×

Assessing content based on assessed trust in users

  • US 7,822,631 B1
  • Filed: 11/30/2005
  • Issued: 10/26/2010
  • Est. Priority Date: 08/22/2003
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented method for an electronic marketplace to select customer-supplied item reviews for presentation to other customers based on automatically assessed quality of the item reviews, the assessed quality of an item review being based at least in part on automatically calculated trust scores of customers associated with the item review, the method comprising:

  • tracking activities of customers of the electronic marketplace in performing multiple types of interactions with the electronic marketplace, the tracking being performed by one or more programmed computing systems of the electronic marketplace;

    for each of the customers, automatically calculating a trust score for the customer based at least in part on the tracked activities of the customer, the calculated trust score for a customer reflecting a level of trust granted by the electronic marketplace regarding future activities of the customer, the automatic calculating of the trust score for the customer being performed by the one or more programmed computing systems;

    receiving multiple textual item reviews that are each supplied by one of the customers and that are each related to one of multiple items provided by the electronic marketplace to customers; and

    for each of the item reviews, using the item review by,automatically assessing an initial quality of the item review based at least in part on the calculated trust score for the customer from whom the item review was received, the assessing such that a calculated trust score reflecting a high level of trust will produce a higher assessed initial quality than a calculated trust score reflecting a low level of trust, the calculated trust score of the customer being based at least in part on types of interactions with the electronic marketplace that are unrelated to providing content or to evaluating content, the automatic assessing of the initial quality being performed by the one or more programmed computing systems;

    automatically determining whether to present the item review to other customers based at least in part on whether the automatically assessed initial quality of the item review is sufficiently high, the automatic determining being performed by the one or more programmed computing systems; and

    if the item review is presented to other customers,receiving evaluations of the item review from each of at least some of the other customers, the received evaluations including numerical ratings of the item review for multiple rating dimensions that are related to distinct aspects of the item review, the numerical rating for a rating dimension indicating an assessment by one of the other customers of a degree to which an aspect of the item review for that rating dimension is satisfied;

    automatically assessing an updated quality of the item review based at least in part on the automatically assessed initial quality and based at least in part on the numerical ratings of the received evaluations, the automatic assessing performed so as to weight the numerical ratings received from customers based on calculated trust scores for those customers, the automatic assessing of the updated quality being performed by the one or more programmed computing systems; and

    if it is determined to continue presenting the item review to other customers based at least in part on the automatically assessed updated quality of the item review being sufficiently high, presenting the item review to other customers, so that quality of item reviews that are supplied by and evaluated by customers may be automatically assessed in a manner to reflect automatically calculated levels of trust in the customers based on multiple types of interactions of the customers with the electronic marketplace, and wherein each of the one or more programmed computing systems includes at least one computer processor.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×