Methods and systems for a closest match search

Abstract

A system to generate an index for a closest match search is described. The system receives a corpus of information that includes member information. The system parses the member information to generate signatures for each of the members and stores the signatures in the index. The signatures are unique to the members. Accordingly, the signatures signify the respective members. The system subsequently utilizes the index to identify input information that matches signatures in the index to identify a closest match of the input information to one or more members in the corpus information.

1 . A system to generate an index for a closest match search, the system comprising:
 at least one processor and executable instructions accessible on a computer-readable medium that, when executed, cause the at least one processor to perform operations comprising:
 identify a plurality of index signatures based a plurality of member information including first member information that describes a first member, the plurality of index signatures including a first plurality of index signatures associated with a first plurality of index signature scores, the first plurality of index signatures including a first index signature, the first plurality of index signature scores including a first index signature score, the first index signature score representing a percentage of coverage of the first index signature over the first member information, the first plurality of index signatures to signify the first member and not any other members, and 
 store the first plurality of index signatures in the index to enable a closest match of input information to the first member. 
   
     
     
         2 . The system of  claim 1 , wherein the operations comprise compare a first plurality of candidate signature scores with a predetermined threshold. 
     
     
         3 . The system of  claim 1 , wherein the first plurality of index signatures includes a plurality of features, and wherein the plurality of features includes a first feature of the first member. 
     
     
         4 . The system of  claim 3 , wherein the first feature includes at least one entity, wherein the at least one entity includes a first entity that includes a string of text that is included in the first member information and delimited from other strings of text that are included in the first member information. 
     
     
         5 . The system of  claim 1  wherein the operations comprise:
 receive listing information that includes the input information; and 
 identify a closest match of the input information to at least one of the plurality of index signatures to identify the closest match of the input information to the first member over the other members. 
 
     
     
         6 . The system of  claim 1 , wherein the input information includes a string of text with at least one name-value pair. 
     
     
         7 . The system of  claim 5 , wherein the listing information is for a listing that describes an item for sale on a network-based marketplace, and wherein the input information includes a title of the listing that describes the item for sale on the network-based marketplace. 
     
     
         8 . The system of  claim 1 , wherein the member information includes a catalogue of products for sale on the network-based marketplace. 
     
     
         9 . The system of  claim 8 , wherein the first member includes a first product for sale on the network-based marketplace. 
     
     
         10 . A computer-implemented method to generate an index for a closest match search, the method comprising:
 identifying a plurality of index signatures based on a plurality of member information including first member information that describes a first member, the plurality of index signatures including a first plurality of index signatures associated with a first plurality of index signature scores, the first plurality of index signatures including a first index signature, the first plurality of index signature scores including a first index signature score, the first index signature score representing a percentage of coverage of the first index signature over the first member information, the first plurality of index signatures to signify the first member and not any of the plurality of other members; and   storing, by one or more hardware processors, the first plurality of index signatures in the index, the storing to enable a closest match of input information to the first member.   
     
     
         11 . The computer-implemented method of  claim 10 , wherein the identifying the plurality of index signatures includes comparing a first plurality of candidate signature scores with a predetermined threshold. 
     
     
         12 . The computer-implemented method of  claim 10 , wherein the first plurality of index signatures includes a plurality of features, and wherein the plurality of features includes a first feature of the first member. 
     
     
         13 . The computer-implemented method of  claim 12 , wherein the first feature includes at least one entity, wherein the at least one entity includes a first entity that includes a string of text that is included in the first member information and delimited from other strings of text that are included in the first member information. 
     
     
         14 . The computer-implemented method of  claim 10  further comprising:
 receiving listing information that includes the input information; and 
 identifying a closest match of the input information to at least one of the plurality of index signatures to identify the closest match of the input information to the first member over the other members. 
 
     
     
         15 . The computer-implemented method of  claim 10 , wherein the input information includes a string of text with at least one name-value pair. 
     
     
         16 . The computer-implemented method of  claim 14 , wherein the listing information is for a listing that describes an item for sale on a network-based marketplace, and wherein the input information includes a title of the listing that describes the item for sale on the network-based marketplace. 
     
     
         17 . The computer-implemented method of  claim 10 , wherein the member information includes a catalogue of products for sale on the network-based marketplace. 
     
     
         18 . The computer-implemented method of  claim 17 , wherein the first member includes a first product for sale on the network-based marketplace. 
     
     
         19 . A machine-readable medium having no transitory signals storing a set of instructions that, when executed by one or more processors of a machine, causes the machine to perform operations comprising:
 identifying a plurality of index signatures based on a plurality of member information including first member information that describes a first member, the plurality of index signatures including a first plurality of index signatures associated with a first plurality of index signature scores, the first plurality of index signatures including a first index signature, the first plurality of index signature scores including a first index signature score, the first index signature score representing a percentage of coverage of the first index signature over the first member information, the first plurality of index signatures to signify the first member and not any of the plurality of other members; and   storing the first plurality of index signatures in the index, the storing to enable a closest match of input information to the first member.   
     
     
         20 . The machine-readable medium of  claim 19 , wherein the identifying the plurality of index signatures includes comparing a first plurality of candidate signature scores with a predetermined threshold.