Back to results
US 11,636,409 B2Granted
Real-time recommendation systems and methods
AssigneeOlx Global B.V.
InventorsBhagirath Bhardwaj, Gaurav Bansal, Md Tanweer Alam
Life of this patent
Estimating the term…
Abstract
Pursuant to some embodiments, systems, methods, apparatus and computer program code are provided to provide substantially real-time recommendations to a user operating a user device to interact with job listing data.
Claims17
A method for operating a user device, comprising:
- displaying a first set of job listing data;
- monitoring, by the user device, a user's interactions with at least a first job listing in the first set of job listing data and storing information identifying at least a first interaction event associated with the user's interaction with the at least first job listing, the at least first interaction event including at least a first clickstream event and a timestamp of the at least first clickstream event;
- calculating, by the user device, weight metrics of attributes of the at least first job listing associated with the at least first interaction event, the attributes including at least a job category of the at least first interaction event and the weight metrics representing a level of interest in the at least first job listing;
- updating an event data object on the user device by storing the weight metrics and the attributes on the user device;
- transmitting information associated with the event data object including the weight metrics and the attributes to a recommendation engine as query data to cause the recommendation engine to retrieve a second set of job listing data containing job listings that match the query data;
- receiving, by the user device, the second set of job listing data;
- presenting, by the user device, the second set of job listing data to the user.
4 more elements
A computing system, comprising:
- a network interface configured to receive job listing data from a remote recommendation engine;
- and
- a processor configured to
- display a first set of job listing data to a user;
- monitor a user's interactions with at least a first job listing in the first set of job listing data, and storing information identifying at least a first interaction event associated with the user's interaction with the at least first job listing, the at least first interaction event including at least a first clickstream event and a timestamp of the at least first clickstream event;
- calculate weight metrics of attributes of the at least first job listing associated with the at least first interaction event, the attributes including at least a job category of the at least first interaction event and the weight metrics representing a level of interest in the at least first job listing;
- update an event data object on the computing system by storing the weight metrics and the attributes on the computing system;
- transmit information associated with the event data object including the weight metrics and the attributes to the remote recommendation engine as query data to cause the recommendation engine to retrieve a second set of job listing data containing job listings that match the query data;
- receive the second set of job listing data;
- and
- present the second set of job listing data to the user.
8 more elements
Dependencies are inferred from the claim text.