Abstract: A recommendation ranking system that computes trust for entities based on negative expressions of trust. Negative expressions of trust are used to reduce the trust of entities. However, the system may discount entities that in the aggregate are distrusted. The system may be used with a social network to provide accurate, personalized recommendations for members of the social network. The network may be modeled as a voting network, with each member of the social network represented as a node and expressions of trust between members represented as weights on edges between nodes. Values of trust may be computed for nodes in the network and used to generate a recommendation. Opinions expressed on a topic may be weighted by trust in the node expressing the opinion. The system may be applied in other settings that can be modeled as a voting network, including ranking of Internet search results.

Abstract: A method, system, and computer-readable storage medium for computing a representation of similarity among items in a set of items. Computing a representation of similarity items may comprise generating a first similarity model that represents characteristics of the set of items, the characteristics being indicative of similarity among the items in the set of items. Additionally, computing the representation of similarity may comprise adaptively selecting a subset of the set of items for similarity evaluation based on the first similarity model, receiving a similarity evaluation for the adaptively-selected subset of items, and generating a second similarity model based on the first to similarity model and the received similarity evaluation.

Abstract: A method, system, and computer-readable storage medium for computing a representation of similarity among items in a set of items. Computing a representation of similarity items may comprise generating a first similarity model that represents characteristics of the set of items, the characteristics being indicative of similarity among the items in the set of items. Additionally, computing the representation of similarity may comprise adaptively selecting a subset of the set of items for similarity evaluation based on the first similarity model, receiving a similarity evaluation for the adaptively-selected subset of items, and generating a second similarity model based on the first to similarity model and the received similarity evaluation.

Abstract: A method, system, and computer-readable medium for searching for items in a set of items using a similarity model. These search techniques may be used in an online purchasing system. Receiving orders for items may comprise receiving input from a user indicating a request for a first item in the set of items, picking a first plurality of items from the set of items based on the first item and a similarity model, presenting the first plurality of items to the user, and receiving a request from the user for a second item in the first plurality of items. The similarity model may represent similarity between items in the set of items and may be used to select a subset of items to quickly to lead to identification of an item of interest.

Abstract: A recommendation ranking system that computes trust for entities based on negative expressions of trust. Negative expressions of trust are used to reduce the trust of entities. However, the system may discount entities that in the aggregate are distrusted. The system may be used with a social network to provide accurate, personalized recommendations for members of the social network. The network may be modeled as a voting network, with each member of the social network represented as a node and expressions of trust between members represented as weights on edges between nodes. Values of trust may be computed for nodes in the network and used to generate a recommendation. Opinions expressed on a topic may be weighted by trust in the node expressing the opinion. The system may be applied in other settings that can be modeled as a voting network, including ranking of Internet search results.