CAREER: Learning Local Feature Relevance for Pattern Classification and Clustering

职业：学习模式分类和聚类的局部特征相关性

• 批准号: 0447814
• 负责人: Carlotta Domeniconi
• 金额: $ 40万
• 依托单位: George Mason University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2005
• 资助国家: 美国
• 起止时间: 2005-02-01 至 2011-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0447814&HistoricalAwards=false
• 关键词: CAREER; Learning; Local; Feature; Relevance

Pattern classification is a broad research area with numerous applications ranging from science, engineering, target marketing, medical diagnosis and electronic commerce to weather forecasting. While pattern classification has shown promise in many areas of practical significance, it faces difficult challenges posed by real world problems, of which the most pronounced is the so-called curse of dimensionality: that is, the fact that the sample size required to perform accurate prediction on problems with high dimensionality is beyond feasibility. The goal of this project is to develop novel classification and clustering techniques to mitigate the curse of dimensionality and to reduce bias by estimating feature relevance and selecting features accordingly. In particular, this project has the following specific, measurable objectives: (1) to develop non-linear and flexible metrics for distance-based classifiers via kernel methods; (2) to construct effective ensembles by exploiting local feature relevance to perform adaptive sampling in feature space, and (3) to develop adaptive metrics for subspace clustering by measuring local correlation of data with respect to different dimensions.Given that so many problems of practical interest are high dimensional, this research will have significant impact in fields and applications as diverse as bioinformatics, security and intrusion detection, information and image retrieval. The PI's collaborative effort with biologists for the analysis of microarray data has the potential to contribute new data mining techniques to the HIV genomic knowledge domain, which in turn could eventually contribute to the customized diagnosis and treatment of AIDS.This project involves the training of both undergraduate and graduate students and includes a joint seminar course with the Department of Molecular Biology at George Mason University. The PI aims to facilitate the involvement of underrepresented groups in the science and engineering fields through her active participation in training activities and workshops in cooperation with the Multicultural Research Center at George Mason University.

模式分类是一个广泛的研究领域，从科学、工程、目标营销、医疗诊断、电子商务到天气预报都有广泛的应用。虽然模式分类在许多具有实际意义的领域显示出了希望，但它面临着现实世界问题带来的困难挑战，其中最明显的是所谓的维数诅咒：即对高维数问题进行准确预测所需的样本量超出了可行性。该项目的目标是开发新的分类和聚类技术，以减轻维度的诅咒，并通过估计特征相关性和相应地选择特征来减少偏差。特别是，该项目具有以下具体的，可测量的目标：(1)通过核方法为基于距离的分类器开发非线性和灵活的度量；(2)利用局部特征相关性在特征空间中进行自适应采样，构建有效的集成；(3)通过测量数据相对于不同维度的局部相关性，开发子空间聚类的自适应度量。鉴于许多实际问题都是高维的，本研究将对生物信息学、安全与入侵检测、信息与图像检索等领域和应用产生重大影响。PI与生物学家在微阵列数据分析方面的合作有可能为HIV基因组知识领域贡献新的数据挖掘技术，从而最终有助于定制艾滋病的诊断和治疗。该项目包括对本科生和研究生的培训，并包括与乔治梅森大学分子生物系的联合研讨会课程。该项目旨在通过与乔治梅森大学多元文化研究中心合作积极参与培训活动和讲习班，促进代表性不足的群体参与科学和工程领域。