Double iterative flavored rank

A system for determining relevance of a web node comprising:

• one or more processors;
• memory operably connected to the one or more processors;
• and
• the memory storing one or more modules programmed to:
• compute an unbiased probability vector for a collection of nodes, the collection of nodes comprising one or more seed nodes included within a seed set of nodes, the web node, and one or more evaluation nodes, the unbiased probability vector setting forth a plurality of reachability relationships, each corresponding to a different pair of nodes of the collection of nodes and setting forth a probability that a random surfer beginning on a first node of the different pair of nodes will end up on a second node of the different pair of nodes;
• assign one or more source score values to the one or more seed nodes;
• set a source expansion factor to a value that allows some probability set forth in the unbiased probability vector with respect to the one or more seed nodes to spread out into other nodes within the collection of nodes, while retaining a portion of the probability within the one or more seed nodes;
• derive a destination score value for the web node based on a mapping of a reachability relationship of the plurality of reachability relationships between the one or more seed nodes and the web node;
• and
• derive, as limited by the source expansion factor, a source score value for the web node based on a mapping of a reachability relationship of the plurality of reachability relationships between the web node and the one or more evaluation nodes and based on a derived destination score associated, respectively, with each of the one or more evaluation nodes.

A computer program product for determining relevance of a web node, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

• computing an unbiased probability vector for a collection of nodes, the collection of nodes comprising one or more seed nodes included within a seed set of nodes, the web node, and one or more evaluation nodes, the unbiased probability vector setting forth a plurality of reachability relationships, each corresponding to a different pair of nodes of the collection of nodes and setting forth a probability that a random surfer beginning on a first node of the different pair of nodes will end up on a second node of the different pair of nodes;
• assigning one or more source score values to the one or more seed nodes;
• setting a source expansion factor to a value that allows some probability set forth in the unbiased probability vector with respect to the one or more seed nodes to spread out into other nodes within the collection of nodes, while retaining a portion of the probability within the one or more seed nodes;
• deriving a destination score value for the web node based on a mapping of a reachability relationship of the plurality of reachability relationships between the one or more seed nodes and the web node;
• deriving, as limited by the source expansion factor, a source score value for the web node based on a mapping of a reachability relationship of the plurality of reachability relationships between the web node and the one or more evaluation nodes and based on a derived destination score associated, respectively, with each of the one or more evaluation nodes.

A method of determining relevance of a web node comprising:

• computing, by a computer system, an unbiased probability vector for a collection of nodes, the collection of nodes comprising one or more seed nodes included within a seed set of nodes, the web node, and one or more evaluation nodes, the unbiased probability vector setting forth a plurality of reachability relationships, each corresponding to a different pair of nodes of the collection of nodes and setting forth a probability that a random surfer beginning on a first node of the different pair of nodes will end up on a second node of the different pair of nodes;
• assigning, by the computer system, one or more destination score values to the one or more seed nodes;
• setting, by the computer system, a destination expansion factor to a value that allows some probability set forth in the unbiased probability vector with respect to the one or more seed nodes to spread out into other nodes within the collection of nodes, while retaining a portion of the probability within the one or more seed nodes;
• deriving, by the computer system, a source score value for the web node based on a mapping of a reachability relationship of the plurality of reachability relationships between the one or more seed nodes and the web node;
• and
• deriving, by the computer system as limited by the destination expansion factor, a destination score value for the web node based on a mapping of a reachability relationship of the plurality of reachability relationships between the web node and the one or more evaluation nodes and based on a derived source score associated, respectively, with each of the one or more evaluation nodes.