Mining “Hidden Phrase” Definitions from the Web

Nguyen, Hung Viet; Velamuru, P.; Kolippakkam, Deepak; Davulcu, Hasan; Liu, H.; Ates, Mikaël

doi:10.1007/3-540-36901-5_17

“…In [15], authors present a co-occurrence clustering algorithm that identifies phrases that frequently co-occurs with the target phrase from the meta-tags of Web documents. However, in this paper we address a different problem; we attempt to mine the phrase definitions in terms of extracted item information, thus, the mined definitions can be utilized to connect "search phrases" to real items in all their nuances.…”

Section: Related Workmentioning

confidence: 99%

Boosting Item Findability: Bridging the Semantic Gap Between Search Phrases and Item Information

2005

Proceedings of the Seventh International Conference on Enterprise Information Systems

1

0

View full text Add to dashboard Cite

Most search engines do their text query and retrieval based on keyword phrases. However, publishers cannot anticipate all possible ways in which users search for the items in their documents. In fact, many times, there may be no direct keyword match between a search phrase and descriptions of items that are perfect "hits" for the search. We present a highly automated solution to the problem of bridging the semantic gap between item information and search phrases. Our system can learn rulebased definitions that can be ascribed to search phrases with dynamic connotations by extracting structured item information from product catalogs and by utilizing a frequent itemset mining algorithm. We present experimental results for a realistic e-commerce domain. Also, we compare our rule-mining approach to vectorbased relevance feedback retrieval techniques and show that our system yields definitions that are easier to validate and perform better.

show abstract

“…An episode is a collection of feature vectors with a partial order; authors claimed that their approach is useful in phrase mining in Finnish, a language that has the relaxed order of words in a sentence. In [15], authors present a co-occurrence clustering algorithm that identifies phrases that frequently co-occurs with the target phrase from the meta-tags of Web documents. However, in this paper we address a different problem; we attempt to mine the phrase definitions in terms of extracted item information, thus, the mined definitions can be utilized to connect "search phrases" to real items in all their nuances.…”

Section: Related Workmentioning

confidence: 99%

Boosting Item Findability: Bridging the Semantic Gap Between Search Phrases and Item Information

Davulcu¹,

Nguyen

²

,

Ramachandran

³

Enterprise Information Systems VII

View full text Add to dashboard Cite

Most search engines do their text query and retrieval based on keyword phrases. However, publishers cannot anticipate all possible ways in which users search for the items in their documents. In fact, many times, there may be no direct keyword match between a search phrase and descriptions of items that are perfect "hits" for the search. We present a highly automated solution to the problem of bridging the semantic gap between item information and search phrases. Our system can learn rulebased definitions that can be ascribed to search phrases with dynamic connotations by extracting structured item information from product catalogs and by utilizing a frequent itemset mining algorithm. We present experimental results for a realistic e-commerce domain. Also, we compare our rule-mining approach to vectorbased relevance feedback retrieval techniques and show that our system yields definitions that are easier to validate and perform better.

show abstract