Evolutionary stratified training set selection for extracting classification rules with trade off precision-interpretability

Instance selection and feature extraction is one of the most important task in data mining, due to the huge amount of data is constantly being produced in many fields. If the dataset is very large means most of the existing machine learning algorithms are inapplicable to handle such huge amount of data and computational cost is high. Two of the approaches have been used for solving this problem. One is scaling up algorithms and another one is data reduction. Scaling up data mining algorithm is not always feasible, but data reduction is possible. In this paper we take both, instance selection and feature extraction for data reduction. Instance selection is a technique that will reduce the size of the original training data. Feature extraction is input data having m dimension space that should be mapped into lower dimension space i.e., eliminate those components which are contributing less information. In this paper Cuttlefish optimization algorithm is used for instance selection, while principal component analysis is used for feature extraction. The combination of feature extraction and instance selection will reduce the large amount of computational time of training the classifiers. The optimal extracted subset of data points and reduced feature space are providing almost similar detection rate, accuracy rate, false positive rate and takes less amount of computational time for training the classifiers what we obtained from using original dataset.

A Data mining is the activity that includes the collection, the use of historical data to discover regularity, patterns and relationship in a large data set. The usefulness of data mining is to specify a pattern to be found in the data mining task. The presences of data mining is motivated by the problem of data explosion which had been experienced lately these day where many organization or company collect so many years of data (purchasing data, sales data, damage data item, transaction data, and so on). In this paper data mining methods been used to analyze the damage data of finished products, with the goal of producing a pattern of the damage product. Based on the pattern from product’s damage, can be see and the aspects that affect to the damaged product. The purpose of this study is to show information about the relationship of damage data of finished goods using C4.5 algorithms in PT. Kayu Lapis Asli Murni, and display results in the form of a decision tree mining.
 Index Terms—data mining, C4.5 algorithms, decision tree, damage data of finished products, PT. Kayu Lapis Asli Murni

A common task in data mining is the visualization of multivariate objects on scatterplots, allowing human observers to perceive subtle inter-relations in the dataset such as outliers, groupings or other regularities. Least- squares multidimensional scaling (MDS) is a well known Exploratory Data Analysis family of techniques that produce dissimilarity or distance preserving layouts in a nonlinear way. In this framework, the issue of visualizing large multidimensional datasets through MDS-based methods is addressed. An original scheme providing very accurate layouts of large datasets is introduced. It is a compromise between the computational complexity O(N5/2) and the accuracy of the solution that makes it suitable both for visualization of fairly large datasets and preprocessing in pattern recognition tasks.

In knowledge management process, data mining technique can be used to extract and discover the valuable and meaningful knowledge from a large amount of data. Nowadays, data mining has given a great deal of concern and attention in the information industry and in society as a whole. This technique is an approach that is currently receiving great attention in data analysis and it has been recognized as a newly emerging analysis tool (Osei-Bryson, 2010; Park, 2001; Sinha, 2008; Tso & Yau, 2007; Wan, 2009; Zanakis, 2005; Zhuang et al., 2009). Additionally, among the major tasks in data mining are classification and prediction; concept description; rule association; cluster analysis; outlier analysis; trend and evaluation analysis; statistical analysis and others. Classification and prediction tasks are among the popular tasks in data mining; and widely used in many areas especially for trend analysis and future planning. In fact, classification technique is supervised learning, which is the class level or prediction target is already known. As a result, the classification model which is represented through rules structures will be constructed in the classification process. In this case, the constructed model will be representing the precious knowledge and it can be used for future planning. There are many areas which adapted this approach to solve their problems such as in finance, medical, marketing, stock, telecommunication, manufacturing, health care, customer relationship and etc. However, the data mining application has not attracted much attention from people in Human Resource (HR) field (Chien & Chen, 2008; Ranjan, 2008). Besides that, in our previous study, most of the prediction applications are used to predict stock, demand, rate, risk, event and others; but there are quite limited studies on human prediction. In addition prediction applications are mainly developed in business and industrious fields; and quite restricted studies involved human talent in an organization (Jantan et al., 2009). HR data can provide a rich resource for knowledge discovery and for decision support system development. Recently, an organization has to struggle effectively in term of cost, quality, service or innovation. All these depend on having enough right people with the right skills, employed

