Recommender system and method

• US 7,440,943 B2
• Filed: 12/22/2000
• Issued: 10/21/2008
• Est. Priority Date: 12/22/2000
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Meunier, Jean-Luc, Grasso, Antonietta, Glance, Natalie S.
• Primary Examiner(s): Mofiz; Apu
• Assistant Examiner(s): STACE, BRENT S
• Application Number:
  US09/746,917
• Publication Number:
  US 20020116291A1
• Time in Patent Office: 2,860 Days
• Field of Search: 707/1, 707/3, 707/5, 707/10, 707/6, 705/10, 705/26
• US Class Current: 707/736
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

A system for providing item recommendations includes a memory, a device, responsive to a user request, for recording an item on a hardcopy medium, and a processor, for storing ratings of items and for generating recommendations for new items based on recommendation criteria. In response to the user request, the processor stores an implicit rating for the requested item in the memory, determines whether, based on the implicit rating and the recommendation criteria, to generate an item recommendation, and if the criteria for generating a recommendation is met, generates a recommendation of a new item. The recommender system may further store a representation of the recorded item in the memory. Recommendations may be based on item to item similarities, item to user similarities or user to user similarities.

Claims

1. A system for providing item recommendations, comprising:

a memory;

a device, responsive to a user request, for recording an item on a hardcopy medium, wherein recording the item on a hardcopy medium comprises an implicit rating for the item by the user;

a processor, for storing ratings on items and for generating recommendations for new items based on recommendation criteria;

wherein, responsive to the recording of the item on a hardcopy medium, the processor stores the implicit rating for the recorded item in the memory, determines whether, based on the implicit rating for the recorded item and the recommendation criteria, to generate an item recommendation, and if the criteria for generating a recommendation is met, generates a recommendation of a new item;

wherein the memory stores user profiles for users of the system, wherein each user profile includes a set of user preferences pertaining to items and wherein the processor, responsive to the recorded item, updates the user'"'"'s profile with the implicit rating of the recorded item;

wherein the processor further stores a representation of the recorded item in memory and determines an item similarity for the recorded item with other items stored in the memory by comparing the stored representation of the recorded item with stored representations of other recorded items stored in memory;

wherein the processor characterizes content of the recorded item sing linguistic tools and wherein the processor determines an item to item similarity between two recorded items by calculating a sum or weights of keywords in common divided by a sum of weights of all keywords associated with the two recorded items.

2. A system for providing item recommendations, comprising:

a memory;

a device, responsive to a user request, for recording an item on a hardcopy medium, wherein recording the item on a hardcopy medium comprises an implicit rating for the item by the user;

a processor, for storing ratings of items and for generating recommendations for new items based on recommendation criteria;

wherein, responsive to the recorded item, the processor stores an implicit rating for the recorded item in the memory, determines whether, based on the implicit rating for the recorded item and the recommendation criteria, to generate an item recommendation, and if the criteria for generating a recommendation is met, generates a recommendation of a new item;

wherein the memory stores user profiles for users of the system, wherein each user profile includes a set of user preferences pertaining to items and wherein the processor, responsive to the recorded item, updates the user'"'"'s profile with the implicit rating of the recorded item;

wherein the processor characterizes content of the recorded item using linguistic tools and wherein the processor generates a historical linguistic user profile for each user comprising a list of terms extracted from user recorded items and frequency of occurrence of such extracted terms and wherein the processor generates a current linguistic user profile for each user comprising a list of terms extracted from user recorded items with terms being weighted by a damping coefficient, e<sup>−

α

t</sup>, where t=today−

timestamp of association of the recorded item with the user and α

is a damping coefficient.

3. In a system for providing item recommendations, comprising:

a memory;

a device for recording an item on a hardcopy medium; and

a processor, for storing ratings of items and for generating recommendations for new items based on recommendation criteria, a method for operating the system for generating recommendations, comprising;

providing a user request for recording an item on a hardcopy medium;

storing an implicit rating of the requested item, wherein recording the item on a hardcopy medium comprises an implicit rating for the item by the user;

determining whether, based on the implicit rating for the recorded item and recommendation criteria, to generate an item recommendation;

if the criteria for generating a recommendation is met, generating a recommendation of a new item;

storing user profiles for users providing user requests, wherein each user profile includes a set of user preferences pertaining to items and responsive to the recorded item, updating the requesting user'"'"'s profile with the implicit rating for the recorded item;

calculating an item similarities rating between two recorded items by calculating a sum of weights of keywords in common divided by a sum of weights of all keywords associated with the two recorded items.

4. In a system for providing item recommendations, comprising:

a memory;

a device for recording an item on a hardcopy medium; and

a processor, for storing ratings of items and for generating recommendations for new items based on recommendation criteria, a method for operating the system for generating recommendations, comprising;

providing a user request for recording an item on a hardcopy medium;

storing an implicit rating of the requested item, wherein recording the item on a hardcopy medium comprises an implicit rating for the item by the user;

determining whether, based on the implicit rating for the recorded item and recommendation criteria, to generate an item recommendation;

if the criteria for generating a recommendation is met, generating a recommendation of a new item

storing user profiles for users providing user requests, wherein each user profile includes a set of user preferences pertaining to items and responsive to the recorded item, updating the requesting user'"'"'s profile with the implicit rating for the recorded item;

characterizing content of the recorded item using linguistic tools;

generating a historical linguistic user profile for each user comprising a list of terms extracted from user recorded items and frequency of occurrence of such extracted terms; and

generating a current linguistic user profile for each user comprising a list of terms extracted from user recorded items with terms being weighted by a damping coefficient, e<sup>−

α

t</sup>, where t=today−

timestamp of association of the recorded item with the user and α

is the damping coefficient.