Data mining Algorithm

This paper compares the accuracy of Decision Tree and Bayesian Network algorithms for predicting the academic performance of undergraduate and postgraduate students at two very different academic institutes: Can Tho University (CTU), a large national university in Viet Nam; and the Asian Institute of Technology (AIT), a small international postgraduate institute in Thailand that draws students from 86 different countries. Although the diversity of these two student populations is very different, the data-mining tools were able to achieve similar levels of accuracy for predicting student performance: 73/71% for {fail, fair, good, very good} and 94/93% for {fail, pass} at the CTU/AIT respectively. These predictions are most useful for identifying and assisting failing students at CTU (64% accurate), and for selecting Very Good students for scholarships at the AIT (82% accurate). In this analysis, the Decision Tree was consistently 3-12% more accurate than the Bayesian Network. The results of these case studies give insight into techniques for accurately predicting student performance, compare the accuracy of data mining algorithms, and demonstrate the maturity of open source tools.

This paper describes a text mining tool that performs two tasks, namely document clustering and text summarization. These tasks have, of course, their corresponding counterpart in "conventional" data mining. However, the textual, unstructured nature of documents makes these two text mining tasks considerably more difficult t han their data mining counterparts. In our system document clustering is performed by using

This paper presents a Multi-Agent Market simulator designed for developing new agent market strategies based on a complete understanding of buyer and seller behaviors, preference models and pricing algorithms, considering user risk preferences and game theory for scenario analysis. This tool studies negotiations based on different market mechanisms and, time and behavior dependent strategies. The results of the negotiations between agents are analyzed by data mining algorithms in order to extract rules that give agents feedback to improve their strategies. The system also includes agents that are capable of improving their performance with their own experience, by adapting to the market conditions, and capable of considering other agent reactions.

Educational Data Mining (EDM) is the process of converting raw data from educational systems to useful information that can be used by educational software developers, students, teachers, parents, and other educational researchers. EDM is also an emerging discipline, concerned with developing methods for exploring the unique types of data that come from educational settings, and with using those methods to better understand students and the settings in which they learn. Whether educational data are taken from students' use of interactive learning environments, computer-supported collaborative learning, or administrative data from schools and universities, it often has multiple levels of meaningful hierarchy, which often need to be determined by properties in the data itself, rather than in advance. Issues of time, sequence, and context also play important roles in the study of educational data. The First International Conference on Educational Data Mining (EDM2008) brings together researchers from computer science, education, psychology, psychometrics, and statistics to analyze large data sets to answer educational research questions. The increase in instrumented educational software, as well as state databases of student test scores, has created large repositories of data reflecting how students learn. The EDM conference focuses on computational approaches for using those data to address important educational questions. The broad collection of research disciplines ensures cross-fertilization of ideas, with the central questions of educational research serving as a unifying focus. This annual conference emerges from preceding EDM workshops at the AAAI, AIED, EC-TEL, ICALT, ITS, and UM conferences. This has been an exciting year for Educational Data Mining, with the first international conference for this emerging discipline, and the launch of the Journal of Educational Data Mining (JEDM), with Kalina Yacef as journal editor. We look forward, as well, to EDM2009, which will occur in Cordoba, Spain.

Educational Data Mining (EDM) is the process of converting raw data from educational systems to useful information that can be used by educational software developers, students, teachers, parents, and other educational researchers. EDM is also an emerging discipline, concerned with developing methods for exploring the unique types of data that come from educational settings, and with using those methods to better understand students and the settings in which they learn. Whether educational data are taken from students' use of interactive learning environments, computer-supported collaborative learning, or administrative data from schools and universities, it often has multiple levels of meaningful hierarchy, which often need to be determined by properties in the data itself, rather than in advance. Issues of time, sequence, and context also play important roles in the study of educational data. The First International Conference on Educational Data Mining (EDM2008) brings together researchers from computer science, education, psychology, psychometrics, and statistics to analyze large data sets to answer educational research questions. The increase in instrumented educational software, as well as state databases of student test scores, has created large repositories of data reflecting how students learn. The EDM conference focuses on computational approaches for using those data to address important educational questions. The broad collection of research disciplines ensures cross-fertilization of ideas, with the central questions of educational research serving as a unifying focus. This annual conference emerges from preceding EDM workshops at the AAAI, AIED, EC-TEL, ICALT, ITS, and UM conferences. This has been an exciting year for Educational Data Mining, with the first international conference for this emerging discipline, and the launch of the Journal of Educational Data Mining (JEDM), with Kalina Yacef as journal editor. We look forward, as well, to EDM2009, which will occur in Cordoba, Spain.

