Abstract
Latent relational search is a new search paradigm based on the proportional analogy between two entity pairs. A latent relational search engine is expected to return the entity "Paris" as an answer to the question mark(?) in the query {(Japan, Tokyo),(France, ?)} because the relation between Japan and Tokyo is high lysimilarto that between France and Paris. We propose amethod for extracting entity pairs from a text corpus to buildan index for a high speed latent relational search engine. By representing there lation between two entities in an entity pair usinglexical patterns, the proposed latent relational search engine can precisely measure the relational similarity between two entity pairs and can therefore accurately rank the result list. We evaluate the system using a Web corpus and compare the performance with an existing relational search engine. The results show that the proposed method achieves high precision and MRR while requiring small query processing time. Inparticular, the proposed method achieves an MRR of 0.963 and it retrieves correct answer in the Top 1 result for 95% of queries.
Original language | English |
---|---|
Pages (from-to) | 307-312 |
Number of pages | 6 |
Journal | Transactions of the Japanese Society for Artificial Intelligence |
Volume | 26 |
Issue number | 2 |
DOIs | |
Publication status | Published - 2011 |
Externally published | Yes |
Keywords
- Analogical search
- Latent relational search
- Relational similarity
- Relational web search
ASJC Scopus subject areas
- Artificial Intelligence
- Software