Abstract: A method and system for ranking products using similarity propagation. The method incudes calculating a total score of a product based on its own score and a propagated score. The propagated score is determined by looking at a similarity distance value and then applying a similarity distance factor to a product's score to determine a propagated score for a new product.

Abstract: A method for updating online reservations may include receiving a request for a listing of available hotel rooms, retrieving available hotel room information from multiple servers, displaying the list of available hotel rooms, and receiving user selection of one of the available hotel rooms. The method may further include generating a reservation based on the user selection, monitoring hotel room prices, canceling an original reservation in response to identifying a lower price, generating a new reservation, and crediting a user account based on a price difference between the original reservation and the new reservation.

Abstract: A method and system for providing automated on-site merchant coupons including determining a product of interest, determining if there is a coupon available for the product of interest and then generating and transmitting the coupon if one is available. The determination of whether a coupon is available may be based on a set of rules as designated by a merchant or may be based on a price-matching criteria.

Abstract: A method and system for clustering products by determining a similarity of the products based on their characteristics. Based on the similarity of the products, an n-space mapping can be generated. The n-space mapping can then be used to reflect how products are clustered together based on similarity propagation.

Abstract: A method and system for ranking products using similarity propagation. The method incudes calculating a total score of a product based on its own score and a propagated score. The propagated score is determined by looking at a similarity distance value and then applying a similarity distance factor to a product's score to determine a propagated score for a new product.