Data mining is an interdisciplinary exercise. Statistics, database technology, machine learning, pattern recognition, artificial intelligence, and visualization, all play a role. And just as it is difficult to define sharp boundaries between these disciplines, so it is difficult to define sharp boundaries between each of them and data mining. At the boundaries, one person's data mining is another's statistics, database, or machine learning problem. ID Age Sex Marital Status Education Income 248 54 Male Married High school graduate 100000 249 ?? Female Married High school graduate 12000 250 29 Male Married Some college 23000 251 9 Male Not married Child 0 252 85 Female Not married High school graduate 19798 253 40 Male Married High school graduate 40100 Table 1.1: Examples of Data in Public Use Microdata Sample Data Sets. ID Age Sex Marital Status Education Income 254 38 Female Not married Less than 1st grade 2691 255 7 Male ?? Child 0 256 49 Male Married 11th grade 30000 257 76 Male Married Doctorate degree 30686 A typical task for this type of data would be finding relationships between different variables. For example, we might want to see how well a person's income could be predicted from the other variables. We might also be interested in seeing if there are naturally distinct groups of people, or in finding values at which variables often coincide. A subset of variables and records is available online at the Machine Learning Repository of the University of California, Irvine , www.ics.uci.edu/~mlearn/MLSummary.html. Data come in many forms and this is not the place to develop a complete taxonomy. Indeed, it is not even clear that a complete taxonomy can be developed, since an important aspect of data in one situation may be unimportant in another. However there are certain basic distinctions to which we should draw attention. One is the difference between quantitative and categorical measurements (different names are sometimes used for these). A quantitative variable is measured on a numerical scale and can, at least in principle, take any value. The columns Age and Income in table 1.1 are examples of quantitative variables. In contrast, categorical variables such as Sex, Marital Status and Education in 1.1 can take only certain, discrete values. The common three point severity scale used in medicine (mild, moderate, severe) is another example. Categorical variables may be ordinal (possessing a natural order, as in the Education scale) or nominal (simply naming the categories, as in the Marital Status case). A data analytic technique appropriate for one type of scale might not be appropriate for another (although it does depend on the objective-see Hand (1996) for a detailed discussion). For example, were marital status represented by integers (e.g., 1 for single, 2 for married, 3 for widowed, and so forth) it would generally not be meaningful or appropriate to calculate the arithmetic mean of a sample of such scores using this scale. Similarly, simple linear regression (predicting one quantitative variable as a function of others) will usually be appropriate to apply to quantitative data, but applying it to categorical data may not be wise; other techniques, that have similar objectives (to the extent that the objectives can be similar when the data types differ), might be more appropriate with categorical scales. Measurement scales, however defined, lie at the bottom of any data taxonomy. Moving up the taxonomy, we find that data can occur in various relationships and structures. Data may arise sequentially in time series, and the data mining exercise might address entire time series or particular segments of those time series. Data might also describe spatial relationships, so that individual records take on their full significance only when considered in the context of others. Consider a data set on medical patients. It might include multiple measurements on the same variable (e.g., blood pressure), each measurement taken at different times on different days. Some patients might have extensive image data (e.g., X-rays or magnetic resonance images), others not. One might also have data in the form of text, recording a specialist's comments and diagnosis for each patient. In addition, there might be a hierarchy of relationships between patients in terms of doctors, hospitals, and geographic locations. The more complex the data structures, the more complex the data mining models, algorithms, and tools we need to apply. 1.9 Further Reading Brief, general introductions to data mining are given in Fayyad, Piatetsky-Shapiro, and Smyth (1996), Glymour et al. (1997), and a special issue of the Communications of the ACM, Vol. 39, No. 11. Overviews of certain aspects of predictive data mining are given by Adriaans and Zantige (1996) and Weiss and Indurkhya (1998). Witten and Franke (2000) provide a very readable, applications-oriented account of data mining from a machine learning (artificial intelligence) perspective and Han and Kamber (2000) is an accessible textbook written from a database perspective data mining. Th ere are many texts on data mining aimed at business users, notably Berry and Linoff (1997, 2000) that contain extensive practical advice on potential business applications of data mining. Leamer (1978) provides a general discussion of the dangers of data dredging, and Lovell (1983) provides a general review of the topic. From a statistical perspective. Hendry (1995, section 15.1) provides an econometrician's view of data mining. Hand et al. (2000) and Smyth (2000) present comparative discussions of data mining and statistics. Casti (1990, 192-193 and 439) provides a briefly discusses "common folklore" stock market predictors and coincidences. Chapter 2: Measurement and Data 2.1 Introduction Our aim is to discover relationships that exist in the "real world," where this may be the physical world, the business world, the scientific world, or some other conceptual domain. However, in seeking such relationships, we do not go out and look at that domain firsthand. Rather, we study data describing it. So first we need to be clear about what we mean by data. Data are collected by mapping entities in the domain of interest to symbolic representation by means of some measurement procedure, which associates the value of a variable with a given property of an entity. The relationships between objects are represented by numerical relationships between variables. These numerical representations, the data items, are stored in the data set; it is these items that are the subjects of our data mining activities. Clearly the measurement process is crucial. It underlies all subsequent data analytic and data mining activities. We discuss this process in detail in section 2.2. We remarked in chapter 1 that the notion of "distance" between two objects is fundamental. Section 2.3 outlines distance measures between two objects, based on the vectors of measurements taken on those objects. The raw results of measurements may or may not be suitable for direct data mining. Section 2.4 briefly comments on how the data might be transformed before analysis. We have already noted that we do not want our data mining activities simply to discover relationships that are mere artifacts of the way the data were collected. Likewise, we do not want our findings to be properties of the way the data are defined: discovering that people with the same surname often live in the same household would not be a major breakthrough. In section 2.5 we briefly introduce notions of the schema of data-the a priori structure imposed on the data. No data set is perfect, and this is particularly true of large data sets. Measurement error, missing data, sampling distortion, human mistakes, and a host of other factors corrupt the data. Since data mining is concerned with detecting unsuspected patterns in data, it is very important to be aware of these imperfections-we do not want to base our conclusions on patterns that merely reflect flaws in data collection or of the recording processes. Section 2.6 discusses quality issues in the context of measurements on cases or records and individual variables or fields. Section 2.7 discusses the quality of aggregate collections of such individuals (i.e., samples). Section 2.8 presents concluding remarks, and section 2.9 gives pointers to more detailed reading.

