Date of Award
2013
Publication Type
Master Thesis
Degree Name
M.Sc.
Department
Computer Science
Keywords
Web studies, Information science, Computer science
Supervisor
Joan Morrisey
Rights
info:eu-repo/semantics/openAccess
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Abstract
To improve the quality of the search result returned by the internet which makes users have to look through a huge amount of links for the real answers, we utilized the high quality links Google produces and the Information Retrieval technology to implement a Question Answering (QA) system. This system analyzes and downloads the text contents from the relevant web pages Google searches based on the users' questions to build a dynamic knowledge collection; retrieves the relevant passages from the collection and sends the ranked passages back. The users can further refine their questions in the query refinement step for the better answers. A novel search strategy was designed to detect the semantic connections between the question and the documents. This answer retrieval also involves the TF-IDF algorithm and Vector Space Model for the document indexing. We have modified the original Cosine Coefficient Similarity Measurement to rank the candidate answers.
Recommended Citation
Zhao, Ruoxuan, "Improving Retrieval of Information from the Internet" (2013). Electronic Theses and Dissertations. 4766.
https://scholar.uwindsor.ca/etd/4766