Due to today’s advancement in technology and businesses, fraud detection has become a critical component of financial transactions. Considering vast amounts of data in large datasets, it becomes more difficult to detect fraud transactions manually. In this research, we propose a combined method using both data mining and statistical tasks, utilizing feature selection, resampling and cost-sensitive learning for credit card fraud detection. In the first step, useful features are identified using genetic algorithm. Next, the optimal resampling strategy is determined based on the design of experiments (DOE) and response surface methodologies. Finally, the cost sensitive C4.5 algorithm is used as the base learner in the Adaboost algorithm. Using a real-time data set, results show that applying the proposed method significantly reduces the misclassification cost by at least 14% compared with Decision tree, Naive bayes, Bayesian Network, Neural network and Artificial immune system.

Learning classifier systems (LCSs) are rule-based inductive learning systems that have been widely used in the field of supervised and reinforcement learning over the last few years. This paper employs sUpervised Classifier System (UCS), a supervised learning classifier system, that was introduced in 2003 for classification tasks in data mining. We present an adaptive framework of UCS on top of a self-organized map (SOM) neural network. The overall classification problem is decomposed adaptively and in real time by the SOM into subproblems, each of which is handled by a separate UCS. The framework is also tested with replacing UCS by a feedforward artificial neural network (ANN). Experiments on several synthetic and real data sets, including a very large real data set, show that the accuracy of classifications in the proposed distributed environment is as good or better than in the nondistributed environment, and execution is faster. In general, each UCS attached to a cell in the SOM has a much smaller population size than a single UCS working on the overall problem; since each data instance is exposed to a smaller population size than in the single population approach, the throughput of the overall system increases. The experiments show that the proposed framework can decompose a problem adaptively into subproblems, maintaining or improving accuracy and increasing speed.

We investigate the use of biased sampling according to the density of the data set to speed up the operation of general data mining tasks, such as clustering and outlier detection in large multidimensional data sets. In density-biased sampling, the probability that a given point will be included in the sample depends on the local density of the data set. We propose a general technique for density-biased sampling that can factor in user requirements to sample for properties of interest and can be tuned for specific data mining tasks. This allows great flexibility and improved accuracy of the results over simple random sampling. We describe our approach in detail, we analytically evaluate it, and show how it can be optimized for approximate clustering and outlier detection. Finally, we present a thorough experimental evaluation of the proposed method, applying density-biased sampling on real and synthetic data sets, and employing clustering and outlier detection algorithms, thus highlighting the utility of our approach.

A scalable decision tree system and its application in pattern recognition and intrusion detection

The subject of the paper is methods of classification of data mining and data processing tasks in the economy, as well as their classification characteristics. Methodology. The research used the general scientific methodology of analysis, synthesis, generalization, comparison. Taxonomy methods are used in the compilation of the classification. The selection of actual economic problems is carried out by researching scientific publications on analysis and data processing. The purpose of the article is to compile an actual classification of data mining and data processing tasks in the economy and to refine the terminology of this branch of science. Results. The classification of data mining tasks is proposed, consisting of four levels. All economic tasks of data mining are divided into two large groups: predictive and descriptive. Each group is subdivided into several classes, which combine tasks with similar taxonomic features. These are the classes of tasks classification, regression, clustering, link analysis, and outlier analysis. Classes of tasks are divided into types. An important criterion for this is the dimensionality of the input data representation. It means the number of neighbours for each individual data element. The data can be presented in the form of series, matrices, and graphs. The methods of analysing each form of data presentation vary considerably. The definition of data processing is clarified and a classification of relevant tasks is proposed. At the top level of classification, the data processing tasks are divided into two groups, depending on whether the order of the elements in the input data changes or not. The main classes of data processing tasks are identified, such as ranking, sorting, filtering, cleansing, recovery, and quantization. Practical use. The results of the research can be used to model intellectual decision-support systems. They allow improving the processes of problem formulation in data mining and data processing tasks. Also, it helps in the formalization of the procedures for selecting methods of their solutions. This leads to an increase in economic efficiency.