Rare association rules are those that only appear infrequently even though they are highly associated with very specific data. In consequence, these rules can be very appropriate for using with educational datasets since they are usually imbalanced. In this paper, we explore the extraction of rare association rules when gathering student usage data from a Moodle system. This type of rule is more difficult to find when applying traditional data mining algorithms. Thus we show some relevant results obtained when comparing several frequent and rare association rule mining algorithms. We also offer some illustrative examples of the rules discovered in order to demonstrate both their performance and their usefulness in educational environments.

Several data mining algorithms use iterative optimization methods for learning predictive models. It is not easy to determine upfront which optimization method will perform best or converge fast for such tasks. In this paper, we analyze Meta Algorithms (MAs) which work by adaptively combining iterates from a pool of base optimization algorithms. We show that the performance of MAs are competitive with the best convex combination of the iterates from the base algorithms for online as well as batch convex optimization problems. We illustrate the effectiveness of MAs on the problem of portfolio selection in the stock market and use several existing ideas for portfolio selection as base algorithms. Using daily S&P500 data for the past 21 years and a benchmark NYSE dataset, we show that MAs outperform existing portfolio selection algorithms with provable guarantees by several orders of magnitude, and match the performance of the best heuristics in the pool.

In this paper, we show how using data mining algorithms can help discovering pedagogically relevant knowledge contained in databases obtained from Web-based educational systems. These findings can be used both to help teachers with managing their class, understand their students' learning and reflect on their teaching and to support learner reflection and provide proactive feedback to learners.

