data mining thesis 2010

Dissertations/Thesis Advisor: 35. Metagenome Analysis using Expectation Maximization, Masters in Computer Science Thesis; Student: Mihir Karnik (2014) 34. Development of a Clique-based Dimensionality Reduction Technique and its Applications to Gene Expression, Masters in Computer Science Thesis; Student: Kankana Shukla (2014) 33. Classification of Diabetes Maculopathy Images based on Data-adaptive Neuro-Fuzzy Inference Classifier, Masters in Computer Science Practicum; Student: Sulaimon Ibrahim (2013) 32. Agent-based Social Network Dara Generation Using Semantic Tagging and Tag-based Social-link Algorithms, Masters in Computer Science Thesis; Student: Joshua Hitchins (2013) 31. Adaptive Grid Based Localized Learning for Multidimensional Dara, Ph.D. in Computational Analysis and Modeling Dissertation; Student: Sheetal Saini (2012) 30. Folksonomy based Ad Hoc Community Detection in Online Social Networks, Masters in Computer Science Thesis; Student: Vasanth Raghu Nair (2012) 29. Data Adaptive Rule Based Classification System, Masters in Computer Science Thesis; Student: Mohit Jain (2012) 28. Data Mining based Learning Algorithms for Semi-supervised Object Identification and Tracking; Ph.D. in Computational Analysis and Modeling Dissertation; Student: Michael P. Dessauer (2010) 27. Associative Pattern Mining for Supervised Learning; Ph.D. in Computational Analysis and Modeling Dissertation; Student: Harpreet Singh (2010) 26. Unsupervised Similarity Mining in High Dimensional Data; MS-CS Thesis; Afolabi Olomola (2010) 25. Automated Valuation Models Using Data Mining Techniques- An Application to Real Estate Valuation; MS-CS, Student: Preet K. Sekhon (2009) 24. An Integrated Approach for Identification of Cell-cyclic Genes in the Saccharomyces Cerevisiae; MS-CS Thesis, Student: Alan E. Alex (2009) 23. Integrated Mining of Feature Spaces for Bioinformatics Domain.
At the conference ACM SIGKDD, the first one and probably the most famous venue for data mining research, the best Ph.D dissertation award is selected each year. This year, we are proud to announce that our former Ph.D student Loïc Cerf ( Constraint-Based Mining of Closed Patterns in Noisy N-Ary Relations , Ph. D Thesis Lyon, July 2010)) has been nominee for this award by a prestigious committee. Even thoughhe did not got the award, its citation at the conference has shed light on the data mining research at LIRIS (ComBiNING team).To know more .
Yang, Jianhua (2010) Intelligent data mining using artificial neural networks and genetic algorithms : techniques and applications. PhD thesis, University of Warwick. Official URL: Abstract Data Mining (DM) refers to the analysis of observational datasets to findrelationships and to summarize the data in ways that are both understandableand useful. Many DM techniques exist. Compared with other DM techniques,Intelligent Systems (ISs) based approaches, which include Artificial NeuralNetworks (ANNs), fuzzy set theory, approximate reasoning, and derivative-freeoptimization methods such as Genetic Algorithms (GAs), are tolerant ofimprecision, uncertainty, partial truth, and approximation. They provideflexible information processing capability for handling real-life situations. Thisthesis is concerned with the ideas behind design, implementation, testing andapplication of a novel ISs based DM technique. The unique contribution of thisthesis is in the implementation of a hybrid IS DM technique (Genetic NeuralMathematical Method, GNMM) for solving novel practical problems, thedetailed description of this technique, and the illustrations of severalapplications solved by this novel technique.GNMM consists of three steps: (1) GA-based input variable selection, (2) Multi-Layer Perceptron (MLP) modelling, and (3) mathematical programming basedrule extraction. In the first step, GAs are used to evolve an optimal set of MLPinputs. An adaptive method based on the average fitness of successivegenerations is used to adjust the mutation rate, and hence theexploration/exploitation balance. In addition, GNMM uses the elite group andappearance percentage to minimize the randomness associated with GAs. Inthe second step, MLP modelling serves as the core DM engine in performingclassification/prediction tasks. An Independent Component Analysis.
Dissertation for the degree of Doctor of Science in Technology to be presented with due permission of the Faculty of Information and Natural Sciences for public examination and debate in Auditorium E at the Aalto University School of Science and Technology (Espoo, Finland) on the 23rd of January, 2010, at 12 noon. Dissertation in PDF format (ISBN 978-952-60-3004-3)   [1837 KB] Dissertation is also available in print (ISBN 978-952-60-3003-6) Abstract The idea of frequent pattern discovery is to find frequently occurring events in large databases. Such data mining techniques can be useful in various domains. For instance, in recommendation and e-commerce systems frequently occurring product purchase combinations are essential in user preference modeling. In the ecological domain, patterns of frequently occurring groups of species can be used to reveal insight into species interaction dynamics. Over the past few years, most frequent pattern mining research has concentrated on efficiency (speed) of mining algorithms. However, it has been argued within the community that while efficiency of the mining task is no longer a bottleneck, there is still an urgent need for methods that derive compact, yet high quality results with good application properties. The aim of this thesis is to address this need. The first part of the thesis discusses a new type of tree pattern class for expressing hierarchies of general and more specific attributes in unstructured binary data. The new pattern class is shown to have advantageous properties, and to discover relationships in data that cannot be expressed alone with the more traditional frequent itemset or association rule patterns. The second and third parts of the thesis discuss the use of entropy as a score measure for frequent pattern mining. A new pattern class is defined, low-entropy sets, which allow to express more general types.