informatics has received a lot of attention from researchers in last decade, and it has developed various sub-domains, so it is interesting as well as essential to review state of the art health informatics research. As knowledge discovery researchers and practitioners have applied an array of data mining techniques for knowledge extraction from health data, so the application of association rule mining techniques to health informatics domain has been focused and studied in detail in this survey. Through critical analysis of applications of association rule mining literature for health informatics from 2005 to 2014, it has been explored that, instead of the more efficient alternative approaches, the Apriori algorithm is still a widely used frequent itemset generation technique for application of association rule mining for health informatics. Moreover, other limitations related to applications of association rule mining for health informatics have also been identified and recommendations have been made to mitigate those limitations. Furthermore, the algorithms and tools utilized for application of association rule mining have also been identified, conclusions have been drawn from the literature surveyed, and future research directions have been presented.

An Optimized Algorithm for Association Rule Mining Using FP Tree

Data mining is used to deal with the huge size of the data stored in the database to extract the desired information and knowledge. It has various techniques for the extraction of data; association rule mining is the most effective data mining technique among them. It discovers hidden or desired pattern from large amount of data. Among the existing techniques the frequent pattern growth (FP growth) algorithm is the most efficient algorithm in finding out the desired association rules. It scans the database only twice for the processing. The issue with the FP growth algorithm is that it generates a huge number of conditional FP trees. In the proposed work, we have designed a new technique which mines out all the frequent item sets without the generation of the conditional FP trees. Unlike FP tree it scans the database only once which reduces the time efficiency of the algorithm. It also finds out the frequency of the frequent item sets to find out the desired association rules.

Exploiting Data Mining Techniques For Improving the Efficiency of Time Series Data

The research work in data mining has achieved a high attraction due to the importance of its applications This paper addresses some theoretical and practical aspects on Exploiting Data Mining Techniques for Improving the Efficiency of Time Series Data using SPSS-CLEMENTINE. This paper can be helpful for an organization or individual when choosing proper software to meet their mining needs. In this paper, we propose utilizes the famous data mining software SPSS Clementine to mine the factors that affect information from various vantage points and analyse that information. However the purpose of this paper is to review the selected software for data mining for improving efficiency of time series data. Data mining techniques is the exploration and analysis of data in order to discover useful information from huge databases. So it is used to analyse a large audit data efficiently for Improving the Efficiency of Time Series Data. SPSS- Clementine is object-oriented, extended module interface, which allows users to add their own algorithms and utilities to Clementine’s visual programming environment. The overall objective of this research is to develop high performance data mining algorithms and tools that will provide support required to analyse the massive data sets generated by various processes that is used for predicting time series data using SPSS- Clementine. The aim of this paper is to determine the feasibility and effectiveness of data mining techniques in time series data and produce solutions for this purpose.

Data Mining Techniques for Informative Motif Discovery

The discovery of motifs in biological sequence is a much explored and still exploring area of research in functional genomics since they control the expression or regulation of a group of genes involved in a similar cellular function. This paper explores the use of data mining techniques by various researchers as a solution to discover motifs in biological sequence. Although data mining techniques has not been applied extensively by researchers as compared to other algorithms. But in recent years data mining techniques has caused a wide attention by the researchers to find motifs in biological sequence. This paper is an attempt towards exploring the effectiveness of data mining techniques for motif discovery.

Application of data mining techniques in building predictive models for oil and gas problems: a case study on casing corrosion prediction

This paper describes the use of (supervised) data mining to predict casing corrosion in carbon geological storage projects. This study discusses: 1) data pre-processing such as missing value handling and discretisation; 2) feature selection methods such as correlation coefficient, signal-to-noise ratio, information gain, Gini index, and the k-nearest neighbour (KNN) approach; 3) classification techniques including decision trees (C4.5 and CART) and Bayesian networks; 4) evaluation methods like cross-validation as four successive steps of supervised learning. The experimental analysis of the casing corrosion problem based on the given supervised learning framework shows the effectiveness of data mining techniques in finding features relevant to the problem under study and in building models to predict and identify casing corrosion.

