Search ranking estimation
View Patent ↗A searcher can be configured to improve relevance ranking of search results through iterative weighting of search ranking results. A Search Auto Categorizer (SAC) operates on a base query to return a probabilistic distribution of leaf categories of a taxonomy in which relevant products may reside. A Search Logic Unit (SLU) can compute a relevance of any particular leaf category to the base query. The SLU can then determine an initial relevance of a particular product to the query based on the probabilistic distribution and the relevance of leaf category to query. The SLU weights the relevance of a product to the query to establish an updated probabilistic distribution. The SLU then repeats the relevance and weighting until convergence upon a relevance list.
1. In a computerized search system in which queries are submitted by users who receive, in response, a list of documents selected from a corpus of documents wherein the list comprises documents deemed responsive to a user's query, a method of determining relevance of the documents comprising:
obtaining the query from the user;
determining a distribution of nodes of a taxonomy that have non zero probabilities of containing documents relevant to the query; determining a probability that a document matching the query resides in a particular node;
determining a relevance of documents to the query by multiplying a particular node probability from the distribution of nodes by the probability that a document matching the query resides in the particular node;
for each particular node containing a particular document, determining a value of the relevance of the particular document to the query by multiplying a determined relevance of the particular document matching the query by the initial probability that the particular node includes relevant documents;
generating updated probabilities that the distribution of nodes of the taxonomy contain relevant documents by multiplying individual ones of the relevance of documents to the query by a weighting factor, wherein the weighting factors used to generate the updated probabilities decay monotonically with the relevance value of the documents to the query;
generating an updated distribution of nodes by weighting each of the relevance of documents to the query;
determining an updated relevance of documents to the query based on the updated distribution of nodes, updated probabilities, and the relevance of documents matching the query; and
summing the updated relevance values to determine the relevance of the documents to the query.
2. The method of claim 1 , wherein determining the relevance of the documents matching the query comprises determining a probability that the particular document matching the query resides in the particular node.
3. The method of claim 1 , wherein determining the relevance of documents to the query further comprises summing all the relevance values of the particular document over all the distribution of nodes where the particular document is found.
4. The method of claim 1 , wherein determining the updated relevance of documents to the query further comprises summing all the updated relevance values of the particular document over all the distribution of nodes where the particular document is found.
5. The method of claim 1 , wherein each of the weighting factors is less than or equal to one.
6. The method of claim 5 , wherein the sum of all the weighting factors is one.
7. The method of claim 1 , further comprising repeatedly generating the updated probabilities based on the previous updated probabilities that the distribution of nodes of the taxonomy contain relevant documents and re-determining the updated relevance of documents to the query.
8. The method of claim 7 , wherein generating the updated probabilities that the distribution of nodes of the taxonomy contain relevant documents ceases when a change in the updated probabilities is less than a convergence threshold.
9. The method of claim 1 , further comprising repeatedly generating the updated probabilities based on the previous updated probabilities and re-determining the updated relevance of documents to the query.
10. The method of claim 9 , wherein generating the updated probabilities ceases when a change in the updated probabilities is less than a convergence threshold.
11. The method of claim 1 , wherein the distribution of nodes are leaf categories.
12. The method of claim 11 , wherein the leaf categories contain indexed documents related to one another.