Deriving dynamic consumer defined product attributes from input queries
Methods and systems of defining product attributes may involve receiving a search query and extracting a user expectation from the search query. In addition, an attribute may be defined for a product based on the user expectation. In one example, consumer generated content such as forum content, review content, blog content and social networking content, is used to define the attribute.
1. A computer implemented method comprising:
receiving a search query;
distinguishing one or more nouns in the search query from a product in the search query;
identifying one or more synonyms associated with the one or more nouns;
extracting a user expectation from the search query by a processor based on the one or more nouns and the one or more synonyms;
mining consumer generated content for at least one of the product and the user expectation;
determining an occurrence frequency of at least one of the product and the user expectation in the consumer generated content;
generating an attribute for the product based on the user expectation if the occurrence frequency exceeds a threshold and the attribute is not present in a product catalog;
adding metadata to the attribute, wherein the metadata includes at least one of occurrence frequency data and opinion data;
adding the attribute to the product catalog;
conducting a search of the product catalog based on the search query; and
generating a result based on the search.
2. The computer implemented method of claim 1 , wherein at least one of forum content, review content, blog content and social networking content is mined for the user expectation.
3. The computer implemented method of claim 1 , wherein determining the occurrence frequency includes consulting one or more sentiment rules.
4. The computer implemented method of claim 1 , wherein extracting the user expectation includes consulting an industry-specific knowledge base.