Ontology Based Data Analysing Approach for Actionable Knowledge Discovery

In Data Mining, the effectiveness of association rules is limited by the huge quantity of delivered rules. In this manuscript, we propose a new approach to prune and filter discovered rules. An interactive and iterative framework is designed to assist the user along the analyzing task. The manuscript focus on medical data test set for detailed analysis and formation of the ontologies. In the proposed approach, the data set of medical records is entered in the back - end database. The development of associations and the ontology is fully dynamic. In case, any symptom is searched, the proposed approach search from the back end database and creates the ontology that is dynamic in execution. It refers that the unsupervised approach of ontology generation is implemented so that the unbiased results can be achieved. The implementation of proposed work shows the dynamic results in terms of rules found and their effectiveness in the real world scenario. 3 M) which is intended to make data mining workable in supporting decision-making actions in the real world. Current data mining tools and techniques drive a paradigm shift from traditional data-centred hidden pattern mining to domain-driven actionable knowledge discovery (AKD). AKD must cater for domain knowledge and environmental factors, balance technical significance and business expectations from both objective and subjective perspectives and support automatically converting patterns into deliverables in business friendly and operable forms such as actions or rules. In domain-driven framework, data mining analysts and domain-specific business analysts complement each other with regard to in-depth granularity and constrained environment through interactive system support. The involvement of domain experts and their knowledge can assist in developing highly effective domain specific data mining techniques, and reduce the complexity of knowledge discovery. The involvement of domain-related social intelligence into KDD process not only strengthens technical development and performance, but highlights business expectations and actionable capability of the identified results. In Data Mining, the effectiveness of association rules is strappingly limited by the huge quantity of delivered rules. In this manuscript, we propose a new approach to prune and filter discovered rules. An association rule is described as the implication X → Y where X and Y are sets of items and

Data Mining for Detecting Carelessness or Mala Fide Intention

Fraud is one of the greatest challenges for the organizations, it needs a machine to be equipped with data mining algorithms , so that it can detect a crime pattern before it takes place. This paper will explore the data mining and Knowledge discovery in data base and later one of the most effective data mining techniques called Benford’s law for detecting the fake entries in medical insurance claims, electricity bills, water bills etc will be discussed. Applications of Benford’s law with limitations will be discussed so that machines exhibits some intelligence in its domain and later proposed to embed the Benford law in software to identify all the entries made by carelessness or with a mala fide intention .

A set of novel mining tools for efficient biological knowledge discovery

In last decades, Bioinformatics has become an emerging field of science with a wide variety of applications in many research areas. The primary goal of bioinformatics is to detect useful biological knowledge hidden under the large volumes of DNA/RNA sequences and structures, literature and other biological and biomedical data, to gain a greater insight into their relationships and, therefore, to enhance the discovery and the comprehension of biological processes. In order to fully exploit the new opportunities that emerge, novel data and text mining techniques have to be developed to effectively address the fundamental biological issue of managing and uncovering meaningful patterns and correlations from these large biological and biomedical data repositories. In this work, we propose an effective data mining technique for analysing biological and biomedical data. The proposed mining process is efficient enough to be applied to various types of biological and biomedical data. To prove the concept, we experiment with applying the data mining technique into two distinct areas, including biomedical text documents and data. In addition, based on the proposed approach, we develop two mining tools, namely the Bio Search Engine and the Genome-Based Population Clustering.

Gravitational search algorithm–optimized neural misuse detector with selected features by fuzzy grids–based association rules mining

Feature selection is one of the most important techniques for data preprocessing in classification problems. In this paper, fuzzy grids–based association rules mining, as an effective data mining technique, is used for feature selection in misuse detection application in computer networks. The main idea of this algorithm is to find the relationships between items in large datasets so that it detects correlations between inputs of the system and then eliminates the redundant inputs. To classify the attacks, a fuzzy ARTMAP neural network is employed whose training parameters are optimized by gravitational search algorithm. The performance of the proposed system is compared with some other machine learning methods in the same application. Experimental results show that the proposed system, when choosing optimum “feature subset size-adjustment” parameter, performs better in terms of detection rate, false alarm rate, and cost per example in classification problems. In addition, employing the reduced-size feature set results in more than 8.4 percent reduction in computational complexity.

Role of knowledge management and analytical CRM in business: data mining based framework

