Query handling in search systems

1. A method for handling queries, comprising:

extracting keywords related to a first query from one or more information sources;

classifying the keywords according to a plurality of categories defined by a query schema;

combining a plurality of the keywords into a second query according to the query schema; and

executing the second query to generate one or more results.

2. The method of claim 1 , further comprising executing the first query and only performing the steps of extracting, classifying, combining, and executing the second query if the first query generates no results.

3. The method of claim 2 , further comprising searching a query redirect database for a success query associated with the first query and only performing the steps of extracting, classifying, combining, and executing the second query if no such success query exists.

4. The method of claim 1 , further comprising weighting the keywords in accordance with a respective weight associated with each information source.

5. The method of claim 1 , wherein combining the plurality of the keywords into the second query comprises forming a plurality of candidate queries.

6. The method of claim 5 , wherein combining the plurality of keywords into the second query further comprises calculating a probability for each of the keywords.

7. The method of claim 6 , wherein combining the plurality of keywords into the second query further comprises calculating a conditional probability for respective bigrams of keywords.

8. The method of claim 7 , wherein combining the plurality of keywords into the second query further comprises multiplying each of the calculated probabilities together to determine an overall probability for each of the candidate queries.

9. The method of claim 8 , wherein combining the plurality of keywords into the second query further comprises selecting a candidate query having a highest overall probability as the second query.

10. A computer readable storage medium comprising a computer readable program for handling queries, wherein the computer readable program when executed on a computer causes the computer to perform the steps of claim 1 .

11. A method for handling queries, comprising:

extracting keywords related to a first query from one or more information sources;

weighting the keywords in accordance with a respective weight associated with each information source;

classifying the keywords according to a plurality of categories defined by a query schema;

combining a plurality of the keywords into a second query according to the query schema, comprising:

forming a plurality of candidate queries;

calculating a probability for each of the keywords;

calculating a conditional probability for respective bigrams of keywords;

multiplying each of the calculated probabilities together to determine an overall probability for each of the candidate queries; and

selecting a candidate query having a highest overall probability as the second query; and

executing the second query to generate one or more results.

12. A query handling system, comprising:

a new query module comprising a processor configured to extract keywords related to a first query from one or more information sources, to classify the keywords according to a plurality of categories defined by a query schema, to combine a plurality of the keywords into a second query according to the query schema; and

a query execution module configured to execute the second query to generate one or more results.

13. The system of claim 11 , wherein the query execution module is further configured to execute the first query and wherein the new query module is configured to only form the second query if the first query generates no results.

14. The system of claim 12 , further comprising a query redirect database, wherein the new query module is configured to only form the second query if there is no success query associated with the first query in the query redirect database.

15. The system of claim 11 , wherein the new query module is further configured to weight the keywords in accordance with a respective weight associated with each information source.

16. The system of claim 11 , wherein the new query module is further configured to form a plurality of candidate queries.

17. The system of claim 15 , wherein the new query module is further configured to calculate a probability for each of the keywords.

18. The system of claim 16 , wherein the new query module is further configured to calculate a conditional probability for respective bigrams of keywords.

19. The system of claim 17 , wherein the new query module is further configured to multiply each of the calculated probabilities together to determine an overall probability for each of the candidate queries.

20. The system of claim 11 , wherein the new query module is further configured to select a candidate query having a highest overall probability as the second query.