<p>Classifying high-dimensional data are a challenging task in data mining. Gene expression data is a type of high-dimensional data that has thousands of features. The study was proposing a method to extract knowledge from high-dimensional gene expression data by selecting features and classifying. Lasso was used for selecting features and the classification and regression tree (CART) algorithm was used to construct the decision tree model. To examine the stability of the lasso decision tree, we performed bootstrap aggregating (Bagging) with 50 replications. The gene expression data used was an ovarian tumor dataset that has 1,545 observations, 10,935 gene features, and binary class. The findings of this research showed that the lasso decision tree could produce an interpretable model that theoretically correct and had an accuracy of 89.32%. Meanwhile, the model obtained from the majority vote gave an accuracy of 90.29% which showed an increase in accuracy of 1% from the single lasso decision tree model. The slightly increasing accuracy shows that the lasso decision tree classifier is stable.</p>

Efficient algorithms to mine frequent patterns are crucial to many tasks in data mining. Since the Apriori algorithm was proposed in 1994, there have been several methods proposed to improve its performance. However, most still adopt its candidate set generation-and-test approach. In addition, many methods do not generate all frequent patterns, making them inadequate to derive association rules. We propose a pattern decomposition (PD) algorithm that can significantly reduce the size of the dataset on each pass, making it more efficient to mine all frequent patterns in a large dataset. The proposed algorithm avoids the costly process of candidate set generation and saves time by reducing the size of the dataset. Our empirical evaluation shows that the algorithm outperforms Apriori by one order of magnitude and is faster than FP-tree algorithm.

Application of Chi-square discretization algorithms to ensemble classification methods

Uncertainty in the analysis and interpretation of research information

For many learning algorithms, their learning accuracy will increase as the size of training data increases, forming the well-known learning curve. Usually a learning curve can be fitted by interpolating or extrapolating some points on it with a specified model. The obtained learning curve can then be used to predict the maximum achievable learning accuracy or to estimate the amount of data needed to achieve an expected learning accuracy, both of which will be especially meaningful to data mining on large data sets. Although some models have been proposed to model learning curves, most of them do not test their applicability to large data sets. In this paper, we focus on this issue. We empirically compare six potentially useful models by fitting learning curves of two typical classification algorithms--C4.5 (decision tree) and LOG (logistic discrimination) on eight large UCI benchmark data sets. By using all available data for learning, we fit a full-length learning curve; by using a small portion of the data, we fit a part-length learning curve. The models are then compared in terms of two performances: (1) how well they fit a full-length learning curve, and (2) how well a fitted part-length learning curve can predict learning accuracy at the full length. Experimental results show that the power law (y = a - b * x-c) is the best among the six models in both the performances for the two algorithms and all the data sets. These results support the applicability of learning curves to data mining.

Data mining tasks are often composed by multiple stages that may be linked each other to form various execution flows. Moreover, data mining tasks are often distributed since they involve data and tools located over geographically distributed environments, like the Grid. Therefore, it is fundamental to exploit effective formalisms, such as workflows, to model data mining tasks that are both multi-staged and distributed. The goal of this work is defining a workflow formalism and providing a visual software environment, named DIS3GNO, to design and execute distributed data mining tasks over the Knowledge Grid, a service-oriented framework for distributed data mining on the Grid. DIS3GNO supports all the phases of a distributed data mining task, including composition, execution, and results visualization. The paper provides a description of DIS3GNO, some relevant use cases implemented by it, and a performance evaluation of the system.KeywordsData MiningComputing NodeGrid ResourceExecution PlanData Mining TaskThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.