Method of Recommending Items to a User Based on User Interest

• US 20120078876A1
• Filed: 12/09/2011
• Published: 03/29/2012
• Est. Priority Date: 05/09/2006
• Status: Active Grant

• Current Assignee: TRIAD National Security, LLC
• Original Assignee: Los Alamos National Security LLC (Government of the United States of America)
• Inventors: Bollen, Johan, Van De Sompel, Herbert
• Granted Patent: US 8,577,831 B2
• Time in Patent Office: Days
• US Class Current: 707/709
• CPC Class Codes: G06F 16/38 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/9566 URL specific, e.g. using aliases, detecting broken or misspelled links

Abstract

Although recording of usage data is common in scholarly information services, its exploitation for the creation of value-added services remains limited due to concerns regarding, among others, user privacy, data validity, and the lack of accepted standards for the representation, sharing and aggregation of usage data. A technical, standards-based architecture for sharing usage information is presented. In this architecture, OpenURL-compliant linking servers aggregate usage information of a specific user community as it navigates the distributed information environment that it has access to. This usage information is made OAI-PMH harvestable so that usage information exposed by many linking servers can be aggregated to facilitate the creation of value-added services with a reach beyond that of a single community or a single information service.

Claims

1-17. -17. (canceled)

18. A method of recommending items to a user comprising:

identifying and displaying one or more items that is known to be of interest to a user based on a query input by the user;

aggregating usage information into a superlog, wherein the usage information is online transactions involving items in Digital Libraries and performed by users from worldwide institutions;

recording the usage information by linking servers used by these institutions;

deriving a network representation of said superlog by representing each item involved in a transaction as a node on a graph and analyzing the contained transactions to derive relationships between nodes;

performing computations on the one or more items known to be of interest to the user in relation to the network representation of the superlog with the objective of obtaining a result for the identified item; and

displaying a result based on the one or more items known to be of interest to the user, wherein said result consists of a listing of item-level recommendations based on the usage information,

wherein;

the aggregating step consists of aggregating usage information using a metadata harvester;

said network representation comprises connections between said nodes based on co-retrieval of items represented by said nodes, said connections being represented as weighted edges between said nodes;

said superlog and the network representation of said superlog is derived only from usage data; and

computations on the one or more items known to be of interest to the user with the objective of obtaining the result are performed using local neighborhood search algorithms incorporating a spreading activation method.

19. A method as recited in claim 18, wherein said network representation comprises a plurality of nodes;

and wherein each node represents an item.

20. A method as recited in claim 18, wherein said weighted edges are generated using a data mining technique.

21. A method as recited in claim 20, wherein said data mining technique is the Retrieval Coherence Assumption.

22. A method as recited in claim 18, wherein said usage information is obtained in an XML format.

23. A method as recited in claim 18, wherein the aggregating step is performed using the OAI-PMH protocol.

24. An apparatus, comprising:

a special purpose log database;

a data harvester;

at least one linking server, which is able to track sequences of requests across a variety of information services;

wherein each said linking server comprises;

a log repository;

said log repository comprising usage data, wherein usage data is online transactions involving items in Digital Libraries and performed by users from worldwide institutions;

wherein said data harvester collects usage data from a plurality of said linking servers and a protocol for collecting usage data consists of using a metadata harvester; and

;

wherein collected usage data is merged into said special purpose log database, said special purpose log database including a network representation of the collected usage data representing each item involved in one or more requests as a node on a graph and comprising connections between said nodes based on co-retrieval of items represented by said nodes, said connections being represented as weighted edges between said nodes;

means to display merged usage data in said special purpose log database; and

means to extract data from the special purpose log database based on a query input by a user and to display the extracted data to the user, said extracted data including one or more items;

means for creating and displaying a listing of item-level recommendations for the user, based on the usage information of the extracted data and corresponding to the one or more items in the extracted data,

wherein;

the special purpose log database is derived only from usage data; and

the listing of item-level recommendations is created using a local neighborhood search algorithm incorporating a spreading activation method.