ITR: Information Extraction from Massive Data Sets

University Of Connecticut, Storrs CT

Investigators

Abstract

Advances in information technology have resulted in the generation of voluminous data in every walk of life. Efficient techniques are needed to process these data. In particular, tools are needed to extract useful information from massive data sets. One of the objectives of the ITR Program is "extending the capability to process, manage, and communicate information on a global scale beyond what can be imagined today". Society at large can benefit immensely from advances in this arena. For example, information extracted from biological data can result in gene identification, diagnosis for diseases, drug design, etc. Market-data information can be used for custom-designed catalogues for customers, supermarket shelving, and so on. Weather prediction and protecting the environment from pollution are possible with the analysis of atmospheric data. Estimates indicate that more than 40% of online transactions are fraudulent. Analyzing the log data can reveal information that can be used to detect fraudulent attempts. The state of the art in information extraction is the use of disparate ad-hoc application-specific techniques. For example, association-rule algorithms are used for processing market data, Sequence-analysis techniques are employed in handling biological data, etc. Unifying techniques are needed for processing data. Such unifying information extraction techniques could benefit from and be of benefit to the various communities that deal with massive data. Historically, communication among these communities has been very sparse. The proposed research will examine closely the techniques employed by these communities and to develop novel techniques that will be applicable to all kinds of data. It is anticipated that the new techniques will also vastly improve the individual techniques currently employed.