Data privacy is a major threat to the widespread deployment of data grids in domains such as health care and finance. We propose a novel technique for obtaining knowledge-by way of a data mining model-from a data grid, while ensuring that the privacy is cryptographically secure. To the best of our knowledge, all previous approaches for solving this problem fail in the presence of malicious participants. In this paper we present an algorithm which, in addition to being secure against malicious members, is asynchronous, involves no global communication patterns, and dynamically adjusts to new data or newly added resources. As far as we know, this is the first privacy-preserving data mining algorithm to possess these features in the presence of malicious participants. Simulations of thousands of resources prove that our algorithm quickly converges to the correct result. The simulations also prove that the effect of the privacy parameter on the convergence time is logarithmic. * We thank Intel Corporation and the Mafat Institute for Research and Development for their generous support of this research.

As a result of today's competitive business environment, companies have been trying to improve the utilization of funds effectively in their budgets for information technology investments. These companies retrieve more information with the same set of resources by means of business intelligence methods. According to Rubin (Chabrow, 2004), IT budgets are not simply declining or leveling off, rather, companies are shifting from a pure cost-cut mode to a model that emphasizes agility and efficiency. The tremendous daily growth of the company data requires more funds and investment for establishing the technologies and infrastructure necessary for gathering fast and crucial information that supports the decision-making process. This necessity gave birth to various business intelligence methods, which mainly aim to process the mass amounts of collected data from their existing application, and represent it in a way that companies can apply to their daily competitive decisions.

This application primarily concerns the implementation of business intelligence for a retail business company. The aim is to implement built-in business intelligence solutions of the Microsoft SQL Server that holds the commercial information of the company for the past three years. The customer company has already been using Microsoft products. The key items used for analyzing data are sales, momentary inventory, and logistics information.

The application can be grouped into five main areas: Building the data warehouse, constructing OLAP cubes, applying data mining algorithms on OLAP cubes, representing the results in reports with reporting services, and implementation.

This study is a part of a project (No: 200878) which is supported by Dokuz Eylül University Scientific Research Fund.

... If one describes an animal as having wings and feathers, it is easy to classify it as a bird, and if ones asks you to describe a bird, you will likely get the major ... Talbert, DA, Honeycutt, M., Smith, A., Talbert, SR: An adaptive, population-sensitive trauma triage decision support model ...

In a data warehousing process, the data preparation phase is crucial. Mastering this phase allows multidimensional analysis or the use of data mining algorithms, as well as substantial gains in terms of time and performance when performing such analyses. Furthermore, a data warehouse can require external data. The web is a prevalent data source in this context, though the data broadcasted on this medium are very heterogeneous. In this chapter, we propose a modeling process for integrating all these diverse, heterogeneous data into a unified format. Furthermore, the very schema definition provides first-rate metadata in our data warehousing context. At the conceptual level, a complex object is represented in UML as a superclass of any useful data source (databases, plain or tagged texts, images, sounds, video clips, etc.). Our logical model is an XML schema that can be described with a DTD or the XML-Schema language. Eventually, we have designed a Java prototype that transforms our m...

Mining of association rules is to find associations among data items that appear together in some transactions or business activities. As of today, algorithms for association rule mining, as well as for other data mining tasks, are mostly applied to relational databases. As XML being adopted as the universal format for data storage and exchange, mining associations from XML data

This thesis is devoted to the design of data mining algorithms for evolving data streams and for the extraction of closed frequent trees. First, we deal with each of these tasks separately, and then we deal with them together, developing classification methods for data streams containing items that are trees. In the data stream model, data arrive at high speed,

The vast majority of data mining algorithms require the setting of many input parameters. The dangers of working with parameter-laden algorithms are twofold. First, incorrect settings may cause an algorithm to fail in finding the true patterns. Second, a perhaps more insidious problem is that the algorithm may report spurious patterns that do not really exist, or greatly overestimate the significance of the reported patterns. This is especially likely when the user fails to understand the role of parameters in the data mining E. Keogh et al. process. Data mining algorithms should have as few parameters as possible. A parameter-light algorithm would limit our ability to impose our prejudices, expectations, and presumptions on the problem at hand, and would let the data itself speak to us. In this work, we show that recent results in bioinformatics, learning, and computational theory hold great promise for a parameter-light data-mining paradigm. The results are strongly connected to Kolmogorov complexity theory. However, as a practical matter, they can be implemented using any off-the-shelf compression algorithm with the addition of just a dozen lines of code. We will show that this approach is competitive or superior to many of the state-of-the-art approaches in anomaly/interestingness detection, classification, and clustering with empirical tests on time series/DNA/text/XML/video datasets. As a further evidence of the advantages of our method, we will demonstrate its effectiveness to solve a real world classification problem in recommending printing services and products.

