Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: English
- Document No: 41970 (52)
- University: Sharif University of Technology, International Campus, Kish Island
- Department: Science and Engineering
- Advisor(s): Sadighi Moshkinani, Mohsen
- Abstract:
- The rapid growth of information on the web and the need for information sharing on one hand and also as news plays an important role in our life and internet becomes the biggest repository for keeping this news on the other hand, lead us to research in this domain.
In this thesis, we introduce a new framework for searching news by considering the relation between news and events. This framework called NewsSe. NewsSe considers news as a series of events in order to cover all aspects of news. NewsSe uses Domain Ontology and Event Ontology to extract the concepts and relations existed in news. NewsSe consists of 4 different modules. NewsCr is a crawler which uses a new methodology for decreasing the number of crawls. NewsDex is an indexer which uses a new algorithm for indexing by focusing on the concepts and events related to news. NewsRank introduces a new ranking algorithm which is used just for ranking news. NewsQ is a query processor which uses a new algorithm for enriching the user’s queries. The heart of this new algorithm is a rule-based Inference Engine.Experience in using semantic techniques in search engines on one hand and considering the relationships between events related to each piece of news on the other hand cause an improvement in ranking results, obtaining more fresh results and cover more concepts related to a particular query compared to other search engines such as Google News and Yahoo News. This study is evaluated by some measures such as Mean Average Precision (MAP), Freshness and also compared with previous expressed search engines by using human evaluation methods - Keywords:
- Semantic Search Engine ; Ontology-Based Search Engine ; Inference Engine
Digital Object List-
محتواي پايان نامه - view
Bookmark