Hybrid approach for query recommendation in conversation systems

• US 7,533,089 B2
• Filed: 06/27/2006
• Issued: 05/12/2009
• Est. Priority Date: 06/27/2006
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Shaw, James C., Pan, Shimei
• Primary Examiner(s): Vy; Hung T
• Application Number:
  US11/475,785
• Publication Number:
  US 20070299824A1
• Time in Patent Office: 1,050 Days
• Field of Search: 707 3- 6
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/56 Natural language generation
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

Techniques are disclosed for combining natural language generation with query retrieval for context appropriate query recommendation. For example, a computer-implemented method for generating a recommended query for a conversation system in response to an original user query, wherein at least a portion of the original user query is not understandable to a query interpretation process, includes the following steps. Recommendation results are computed, in response to the original user query, using a natural language generation-based recommendation process. Recommendation results are computed, in response to the original user query, using a retrieval-based recommendation process. A recommended query is generated based on consideration of at least a portion of the natural language generation-based recommendation results and at least a portion of the retrieval-based recommendation results.

Claims

1. A computer-implemented method for generating a recommended query for a conversation system in response to an original query, wherein at least a portion of the original query is not understandable to a query interpretation process, the method comprising the steps of:

computing query recommendation results, in response to the original query, using a natural language generation-based recommendation process, wherein the natural language generation-based recommendation result computing step comprises the steps of;

selecting content for use in generating the recommended query, wherein the content selection step comprises the steps of;

extracting one or more features from at least one of the original query and interpretation results generated by the query interpretation process, wherein the one or more extracted features characterize a current interpretation problem associated with the original query;

based on the one or more extracted features that characterize the current interpretation problem exhibited by the interpretation results generated by the query interpretation process, determining and applying one or more rules, indexed by the current interpretation problem, for use in revising at least a portion of the original query to generate semantic content for use in generating the recommended query; and

based on the semantic content, generating a grammatically-appropriate sentence which forms the recommended query, wherein the grammatically-appropriate sentence generating step further comprises retrieving one or more of words, phrases and sentence segments from at least one of the original query and a query corpus to convey the semantic content for the recommended query;

computing query recommendation results, in response to the original query, using a retrieval-based recommendation process, wherein the retrieval-based recommendation result computing step comprises the step of computing one or more similarity scores based on the original query and one or more example queries from a query corpus;

generating a recommended query by proportionately merging at least a portion of the natural language generation-based query recommendation results and at least a portion of the retrieval-based query recommendation results; and

sending the recommended query back to a sender of the original query such that a subsequent query can be sent based on the recommended query.