The aim of the Code4Thought project was to deliver a tool supported methodology that would facilitate the evaluation of a software product's quality according to ISO/IEC-9126 software engineering quality standard. It was a joint collaboration between Dynacomp S.A.

Sampling of large datasets for data mining is important for at least two reasons. The processing of large amounts of data results in increased computational complexity. The cost of this additional complexity may not be justifiable. On the other hand, the use of small samples results in fast and efficient computation for data mining algorithms. Statistical methods for obtaining sufficient samples from datasets for classification problems are discussed in this paper. Results are presented for an empirical study based on the use of sequential random sampling and sample evaluation using univariate hypothesis testing and an information theoretic measure. Comparisons are made between theoretical and empirical estimates.

The productivity of a compiler development team depends on its ability not only to the design effective solutions to known code generation problems, but also to uncover potential code improvement opportunities. This paper describes a data mining tool that can be used to identify such opportunities based on a combination of hardware-profiling data and on compiler-generated counters. This data is combined into an Execution Flow Graph (EFG) and then FlowGSP, a new data mining algorithm, finds sequences of attributes associated with subpaths of the EFG. Many examples of important opportunities for code improvement in the IBM R Testarossa compiler are described to illustrate the usefulness of this data mining technique. This mining tool is specially useful for programs whose execution is not dominated by a small set of frequently executed loops. Information about the amount of space and time required to run the mining tool are also provided. In comparison with manual search through the data, the mining tool saved a significant amount of compiler development time and effort.

Many high level representations of time series have been proposed for data mining, including Fourier transforms, wavelets, eigenwaves, piecewise polynomial models, etc. Many researchers have also considered symbolic representations of time series, noting that such representations would potentiality allow researchers to avail of the wealth of data structures and algorithms from the text processing and bioinformatics communities. While many symbolic representations of time series have been introduced over the past ...

Web structure mining has been a well-researched area during recent years. Based on the observation that data on the web may change at any time in any way, some incremental data mining algorithms have been proposed to update the mining results with the corresponding changes. However, none of the existing web structure mining techniques is able to extract useful and hidden knowledge from the sequence of historical web structural changes. While the knowledge from snapshot is important and interesting, the knowledge behind the corresponding changes may be more critical and informative in some applications. In this paper, we propose a novel research area of web structure mining called web structural delta mining. The distinct feature of our research is that our mining object is the sequence of historical changes of web structure (also called web structural deltas). For web structural delta mining, we aim to extract useful, interesting, and novel web structures and knowledge considering their historical, dynamic, and temporal properties. We propose three major issues of web structural delta mining, identifying useful and interesting structures, discovering associations from structural deltas, and structural change pattern based classifier. Moreover, we present a list of potential applications where the web structural delta mining results can be used.

This chapter describes a principled approach to meta-learning that has three distinctive features. First, whereas most previous work on meta-learning focused exclusively on the learning task, our approach applies meta-learning to the full knowledge discovery process and is thus more aptly referred to as meta-mining. Second, traditional meta-learning regards learning algorithms as black boxes and essentially correlates properties of their input (data) with the performance of their output (learned model). We propose to tear open the black box and analyse algorithms in terms of their core components, their underlying assumptions, the cost functions and optimization strategies they use, and the models and decision boundaries they generate. Third, to ground meta-mining on a declarative representation of the data mining (dm) process and its components, we built a DM ontology and knowledge base using the Web Ontology Language (owl). The Data Mining Optimization Ontology (dmop, pronounced dee-mope)) provides a unified conceptual framework for analysing dm tasks, algorithms, models, datasets, workflows and performance metrics, as well as their relationships. The dm knowledge base uses concepts from dmop to describe existing data mining algorithms and their implementations in major dm software packages. Metadata collected from data mining experiments are also described in terms of concepts from the ontology and linked to algorithm and operator descriptions in the knowledge base; they are then stored in data mining experiment data bases to serve as training and evaluation data for the meta-miner. These three features together lay the groundwork for what we call deep or semantic meta-mining, i.e., dm process or workflow mining that is driven simultaneously by meta-data and by the collective expertise of data miners embodied in the data mining ontology and knowledge base. In Section 2, we review the state of the art in the fields of meta-learning and data mining ontologies; at the same time, we motivate the need for ontology-based meta-mining and distinguish our approach from related work in these two areas. Section 3 gives a detailed description of dmop, while Section 4 introduces a novel method for ontology-based discovery of generalized patterns from data mining workflows. Section 5 reports on proof-of-concept experiments conducted to gauge the efficacy of dmop-based workflow mining, and Section 6 concludes.

