Method for matching queries with answer items in a knowledge base

Abstract

The present invention includes an expert system in which a search index furnishes answers to incoming queries provided in natural language. A search index for a specific field contains components that facilitate selecting a best fitting stored answer to the incoming query. Furthermore, context of the incoming query (e.g. location of the user, a current web page or service being used/viewed by the user, the time, etc.) may be considered when selecting a best fitting answer.

1 . A system for matching stored queries (matchSQs) to a user natural language query (NLQ), said system comprising:
 a computing platform including communication circuitry, processing circuitry and computer executable code adapted to cause the computing platform to:   (a) receive digital data representing the user NLQ;   (b) search at least one knowledgebase for candidate matchSQs, which candidate matchSQs include words corresponding to words in the user NLQ;   (c) score each match between the NLQ and each individual candidate matchSQs by:
 a. calculating a first Similar Word Score (SWS) between the NLQ and the individual candidate matchSQ by using a defined formula to combine or multiply significance values of words in the individual candidate matchSQ corresponding to words in the NLQ; 
 b. calculating a second SWS between the NLQ and the NLQ by using the defined formula to combine or multiply significance values of all words in the NLQ; 
 c. calculating a third SWS between the individual candidate matchSQ and the individual candidate matchSQ by using the defined formula to combine or multiply significance values of all words in the individual candidate matchSQ; 
 d. calculating a match score between the NLQ and the individual candidate matchSQ by aggregating: (1) a difference between the first SWS and the second SWS and (2) a difference between the first SWS and the third SWS. 
   
     
     
         2 . The system according to  claim 1 , further comprising disqualifying matchSQs based on a comparison of context of the candidate matchSQs to a context of the NLQ. 
     
     
         3 . The system according to  claim 1 , wherein calculating a match score between the NLQ and each individual candidate matchSQ further includes factoring a similarity or dissimilarity in the context of the NLQ and contexts of the candidate matchSQs. 
     
     
         4 . The system according to  claim 1 , wherein the significance values of words represent a frequency of use within one or more texts or knowledgebases. 
     
     
         5 . The system according to  claim 1 , further comprising calculating a query specific significance value for words comprising the NLQ, the query specific significance value of each given word representing a ratio between a significance value of the word and a sum of significance values of words comprising the NLQ. 
     
     
         6 . The system according to  claim 1 , further comprising calculating a query specific significance value for words comprising each individual candidate matchSQ, the query specific significance value of each given word representing a ratio between a significance value of the word and a sum of significance values of words comprising the individual candidate matchSQ. 
     
     
         7 . A system for providing an automated response to a user natural language query (NLQ) made in regard to a subject, said system comprising:
 a computing platform including processing circuitry associated with a tangible digital medium containing computer executable code adapted to cause a processor to:   (a) receive digital data representing the user NLQ;   (b) search at least one knowledgebase for candidate matchSQs, which candidate matchSQs include words corresponding to words in the user NLQ;   (c) score each match between the NLQ and each individual candidate matchSQs by:
 i. calculating a first Similar Word Score (SWS) between the NLQ and the individual candidate matchSQ by using a defined formula to combine or multiply significance values of words in the individual candidate matchSQ corresponding to words in the NLQ; 
 ii. calculating a second SWS between the NLQ and the NLQ by using the defined formula to combine or multiply significance values of all words in the NLQ; 
 iii. calculating a third SWS between the individual candidate matchSQ and the individual candidate matchSQ by using the defined formula to combine or multiply significance values of all words in the individual candidate matchSQ; 
 iv. calculating a match score between the NLQ and the individual candidate matchSQ by aggregating: (1) a difference between the first SWS and the second SWS and (2) a difference between the first SWS and the third SWS; 
   and   communication circuitry adapted to send the computer executable code to the processor.   
     
     
         8 . The system according to  claim 7 , further comprising disqualifying matchSQs based on a comparison of context of the candidate matchSQs to a context of the NLQ. 
     
     
         9 . The system according to  claim 7 , wherein calculating a match score between the NLQ and each individual candidate matchSQ further includes factoring a similarity or dissimilarity in the context of the NLQ and contexts of the candidate matchSQs. 
     
     
         10 . The system according to  claim 7 , wherein the significance values of words represent a frequency of use within one or more texts or knowledgebases. 
     
     
         11 . The system according to  claim 7 , further comprising calculating a query specific significance value for words comprising the NLQ, the query specific significance value of each given word representing a ratio between a significance value of the word and a sum of significance values of words comprising the NLQ. 
     
     
         12 . The system according to  claim 7 , further comprising calculating a query specific significance value for words comprising each individual candidate matchSQ, the query specific significance value of each given word representing a ratio between a significance value of the word and a sum of significance values of words comprising the individual candidate matchSQ.