US2019087879A1PendingUtilityA1

Marketplace listing analysis systems and methods

Assignee: EBAY INCPriority: Feb 8, 2012Filed: Nov 16, 2018Published: Mar 21, 2019
Est. expiryFeb 8, 2032(~5.5 yrs left)· nominal 20-yr term from priority
G06Q 30/0641G06Q 30/0601G06Q 30/00
63
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

Improved searching for network-based publication systems is provided via improved item listings. A set of rules is obtained that include string similarity functions generated from character patterns that contain unconventional character sequences not included in user search queries. Marketplace listing information is obtained from a user, where the marketplace listing information includes a listing title. The listing title is analyzed using the set of rules to identify at least one character pattern in the listing title based on a similarity between the listing title and the at least one character pattern determined using at least one of the string similarity functions. A request for a new listing title is generated based on identifying the at least one character pattern in the listing title.

Claims

exact text as granted — not AI-modified
The invention claimed is: 
     
         1 . A machine readable medium storing instructions that, when used by a machine, cause the machine to perform operations, the operations comprising:
 obtaining a set of rules that include string similarity functions generated from a plurality of character patterns that contain unconventional character sequences not included in user search queries;   obtaining marketplace listing information from a user, the marketplace listing information including a listing title;   analyzing the listing title using the set of rules to identify at least one character pattern in the listing title based on a similarity between the listing title and the at least one character pattern determined using at least one of the string similarity functions; and   generating a request for a new listing title based on identifying the at least one character pattern in the listing title.   
     
     
         2 . The machine readable medium of  claim 1 , wherein the operations further comprise:
 extracting features from the listing title.   
     
     
         3 . The machine readable medium of  claim 1 , wherein the request for a new listing title includes a reason for requesting the new listing title. 
     
     
         4 . The machine readable medium of  claim 1 , wherein the operations further comprise:
 identifying new character patterns that contain unconventional character sequences not included in user search queries; and   updating the plurality of character patterns by adding the new character patterns.   
     
     
         5 . The machine readable medium of  claim 4 , wherein operations further comprise:
 updating the set of rules based on the updated plurality of character patterns.   
     
     
         6 . The machine readable medium of  claim 4 , wherein the new character patterns are identified by analyzing new marketplace listings for additional examples of character patterns that contain unconventional character sequences not included in user search queries 
     
     
         7 . The machine readable medium of  claim 1 , wherein the plurality of character patterns are identified by:
 generating a seed set of character patterns that contain unconventional character sequences not included in user search queries;   identifying common features of the seed set; and   applying one or more machine learning algorithms to identify additional character patterns not contained in the seed set.   
     
     
         8 . A computer-implemented method comprising:
 obtaining a set of rules that include string similarity functions generated from a plurality of character patterns that contain unconventional character sequences not included in user search queries;   obtaining marketplace listing information from a user, the marketplace listing information including a listing title;   analyzing, using one or more processors, the listing title using the set of rules to identify at least one character pattern in the listing title based on a similarity between the listing title and the at least one character pattern determined using at least one of the string similarity functions; and   generating a request for a new listing title based on identifying the at least one character pattern in the listing title.   
     
     
         9 . The method of  claim 8 , wherein the method further comprises:
 extracting features from the listing title.   
     
     
         10 . The method of  claim 8 , wherein the request for a new listing title includes a reason for requesting the new listing title. 
     
     
         11 . The method of  claim 8 , wherein the method further comprises:
 identifying new character patterns that contain unconventional character sequences not included in user search queries; and   updating the plurality of character patterns by adding the new character patterns.   
     
     
         12 . The method of  claim 11 , wherein method further comprises:
 updating the set of rules based on the updated plurality of character patterns.   
     
     
         13 . The method of  claim 11 , wherein the new character patterns are identified by analyzing new marketplace listings for additional examples of character patterns that contain unconventional character sequences not included in user search queries 
     
     
         14 . The method of  claim 8 , wherein the plurality of character patterns are identified by:
 generating a seed set of character patterns that contain unconventional character sequences not included in user search queries;   identifying common features of the seed set; and   applying one or more machine learning algorithms to identify additional character patterns not contained in the seed set.   
     
     
         15 . A system comprising:
 at least one module, executing on one or more computer processors, to:   obtain a set of rules that include string similarity functions generated from a plurality of character patterns that contain unconventional character sequences not included in user search queries;   obtain marketplace listing information from a user, the marketplace listing information including a listing title;   analyze the listing title using the set of rules to identify at least one character pattern in the listing title based on a similarity between the listing title and the at least one character pattern determined using at least one of the string similarity functions; and   generate a request for a new listing title based on identifying the at least one character pattern in the listing title.   
     
     
         16 . The system of  claim 15 , wherein the at least one module is further to extract features from the listing title. 
     
     
         17 . The system of  claim 15 , wherein the request for a new listing title includes a reason for requesting the new listing title. 
     
     
         18 . The system of  claim 15 , wherein the at least one module is further to:
 identify new character patterns that contain unconventional character sequences not included in user search queries; and   update the plurality of character patterns by adding the new character patterns.   
     
     
         19 . The system of  claim 18 , wherein the new character patterns are identified by analyzing new marketplace listings for additional examples of character patterns that contain unconventional character sequences not included in user search queries 
     
     
         20 . The system of  claim 15 , wherein the plurality of character patterns are identified by:
 generating a seed set of character patterns that contain unconventional character sequences not included in user search queries;   identifying common features of the seed set; and   applying one or more machine learning algorithms to identify additional character patterns not contained in the seed set.

Join the waitlist — get patent alerts

Track US2019087879A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.