In this paper we report ongoing research on the Open Academic Analytics Initiative (OAAI), a project aimed at increasing college student retention by performing early detection of academic risk using data mining methods. The paper describes the goals and objectives of the OAAI, and lays out a methodological framework to develop models that can be used to perform inferential queries on student performance using open source course management system data and student academic records. Preliminary results on initial model development using several data mining algorithms for classification are presented.

Mining published articles in biology and medicine is a favored means of identifying potential biomarkers in comparison to conventional reviewing process. This is made possible by the development of public literature databases and data mining algorithms. In this article, we present a method to extract novel protein interactions from online full-text articles for biomarker discovery. By evaluating support and confidence metrics, explicit and implicit protein interactions are extracted from corpus of articles. By properly chosen minimum support and confidence, our method maximizes the identification of known interactions while minimizing the number of novel interactions. Hence, our method provides a manageable size of novel interactions for biological validation.

Natural systems exhibit random, chaotic, and multiply periodic behaviors that are driven by gravity, weather, and man-made disturbances. Modeling them on a large scale is challenging because behaviors vary discontinuously both spatially and in time. Modeling requires calibration and validation data that represent a diversity of causes and effects. Measured variables are either categorical (static) or dynamic (time series). Integrating multiple data types and reducing large numbers of variables to a select set often leads to subjective decision-making that has significant ramifications when applying state-of-the-art multi-step modeling approaches, e.g., land-use models driving finite element flow models. This paper is Part 1 of a two-part treatment that describes an alternative approach that employs a sequence of numerically optimized data mining algorithms. They include 1) signal decomposition to separate static, chaotic and periodic time series components that are attributable to d...

Generalized association rule extraction is a powerful tool to discover a high level view of the interesting patterns hidden in the analyzed data. However, since the patterns are extracted at any level of abstraction, the mined rule set may be too large to be effectively exploited in the decision making process.

This paper describes a text mining tool that performs two tasks, namely document clustering and text summarization. These tasks have, of course, their corresponding counterpart in "conventional" data mining. However, the textual, unstructured nature of documents makes these two text mining tasks considerably more difficult t han their data mining counterparts. In our system document clustering is performed by using

We describe a grid-based approach for enterprisescale data mining that leverages database technology for I/O parallelism, and on-demand compute servers for compute parallelism in the statistical computations. By enterprise-scale, we mean the highly-automated use of data mining in vertical business applications, where the data is stored on one or more relational database systems, and where a distributed architecture comprising of high-performance compute servers or a network of low-cost, commodity processors is used to improve application performance and provide the application deployment flexibility for overall workload management. The approach relies on an algorithmic decomposition of the data mining kernel on the data and compute grids, which makes it possible to exploit the parallelism on the respective grids in a simple way, while minimizing the data transfer between them. The overall approach is compatible with existing database standards for data mining task specification and results reporting, and hence external applications using these standardsbased interfaces do not have to be modified in order to realize the benefits of this grid-based approach.

Data mining algorithms, especially those used for unsupervised learning, generate a large quantity of rules. In particular this applies to the Apriori family of algorithms for the determination of association rules. It is hence impossible for an expert in the field being mined to sustain these rules. To help carry out the task, many measures which evaluate the interestingness of rules have been developed. They make it possible to filter and sort automatically a set of rules with respect to given goals. Since these measures may produce different results, and as experts have different understandings of what a good rule is, we propose in this article a new direction to select the best rules: a two-step solution to the problem of the recommendation of one or more user-adapted interestingness measures. First, a description of interestingness measures, based on meaningful classical properties, is given. Second, a multicriteria decision aid process is applied to this analysis and illustrates the benefit that a user, who is not a data mining expert, can achieve with such methods.

