Sentiment Extraction From Consumer Reviews For Providing Product Recommendations

Abstract

A system and method for recommending a product to a user in response to a query for a product with a feature wherein the recommendation is accompanied by a quotation expressing a sentiment about the feature or the product.

I claim: 
     
         1 . A method for recommending a product to a user comprising:
 receiving a search query;   determining a feature of the query, wherein a feature comprises a characteristic of the product;   determining the product corresponding to the feature wherein determining the product corresponding to the feature comprises:
 building a syntactic template from the query, and; 
 searching a recommendation database with the syntactic template; 
   retrieving the product and a quotation relevant to the product and the feature wherein the quotation comprises expression of a sentiment about the feature or the product, and;   outputting the product and the quotation to the user.   
     
     
         2 . The method of  claim 1  wherein determining a feature of the query comprises applying a semantic template to the query. 
     
     
         3 . The method of  claim 2  wherein the semantic template is applied to the query by a span query. 
     
     
         4 . The method of  claim 1  wherein searching a recommendation database with the syntactic template comprises building a span query from the syntactic template. 
     
     
         5 . A method for recommending a product to a user comprising:
 receiving a search query;   determining a feature of the query wherein a feature comprises a characteristic of a product;   determining a product corresponding to the feature;   retrieving the product and a quotation relevant to the product and the feature wherein the quotation comprises expression of a sentiment about the feature or the product, and;   outputting the product and the quotation to the user.   
     
     
         6 . A method for extracting quotations from a document, the quotations comprising a topic and a sentiment about the topic, comprising:
 receiving a document;   building a span query relevant to the topic and sentiment;   applying the span query to the document; and   storing the result of the span query wherein the result comprises a quotation comprising a topic and a sentiment about the topic.   
     
     
         7 . The method of  claim 6  wherein building a span query comprises:
 building syntactic template from a lexicography relevant to the topic and sentiment; and 
 determining a first semantic template relevant to the syntactic template. 
 
     
     
         8 . The method of  claim 6  further comprising apply a heuristic model in addition to the span query to the document. 
     
     
         9 . The method of  claim 7  wherein determining a first semantic template comprises resolving a conflict between a second and a third semantic template. 
     
     
         10 . The method of  claim 7  wherein resolving a conflict comprises applying defeasible logic programming. 
     
     
         11 . The method of  claim 6  wherein the document is web page. 
     
     
         12 . A computer program product, comprising:
 a computer usable medium having computer readable program code embodied therein to recommend a product to a user, the computer readable program code in the computer program product comprising:   computer readable program code to receive a search query;   computer readable program code to determine a feature of the query, wherein a feature comprises a characteristic of the product;   computer readable program code to determine the product corresponding to the feature wherein the computer readable program code to determine the product corresponding to the feature comprises:
 computer readable program code to build a syntactic template from the query, and; 
 computer readable program code to search a recommendation database with the syntactic template; 
   computer readable program code to retrieve the product and a quotation relevant to the product and the feature wherein the quotation comprises expression of a sentiment about the feature or the product, and;   computer readable program code to output the product and the quotation to the user.   
     
     
         13 . The computer program product of  claim 12  wherein the computer readable program code to determine a feature of the query comprises computer readable program code to apply a semantic template to the query. 
     
     
         14 . The computer program product of  claim 13  wherein the computer readable program code to apply a semantic template comprises computer readable program code to apply the semantic template to the query with a span query. 
     
     
         15 . The computer program product of  claim 12  wherein the computer readable program code to search a recommendation database with the syntactic template comprises computer readable program code to build a span query from the syntactic template. 
     
     
         16 . A computer program product comprising:
 a computer usable medium having computer readable program code embodied therein to recommend a product to a user, the computer readable program code in the computer program product comprising:   computer readable program code to receive a search query;   computer readable program code to determine a feature of the query wherein a feature comprises a characteristic of a product;   computer readable program code to determine a product corresponding to the feature;   computer readable program code to retrieve the product and a quotation relevant to the product and the feature wherein the quotation comprises expression of a sentiment about the feature or the product, and;   computer readable program code to output the product and the quotation to the user.   
     
     
         17 . A computer program product comprising:
 a computer usable medium having computer readable program code embodied therein to extract quotations from a document, the quotations comprising a topic and a sentiment about the topic, the computer readable program code in the computer program product including:   computer readable program code to receive a document;   computer readable program code to build a span query relevant to the topic and sentiment;   computer readable program code to apply the span query to the document; and   computer readable program code to store the result of the span query wherein the result comprises a quotation comprising a topic and a sentiment about the topic.   
     
     
         18 . The computer program product of  claim 17  wherein the computer readable program code to build a span query comprises:
 computer readable program code to build syntactic template from a lexicography relevant to the topic and sentiment; and 
 computer readable program code to determine a first semantic template relevant to the syntactic template. 
 
     
     
         19 . The computer program product of  claim 17  wherein the computer readable program code further comprises computer readable program code to apply a heuristic model to the document. 
     
     
         20 . The computer program product of  claim 18  wherein the computer readable program code to determine a first semantic template comprises computer readable program code to resolve a conflict between a second and a third semantic template. 
     
     
         21 . The computer program product of  claim 18  wherein the computer readable program code to resolve a conflict comprises computer readable program code to apply defeasible logic programming.