Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy (PhD)

Degree Program

Computer Science

Year Degree Awarded

Fall 2014

First Advisor

W. Bruce Croft

Subject Categories

Artificial Intelligence and Robotics

Abstract

Natural language expressions are more familiar to users than choosing keywords for queries. Given that, people can use natural language expressions to represent their sophisticated information needs. Instead of listing keywords, verbose queries are expressed in a grammatically well-formed phrase or sentence in which terms are used together to represent the more specific meanings of a concept, and the relationships of these concepts are expressed by function words.

The goal of this thesis is to investigate methods of using the semantic and syntactic features of natural language queries to maximize the effectiveness of search. For this purpose, we propose the synchronous framework in which we use syntactic parsing techniques for modeling term dependencies. We use the Generative Relevance Hypothesis (GRH) to evaluate valid variations in dependence relationships between queries and documents. This is one of the first results demonstrating that dependency parsing can be used to improve retrieval effectiveness.

We propose a method for classifying concepts in verbose queries as key concepts and secondary concepts that are used in the statistical translation model for query term expansion. Key concepts are the most important terms of queries. We use key concepts as the context for translating terms. Although secondary (key) concepts are not as important as key concepts, they are still important because they provide clues about what kinds of information users are looking for. Using concept classification results, we elaborate a translation model in which the key concepts of queries are used as the context of translation. The secondary concepts of queries are used to selectively apply the translation model to query terms.

We define the important new task of focused retrieval of answer passages that aims to immediately provide answers for users' information needs while the length of answer passage should be suitable for restricted search environments such as mobile devices and voice-based search systems.

Share

COinS