In this research we applied classification models for prediction of students’ performance, and cluster models for grouping students based on their cognitive styles in e-learning environment. Classification models described in this paper should help: teachers, students and business people, for early engaging with students who are likely to become excellent on a selected topic. Clustering students based on cognitive styles and their overall performance should enable better adaption of the learning materials with respect to their learning styles. The approach is tested using well-established data mining algorithms, and evaluated by several evaluation measures. Model building process included data preprocessing, parameter optimization and attribute selection steps, which enhanced the overall performance. Additionally we propose a Moodle module that allows automatic extraction of data needed for educational data mining analysis and deploys models developed in this study.

Data mining on clinical data is a challenging area in the field of medical research, aiming at predicting and discovering patterns of disease occurrence and prognosis based on detected symptoms and reported health conditions. Data mining is the process of recovering related, significant and imperative information from a copious collection of comprehensive data. The main focus of this paper is

Recent studies have demonstrated the prospects of data mining algorithms for addressing the task of seriation in pa-leontological data (ie the age-based ordering of the sites of excavation). A prominent approach is spectral ordering that computes a similarity ...

In this paper, we used data mining techniques for the automatic discovering of useful temporal abstraction in reinforcement learning. This idea was motivated by the ability of data mining algorithms in automatic discovering of structures and patterns, when applied to large data sets. The ...

The study of interictal transient events may substantially complement the analysis of seizures in the presurgical evaluation of intractable epilepsy. A comprehensive methodology of quantifying reproducibility of activation patterns in intracerebral electroencephalography signals is presented. It may be applied to various forms of transient epileptic events under the assumption that a time of occurrence may be assigned to them. In this paper, the method is used on two different forms of interictal events (interictal spikes or sharpwaves and transient bursts of fast activity). The methodology is based on signal processing and data mining algorithms and proceeds in three steps: 1) detection of transient paroxysmal events (monochannel event); 2) identification of quasisynchronous transient paroxysmal events (multichannel events); and 3) automatic extraction of similar activation patterns. Results show that the methodology allows reproducible sequential activation sets to be identified from signals recorded in four patients. Potential advantages of the method are discussed with respect to other approaches.

An innovative knowledge-based methodology for terrorist detection by using Web traffic content as the audit information is presented. The proposed methodology learns the typical behavior ('profile') of terrorists by applying a data mining algorithm to the textual content of terror-related Web sites. The resulting profile is used by the system to perform real-time detection of users suspected of being engaged in terrorist activities. The

Abstract—Most Data mining Algorithms and Tools when applied to Industrial problems such as CRM (Customer Relationship Management) are useful in pointing out Customers who are likely attritors and customers who are Loyal , but they require human experts to post ...

The problem of finding a suitable dataset to test different data mining algorithms and techniques and specifically association rule mining for Market Basket Analysis is a big challenge. A lot of dataset generators have been implemented in order to overcome this problem. ARtool is a tool that generates synthetic datasets and runs association rule mining for Market Basket Analysis. But

We consider the problem of resource allocation in mining multiple data streams. Due to the large volume and the high speed of streaming data, mining algorithms must cope with the effects of system overload. How to realize maximum mining benefits under resource constraints becomes a challenging task. In this paper, we propose a load shedding scheme for classifying multiple data streams. We focus on the following problems: i) how to classify data that are dropped by the load shedding scheme? and ii) how to decide when to drop data from a stream? We introduce a quality of decision (QoD) metric to measure the level of uncertainty in classification when exact feature values of the data are not available because of load shedding. A Markov model is used to predict the distribution of feature values and we make classification decisions using the predicted values and the QoD metric. Thus, resources are allocated among multiple data streams to maximize the quality of classification decisions. Furthermore, our load shedding scheme is able to learn and adapt to changing data characteristics in the data streams. Experiments on both synthetic data and real-life data show that our load shedding scheme is effective in improving the overall accuracy of classification under resource constraints.

