Statistical learning theory

About: Statistical learning theory is a research topic. Over the lifetime, 1618 publications have been published within this topic receiving 158033 citations.

Automatic Classification of Schizophrenia using Resting-state Functional Language Network via An Adaptive Learning Algorithm

Abstract: A reliable and precise classification of schizophrenia is significant for its diagnosis and treatment of schizophrenia. Functional magnetic resonance imaging (fMRI) is a novel tool increasingly used in schizophrenia research. Recent advances in statistical learning theory have led to applying pattern classification algorithms to access the diagnostic value of functional brain networks, discovered from resting state fMRI data. The aim of this study was to propose an adaptive learning algorithm to distinguish schizophrenia patients from normal controls using resting-state functional language network. Furthermore, here the classification of schizophrenia was regarded as a sample selection problem where a sparse subset of samples was chosen from the labele d training set. Using these selected samples, which we call informative vectors, a classifier for the clinic diagnosis of schizophrenia was established. We experimentally demonstrated that the proposed algorithm incorporating resting-state functional language network achieved 83.6% leave-one-out accuracy on resting-state fMRI data of 27 schizophrenia patients and 28 normal controls. In contrast with K-Nearest-Neighbor (KNN), Support Vector Machine (SVM) and l1-norm, our method yielded better classification performance. Moreover, our results suggested that a dysfunction of resting-state functional language network plays an important role in the clinic diagnosis of schizophrenia. Keywords: Schizophrenia, fMRI, resting-state, functional connectivity, language network, pattern classification, sample selection, informative vector

High dimensional probability VI : the Banff volume

Abstract: This is a collection of papers by participants at High Dimensional Probability VI Meeting held from October 9-14, 2011 at the Banff International Research Station in Banff, Alberta, Canada. High Dimensional Probability (HDP) is an area of mathematics that includes the study of probability distributions and limit theorems in infinite-dimensional spaces such as Hilbert spaces and Banach spaces. The most remarkable feature of this area is that it has resulted in the creation of powerful new tools and perspectives, whose range of application has led to interactions with other areas of mathematics, statistics, and computer science. These include random matrix theory, nonparametric statistics, empirical process theory, statistical learning theory, concentration of measure phenomena, strong and weak approximations, distribution function estimation in high dimensions, combinatorial optimization, and random graph theory. The papers in this volume show that HDP theory continues to develop new tools, methods, techniques and perspectives to analyze the random phenomena. Both researchers and advanced students will find this book of great use for learning about new avenues of research.

System for dynamic recommendation using statistical learning theory by analyzing transaction pattern, and offering the dynamic recommendation

Abstract: PURPOSE: A system for a dynamic recommendation using a statistical learning theory is provided to make a user access the desired information efficiently by offering the dynamic recommendation based on the transaction pattern information of the user connecting to a web site. CONSTITUTION: A web log file stores the connection information generated when the user connects to the web site. A database classifies/stores the response and the propensity of the users connecting to the web site. A recommender generates/recommends the recommendation information by extracting the response for the predetermined information of a new user from the web log file when the new user connects to the web site, understanding the propensity from the response, and making a model based on the response of a group matched with the propensity of the new user with use of the statistical learning theory.

Research on Multi-class Classification of Support Vector Machine

Abstract: Support vector machine(SVM) is a new learning method based on statistical learning theory,which can effectively solve the over study problem by using structural risk minimization(SRM) and has better generalization performance.Traditional SVM is developed for binary classification problems,in order to analyze huge and multi-category data for practical problems,a comparison result about the classification speed and accuracy is given through analyzing the theory and realization method of all-together,one-against-rest,one-against-one and directed acyclic graph sup-port vector machine(DAGSVM).Experimental results show that various methods can get different classifier generalization ability,training speed and test speed.The direction of how to solve multi-class classification effectively is proposed.