Speculative query expansion for relevance feedback
View Patent ↗Speculative query expansion is a way to increase the effectiveness of relevance feedback. Instead of picking the best expansion query beforehand, multiple queries are executed and the best expansion query is selected afterward, but with no additional effort (or information) required from the user. From the best expansion query, search results and expanded queries more pertinent to the user can therefore be provided.
1. A computer implemented method executed by a processor for returning search results based on speculative query expansion, comprising:
receiving an initial query;
retrieving a set of documents based on the initial query, the set of documents comprising documents judged either as relevant or non-relevant to the initial query, and unjudged documents;
executing a speculative query expansion process based on the retrieved set of documents, comprising:
a. generating a list of query expansion terms based on the judged documents;
b. assigning a weight for each of the query expansion terms in the generated list;
c. selecting at least one term from the list of query expansion terms;
d. adding the selected at least one term to the initial query to create a speculatively expanded query wherein the speculatively expanded query is indicative of a query possibility;
e. generating a list of documents from the set of documents based on the speculatively expanded query;
f. assigning a pseudo metric score for the speculatively expanded query based on the order of judged documents within the generated list wherein the pseudo metric score is indicative of an application of a metric on the judged documents;
g. if no termination condition is met to terminate the speculative query process is received, returning to step c; and
h. returning the speculatively expanded query with the highest pseudo metric score;
returning search results based on the speculative query expansion process.
2. The computer implemented method of claim 1 , wherein the instruction to terminate the speculative query process is based on a preset number of iterations.
3. The computer implemented method of claim 1 , wherein the judged documents are removed from the generated list.
4. The computer implemented method of claim 1 , wherein the returning the speculatively expanded query further comprises returning the generated list.
5. A non-transitory computer readable medium comprising instructions executed by a processor for returning search results based on speculative query expansion, the instructions comprising:
receiving an initial query;
retrieving a set of documents based on the initial query, the set of documents comprising documents judged either as relevant or non-relevant to the initial query, and unjudged documents;
executing a speculative query expansion process based on the retrieved set of documents, comprising:
a. generating a list of query expansion terms based on the judged documents;
b. assigning a weight for each of the query expansion terms in the generated list;
c. selecting at least one term from the list of query expansion terms;
d. adding the selected at least one term to the initial query to create a speculatively expanded query, wherein the speculatively expanded query is indicative of a query possibility;
e. generating a list of documents from the set of documents based on the speculatively expanded query;
f. assigning a pseudo metric score for the speculatively expanded query based on the order of marked documents within the generated list, wherein the pseudo metric score is indicative of an application of a metric on the judged documents;
g. if no termination condition is met to terminate the speculative query process is received, returning to step c; and
h. returning the speculatively expanded query with the highest metric score;
returning search results based on the speculative query expansion process.
6. The non-transitory computer readable medium of claim 5 , wherein the instruction to terminate the speculative query process is based on a preset number of iterations.
7. The non-transitory computer readable medium of claim 5 , wherein the retrieving of documents further comprises conducting a search for the set of documents based on the initial query.