Item listing categorization system

Abstract

Techniques for categorizing item listings based on parsing item listing titles are described. According to various embodiments, listing titles of one or more item listings on a marketplace website are accessed, the item listings being associated with a particular product category in a product category structure of the marketplace website. Words in each of the listing titles may then be converted to semantic tokens in a token symbol space, based on a tokenization process. Thereafter, n-gram modeling may be performed on the tokens corresponding to each of the listing titles of the item listings in the particular product category. One or more dominant n-gram models associated with the listing titles of the item listings in the particular product category may then be identified.

1 . (canceled) 
     
     
         2 . A computer-implemented method comprising:
 receiving a user specification of a publication title in connection with a user request to post a publication on a web site;   identifying, by at least one processor, a dominant n-gram model associated with the publication title, the dominant n-gram model identifying a sequence of one or more semantic tokens that appear in a statistically significant portion of the publication title in a particular category;   associating the publication title with the particular category based on the dominant n-gram model; and   transmitting a notification that the publication is being assigned to the particular category to the user posting the publication.   
     
     
         3 . The method of  claim 2 , wherein the tokens include a product token, an accessory token, a for token, a with token, and an unknown token. 
     
     
         4 . The method of  claim 2 , wherein the publication is an item listing. 
     
     
         5 . The method of  claim 4 , further comprising matching the item listing title with the dominant n-gram model associated with the particular product category. 
     
     
         6 . The method of  claim 2 , wherein the publication title further includes a user specification of a second category distinct from the particular category, and wherein the method further comprises notifying the user that the publication title is associated with the particular category and is not associated with the second category. 
     
     
         7 . The method of  claim 4 , wherein the particular category is an accessory items category. 
     
     
         8 . The method of  claim 7 , wherein one of the dominant n-gram models for the accessory items category includes a product token followed by a for token followed by an accessory token. 
     
     
         9 . The method of  claim 7 , wherein one of the dominant n-gram models for the accessory items category includes a product token followed by an accessory token. 
     
     
         10 . The method of  claim 2 , wherein the particular category is a bundled items category. 
     
     
         11 . The method of  claim 10 , wherein one of the dominant n-gram models for the bundled items category includes a product token followed by a with token followed by an accessory token. 
     
     
         12 . A system comprising:
 a processor and executable instructions accessible on a computer-readable medium that, when executed, cause the processor to perform operations comprising:
 receive a user specification of a publication title in connection with a user request to post a publication on a website; 
 identify, by at least one processor, a dominant n-gram model associated with the publication title, the dominant n-gram model identifying a sequence of one or more semantic tokens that appear in a statistically significant portion of the publication title in a particular category; 
 associate the publication title with the particular category based on the dominant n-gram model; and 
 transmit a notification that the publication is being assigned to the particular category to the user posting the publication. 
   
     
     
         13 . The method of  claim 12 , wherein the tokens include a product token, an accessory token, a for token, a with token, and an unknown token. 
     
     
         14 . The method of  claim 12 , wherein the publication is an item listing. 
     
     
         15 . The method of  claim 14 , further comprising matching the item listing title with the dominant n-gram model associated with the particular product category. 
     
     
         16 . The method of  claim 12 , wherein the publication title further includes a user specification of a second category distinct from the particular category, and wherein the method further comprises notifying the user that the publication title is associated with the particular category and is not associated with the second category. 
     
     
         17 . The method of  claim 14 , wherein the particular category is an accessory items category. 
     
     
         18 . The method of  claim 17 , wherein one of the dominant n-gram models for the accessory items category includes a product token followed by a for token followed by an accessory token. 
     
     
         19 . The method of  claim 17 , wherein one of the dominant n-gram models for the accessory items category includes a product token followed by an accessory token. 
     
     
         20 . The method of  claim 12 , wherein the particular category is a bundled items category. 
     
     
         21 . The method of  claim 20 , wherein one of the dominant n-gram models for the bundled items category includes a product token followed by a with token followed by an accessory token.