Enhanced answers in DeepQA system according to user preferences
A semantic search engine is enhanced to employ user preferences to customize answer output by, for a first user, extracting user preferences and sentiment levels associated with a first question; receiving candidate answer results of a semantic search of the first question; weighting the candidate answer results according to the sentiment levels for each of the user preferences; and producing the selected candidate answers to the first user. Optionally, user preferences and sentiment levels may be accumulated over different questions for the same user, or over different users for similar questions. And, supplemental information may be retrieved relative to a user preference in order to further tune the weighting per the preferences and sentiment levels.
1. A method for using user preferences to customize answer output comprising the steps of:
performing, by a computer, using a semantic search engine with machine learning, a first semantic search on a first search query for a first user;
receiving, by a computer, a first plurality of candidate answers from the first semantic search;
responsive to determining, by a computer that the first search query contains multiple-parts or multiple-steps, retrieving, by a computer, a first set of supplemental information regarding at least one of the first plurality of candidate answers;
selecting, by a computer, one or more of the first plurality of candidate answers according to the first set of supplemental information;
updating, by a computer, the machine learning of the semantic search engine according to the selecting; and
producing, by a computer, an output including the selected one or more candidate answers.
2. The method as set forth in claim 1 further comprising the steps of:
performing, by a computer, a second semantic search on a second search query for the first user, or on a first search query for a second user, or both;
receiving, by a computer, using a semantic search engine with machine learning, a second plurality of candidate answers from the second semantic search; and
selecting, by a computer, one or more of the second set of candidate answers according to the first set of supplemental information, or according to natural language preferences of the first user, or according to natural language preferences of the second user, or a combination thereof;
wherein the producing, by a computer, an output further includes the selected one or more candidate answers from the second semantic search.
3. The method as set forth in claim 1 wherein the selecting comprises a weighting process.
4. A computer program product for using user preferences to customize answer output comprising:
a computer readable, tangible storage device; and
program instructions stored by the computer readable, tangible storage device, which, when executed by a processor, cause a computing platform to:
perform, using a semantic search engine with machine learning, a first semantic search on a first search query for a first user;
receive a first plurality of candidate answers from the first semantic search;
responsive to determining that the first search query contains multiple-parts or multiple-steps, retrieving, by a computer, a first set of supplemental information regarding at least one of the first plurality of candidate answers;
select one or more of the first plurality of candidate answers according to the first set of supplemental information;
updating, by a computer, the machine learning of the semantic search engine according to the selecting; and
produce an output including the selected one or more candidate answers.
5. The computer program product as set forth in claim 4 wherein the program instructions further comprise instructions to cause the computing platform to:
perform a second semantic search on a second question for the first user, or on a first question for a second user, or both;
receive a second plurality of candidate answers from the second semantic search; and
select one or more of the second set of candidate answers according to the first set of supplemental information, or according to natural language preferences of the first user, or according to natural language preferences of the second user, or a combination thereof;
wherein the producing, by a computer, an output further includes the selected one or more candidate answers from the second semantic search.
6. The computer program product as set forth in claim 4 wherein the selecting comprises a weighting process.
7. A system for using user preferences to customize answer output comprising:
a computing platform having a processor;
a computer readable, tangible storage device; and
program instructions stored by the computer readable, tangible storage device, which, when executed by the processor, cause a computing platform to:
perform, using a semantic search engine with machine learning, a first semantic search on a first search query for a first user;
receive a first plurality of candidate answers from the first semantic search;
responsive to determining that the first search query contains multiple-parts or multiple-steps, retrieving, by a computer, a first set of supplemental information regarding at least one of the first plurality of candidate answers;
select one or more of the first plurality of candidate answers according to the first set of supplemental information;
updating, by a computer, the machine learning of the semantic search engine according to the selecting; and
produce an output including the selected one or more candidate answers.
8. The system as set forth in claim 7 wherein the program instructions further comprise instructions to cause the computing platform to:
perform a second semantic search on a second question for the first user, or on a first question for a second user, or both;
receive a second plurality of candidate answers from the second semantic search; and
select one or more of the second set of candidate answers according to the first set of supplemental information, or according to natural language preferences of the first user, or according to natural language preferences of the second user, or a combination thereof;
wherein the producing, by a computer, an output further includes the selected one or more candidate answers from the second semantic search.
9. The system as set forth in claim 7 wherein the selecting comprises a weighting process.