Criteria-specific authority ranking
View Patent ↗A plurality of linkable nodes may be ranked based on their intrinsic scores. The intrinsic scores may be assigned to each node based on an analysis of the respective node with respect to one or more criteria. Links may be identified between the nodes, and a ranking score for each node may be generated based on the intrinsic scores of nodes linked therewith.
1. A method of ranking a plurality of interlinked nodes, the method comprising the steps of:
(a) analyzing each node so as to measure only a property of the node that is independent of its relation to other nodes;
(b) assigning an intrinsic score to each node based on the analysis;
(c) initiating ranking scores for all nodes: and
(d) using a processor, iteratively updating the ranking scores for the nodes, in each iteration and for each node, by determining a sum of previous ranking scores, multiplied by a term comprising the respective intrinsic score, of all nodes linked therewith, and averaging the previous ranking score for the node, weighted by the respective intrinsic score for the node and an additional weighting factor common to all nodes, with the determined sum.
2. The method of claim 1 , wherein the links are directed.
3. The method of claim 1 , wherein the links are undirected.
4. The method of claim 1 , wherein, in step (a), the nodes are analyzed with respect to a user-specified criterion.
5. The method of claim 4 , wherein the user-specified criterion comprises the user's age.
6. The method of claim 4 , wherein the user-specified criterion comprises a topic.
7. The method of claim 1 , wherein the nodes represent documents.
8. The method of claim 7 , wherein a link represents a reference in one document to another document.
9. The method of claim 8 , wherein the documents are web sites and the links represent hyperlinks.
10. The method of claim 7 , wherein the intrinsic scores reflect the readability of the documents.
11. The method of claim 1 , wherein iteratively updating the ranking scores comprises multiplying a vector of the ranking scores by a matrix that reflects (i) the intrinsic scores of the documents and (ii) links between the documents.
12. A method of ranking a plurality of interlinked nodes that form a network, the method comprising the steps of:
(a) analyzing each node so as to measure only a property of the node that is independent of its relation to other nodes, and assigning an intrinsic score to the node based on the analysis; and
(b) using a processor, iteratively computing ranking scores for the nodes by simulating a random walk through the network and when passing through a node during the simulated random walk, incrementing a counter for that node by the intrinsic score of the immediately preceding node,
wherein the counters for the nodes at termination of the random walk constitute the ranking scores of the nodes, up to normalization,
wherein the random walk comprises a series of steps and wherein simulating the random walk comprises, at each step, randomly selecting a new node from a list of nodes comprising all nodes to which the immediately preceding node links.
13. The method of claim 12 , wherein the list of nodes associated with each step further comprises the immediately preceding node.
14. A system for ranking a plurality of interlinked nodes, comprising:
a processor; and
in communication with the processor, computer memory storing instructions, executable by the processor, for
(i) analyzing each node so as to measure only a property of the node that is independent of its relation to other nodes;
(ii) assigning an intrinsic score to each node based on the analysis;
(iii) initiating ranking scores for all nodes; and
(iv) iteratively updating the ranking scores for the nodes, in each iteration and for each note, by determining a sum of previous ranking scores, multiplied by a term comprising the respective intrinsic score, of all nodes linked therewith, and averaging the previous ranking score for the node, weighted by the respective intrinsic score for the node and an additional weighting factor common to all nodes, with the determined sum.