PurposeThe purpose of the paper is to provide a thorough analysis of the concepts of business intelligence (BI), knowledge management (KM) and analytical CRM (aCRM) and to establish a framework for integrating all the three to each other. The paper also seeks to establish a KM and aCRM based framework using data mining (DM) techniques, which helps in the enterprise decision‐making. The objective is to share how KM and aCRM can be integrated into this seamless analytics framework to sustain excellence in decision making using effective data mining techniques and to explore how working on such aCRM system can be effective for enabling organizations delivering complete solutions.Design/methodology/approachThis paper is based on focused and dedicated study of the literature present on the aCRM, KM and data mining techniques. The paper considered how to develop a strategy and operational framework that would build aCRM on the foundation of existing DM techniques and KM approach to meet the business challenges. Based on this research, a customized, integrated framework, to match the needs of business was designed.FindingsKM focuses on managing knowledge within the organization and aCRM focuses on gaining analytical information from the customer data. Both KM and aCRM help in the decision making process and understanding. This knowledge is difficult to uncover. Hence, this paper explains the importance of data mining tools and techniques to uncover knowledge by the integration between KM and aCRM. This paper presents an integrated KM and aCRM based framework using DM techniques.Research limitations/implicationsAll the firms may not be in favor of adopting KM while implementing aCRM. The KM requires a convalesce of organizational culture, technology innovations, effective work force in culminating knowledge dissemination in all business domains.Practical implicationsThe organizations implementing this knowledge enabled aCRM framework would be easily able to convert their business knowledge via the analytical CRM to solve many business issues, such as increase response rates from direct mail, telephone, e‐mail, and internet delivered marketing campaigns, increased sales and increased services. With aCRM, firms can identify their most profitable customers and use this knowledge for promotional schemes for those customers as well as identify future customers with prediction on ROI.Originality/valueThe need for the integration of KM and aCRM is clear. It is written for practitioners who are looking for approaches to improve business performance and maintain high profits for their business by incorporating knowledge‐enabled aCRM in their setup.

Applying Data Visualization and Mining Techniques to Improvise Luxury Goods Promotion through E-Retail

Data mining has been used by researchers and key organizations for variety of purpose to improve the business intelligence processes. The research paper explains the use of data mining to promote selling of luxury goods with in e-retailing value chain in order to facilitate process of data analysis and decision making. We have analyzed how launching a product in e-retail domain requires sufficient detailed results taken from effective data mining technique. In order to determine the regions with closed and heavily populated wealth clusters, clustering technique was applied. To achieve the desired objective first we have obtained the database of most affluent people across the globe and further applied clustering and data visualization technique to do some predictive analysis which can further help in focusing product marketing in e-retailing. The result achieved describes region fragmentation on basis of wealth and accordingly the e-retail concept can be implemented for those specific regions. This paper illustrates cutting edge advantages of product selling through e-retail and the challenges encountered to implement data mining to achieve the desired objective.

Discovering Interesting Motif-Sets for Multi-Class Protein Sequence Classification

In this article, we propose an effective data mining technique for multi-class protein sequence classification. The technique, which can discover discriminative motif-sets for classification, performs its tasks in two phases. In Phase 1, it makes use of a popular motif discovery algorithm called MEME (Multiple Expectation Maximization for Motif Elicitation) to discover a set of highly conserved motifs in each protein family of training sequences. The highly conserved motif-sets discovered in each family may overlap with each other and may therefore not be unique enough to allow them to be used for classification. Phase 2, therefore, makes use of a pattern discovery approach to discover the interesting motif-sets in each protein family that are useful for classification with a single classifier. Based on these motif-sets, the functional family of each independent testing sequence can then be determined. For experimentation, the proposed technique has been tested with different sets of protein sequences. Experimental results show that it outperforms other existing protein sequence classifiers and can effectively classify proteins into their corresponding functional families. In addition, the motif-sets discovered during the training process have been found to be biologically meaningful.

Time-resolved metabolomics reveals metabolic modulation in rice foliage

BackgroundTo elucidate the interaction of dynamics among modules that constitute biological systems, comprehensive datasets obtained from "omics" technologies have been used. In recent plant metabolomics approaches, the reconstruction of metabolic correlation networks has been attempted using statistical techniques. However, the results were unsatisfactory and effective data-mining techniques that apply appropriate comprehensive datasets are needed.ResultsUsing capillary electrophoresis mass spectrometry (CE-MS) and capillary electrophoresis diode-array detection (CE-DAD), we analyzed the dynamic changes in the level of 56 basic metabolites in plant foliage (Oryza sativa L. ssp. japonica) at hourly intervals over a 24-hr period. Unsupervised clustering of comprehensive metabolic profiles using Kohonen's self-organizing map (SOM) allowed classification of the biochemical pathways activated by the light and dark cycle. The carbon and nitrogen (C/N) metabolism in both periods was also visualized as a phenotypic linkage map that connects network modules on the basis of traditional metabolic pathways rather than pairwise correlations among metabolites. The regulatory networks of C/N assimilation/dissimilation at each time point were consistent with previous works on plant metabolism. In response to environmental stress, glutathione and spermidine fluctuated synchronously with their regulatory targets. Adenine nucleosides and nicotinamide coenzymes were regulated by phosphorylation and dephosphorylation. We also demonstrated that SOM analysis was applicable to the estimation of unidentifiable metabolites in metabolome analysis. Hierarchical clustering of a correlation coefficient matrix could help identify the bottleneck enzymes that regulate metabolic networks.ConclusionOur results showed that our SOM analysis with appropriate metabolic time-courses effectively revealed the synchronous dynamics among metabolic modules and elucidated the underlying biochemical functions. The application of discrimination of unidentified metabolites and the identification of bottleneck enzymatic steps even to non-targeted comprehensive analysis promise to facilitate an understanding of large-scale interactions among components in biological systems.