Data Mining techniques are commonly used to extract patterns, like association rules and decision trees, from huge volumes of data. The comparison of patterns is a fundamental issue, which can be exploited, among others, to synthetically measure dissimilarities in evolving or different datasets and to compare the output produced by different data mining algorithms on a same dataset. In this paper, we present the PANDA framework for computing the dissimilarity of both simple and complex patterns, defined upon raw data and other patterns, respectively. In PANDA the problem of comparing complex patterns is decomposed into simpler sub-problems on the component (simple or complex) patterns and soobtained partial solutions are then smartly aggregated into an overall dissimilarity score. This intrinsically recursive approach grants PANDA with a high flexibility and allows it to easily handle patterns with highly complex structures. PANDA is built upon a few basic concepts so as to be generic and clear to the end user. We demonstrate the generality and flexibility of PANDA by showing how it can be easily applied to a variety of pattern types, including sets of itemsets and clusterings.

This paper introduces Cougar^2, an innovative open source Java framework and toolset that assists researchers in designing, developing, and using machine learning and data mining algorithms. The primary mission for Cougar^2 is to provide an intuitive API to the research community with the abstraction and flexibility necessary to allow painless extension of the core framework. The Cougar^2 framework introduces and employs the Factory, Algorithm, and Model (FAM) paradigm which represents a novel combination of established object-oriented principles, design patterns, strategic abstraction, and domain knowledge geared for any machine learning or data mining task. Cougar^2 has been used successfully for both state of the art spatial data mining research (regional knowledge discovery and clustering) and as the main development tool in a data mining graduate course over the past two years.

In the remote sensing field, a frequently recurring question is: Which computational intelligence or data mining algorithms are most suitable for the retrieval of essential information given that most natural systems exhibit very high non-linearity. Among potential candidates might be empirical regression, neural network model, support vector machine, genetic algorithm/genetic programming, analytical equation, etc. This paper compares three types of data mining techniques, including multiple non-linear regression, artificial neural networks, and genetic programming, for estimating multi-temporal turbidity changes following hurricane events at Lake Okeechobee, Florida. This retrospective analysis aims to identify how the major hurricanes impacted the water quality management in 2003-2004. The Moderate Resolution Imaging Spectroradiometer (MODIS) Terra 8-day composite imageries were used to retrieve the spatial patterns of turbidity distributions for comparison against the visual patterns discernible in the in-situ observations. By evaluating four statistical parameters, the genetic programming model was finally selected as the most suitable data mining tool for classification in which the MODIS band 1 image and wind speed were recognized as the major determinants by the model. The multi-temporal turbidity maps generated before and after the major hurricane events in 2003-2004 showed that turbidity levels were substantially higher after hurricane episodes. The spatial patterns of turbidity confirm that sediment-laden water travels to the shore where it reduces the intensity of the light necessary to submerged plants for photosynthesis. This reduction results in substantial loss of biomass during the post-hurricane period.

Generalized association rule extraction is a powerful tool to discover a high level view of the interesting patterns hidden in the analyzed data. However, since the patterns are extracted at any level of abstraction, the mined rule set may be too large to be effectively exploited in the decision making process.

Classification algorithms of data mining have been successfully applied in the recent years to predict cancer based on the gene expression data. Micro-array is a powerful diagnostic tool that can generate handful information of gene expression of all the human genes in a cell at once. Various classification algorithms can be applied on such micro-array data to devise methods that can predict the occurrence of tumor. However, the accuracy of such methods differ according to the classification algorithm used. Identifying the best classification algorithm among all available is a challenging task. In this study, we have made a comprehensive comparative analysis of 14 different classification algorithms and their performance has been evaluated by using 3 different cancer data sets. The results indicate that none of the classifiers outperformed all others in terms of the accuracy when applied on all the 3 data sets. Most of the algorithms performed better as the size of the data set is increased. We recommend the users not to stick to a particular classification method and should evaluate different classification algorithms and select the better algorithm.

To answer user queries, a data integration system employs a set of semantic mappings between the mediated schema and the schemas of data sources. In dynamic environments sources often undergo changes that invalidate the mappings. Hence, once the system is deployed, the administrator must monitor it over time, to detect and repair broken mappings. Today such continuous monitoring is extremely labor intensive, and poses a key bottleneck to the widespread deployment of data integration systems in practice. We ...