Three Data Mining Techniques To Improve Lazy Structure−Activity Relationships for Noncongeneric Compounds

We present three simple, yet effective data mining techniques for lazy structure-activity relationships (SARs) of noncongeneric compounds. In lazy SARs, classifications are particularly tailored for each test compound. Therefore, it is possible to make the most of the structure of a test compound. In our case, we derive its substructures and use them to determine similar structures. To obtain a well-balanced and representative set of structural descriptors, we enrich this set by strongly activating or deactivating fragments from the training set and subsequently remove redundant fragments. Finally, we perform k-Nearest Neighbor classification for several values of k and take a vote among the resulting predictions. These techniques (enrichment, removing redundancy, and voting) are integrated into the system iSAR (instance-based structure-activity relationships) and tested individually to show the relative contribution to the system's performance. Experiments on three data sets indicate that this simple and lightweight approach performs at least on the same level as other, more complex approaches.

Feature selection and classification using flexible neural tree

The purpose of this research is to develop effective machine learning or data mining techniques based on flexible neural tree FNT. Based on the pre-defined instruction/operator sets, a flexible neural tree model can be created and evolved. This framework allows input variables selection, over-layer connections and different activation functions for the various nodes involved. The FNT structure is developed using genetic programming (GP) and the parameters are optimized by a memetic algorithm (MA). The proposed approach was applied for two real-world problems involving designing intrusion detection system (IDS) and for breast cancer classification. The IDS data has 41 inputs/features and the breast cancer classification problem has 30 inputs/features. Empirical results indicate that the proposed method is efficient for both input feature selection and improved classification rate.

レセプトデータを使用した医薬品有害事象検出データマイニング手法の開発

Objective : To detect signals of potential drug adverse events (DAEs) through data mining of health insurance claims.Design and Data : Retrospective observational study. The data used were the database of health insurance claims collected and maintained by the Japan Medical Data Center consisting of 312, 797 medical and pharmaceutical claims in one year (August 2003 through July 2004) linked uniquely for 35, 410 patients using an encryption technique to ensure privacy.Methods : We counted all combinations (cross product or Cartesian product) of drugs and diagnoses appearing in the same claims and counted the number of times a given drug was prescribed preceding the suspected diagnosis in all combinations of the drug and the diagnosis appearing in a claim, i.e., the prescription date precedes the diagnosing date (the preceding number). We calculated the expected preceding number from the overall prevalence of drugs and diagnoses, and then calculated the observed and expected ratio, which was used as the signal indices. We calculated the signal indices on the health insurance claims data to detect DAEs of psychiatric drugs.Results : Amoxapine and trazodone HCL showed high signal indices with paralytic ileus and convulsion (epilepsy) as documented in their package inserts. However, paroxetine HCL and etizolam showed high signal indices with these potential adverse events although no such DAEs are documented in their package inserts.Conclusions : The undocumented high signal indices observed between the drugs and diagnoses indicate the potential DAEs and warrant in-depth pharmacovigilance. Given the strength of health insurance claims with a well-defined source population and accurate drug exposure, the proposed signal index will likely prove to be an effective data mining technique when combined with nested case-control analysis and counter-matching.

Distributed Multivariate Regression Using Wavelet-Based Collective Data Mining

This paper presents a method for distributed multivariate regression using wavelet-based collective data mining (CDM). The method seamlessly blends machine learning and the theory of communication with the statistical methods employed in parametric multivariate regression to provide an effective data mining technique for use in a distributed data and computation environment. The technique is applied to two benchmark data sets, producing results that are consistent with those obtained by applying standard parametric regression techniques to centralized data sets. Evaluation of the method in terms of mode accuracy as a function of appropriateness of the selected wavelet function, relative number of nonlinear cross-terms, and sample size demonstrates that accurate parametric multivariate regression models can be generated from distributed, heterogeneous, data sets with minimal data communication overhead compared to that required to aggregate a distributed data set. Application of this method to linear discriminant analysis, which is related to parametric multivariate regression, produced classification results on the Iris data set that are comparable to those obtained with centralized data analysis.