Semantic Decoding of User Queries

• US 20090157401A1
• Filed: 06/23/2008
• Published: 06/18/2009
• Est. Priority Date: 11/12/1999
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Ian M. Bennett
• Inventors: Bennett, Ian M.
• Granted Patent: US 8,229,734 B2
• Time in Patent Office: Days
• US Class Current: 704/243
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 40/216 using statistical methods
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/42 Data-driven translation
• CPC Class Codes: G09B 19/06 Foreign languages with audible presentation of material to be studied G09B5/04
• CPC Class Codes: G09B 5/04 with audible presentation of the material to be studied sound recording or reproducing G11B
• CPC Class Codes: G09B 7/00 Electrically-operated teaching apparatus or devices working with questions and answers mechanically operated G09B3/00; computing arrangements G06F
• CPC Class Codes: G09B 7/02 of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student
• CPC Class Codes: G10L 15/142 Hidden Markov Models [HMMs]
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/285 Memory allocation or algorithm optimisation to reduce hardware requirements
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: H04M 2250/74 : with voice recognition means voice activated dialling H04M1/271; voice recognition algorithms G10L15/00

Abstract

An intelligent query system for processing voiced-based queries is disclosed, which uses semantic based processing to identify the question posed by the user by understanding the meaning of the user'"'"'s utterance. Based on identifying the meaning of the utterance, the system selects a single answer that best matches the user'"'"'s query. The answer that is paired to this single question is then retrieved and presented to the user. The system, as implemented, accepts environmental variables selected by the user and is scalable to provide answers to a variety and quantity of user-initiated queries.

Claims

1. A method of performing semantic decoding of a user question presented in a speech:

utterance to a networked online natural language processing system to determine a meaning of the user question, the method comprising the steps of;

(a) receiving a set of words forming the user question at the networked online natural language processing system;

wherein the user question can be associated with one or more known queries, and which known queries have a meaning that is understood by the networked online natural language processing system;

(b) determining a frequency of occurrence of one or more terms identified in the user question and in each of said one or more known queries;

(c) determining a percentage of terms which appear in the user question that also appear in each of said one or more known queries;

(d) measuring a semantic similarity between said set of words in the user question, and a corresponding set of words in each of said one or more known queries;

(e) combining the results of steps (b) through (d) to identify which one of said one or more known queries is to be used in identifying the meaning of the user question.

2. The method of claim 1, wherein each of steps (b) through (d) includes an associated weighting factor.

3. The method of claim 2, wherein a composite semantic metric is calculated based on steps (b) through (d) and a corresponding set of weighting factors.

4. The method of claim 1, wherein during step (c) a semantic similarity W between the user question (UQ) and each of said one or more known queries is calculated using WordNet, and by computing a semantic distance between UQ (q) and a known query (d) as follows:

Sem(T_uq, T_r)=[I(uq,r)+I(r,uq)]/[Abs[T_uq]+Abs[T_r]]

where I(uq,r) and I(r,uq) are values corresponding to the inverse semantic distances computed at a given sense and level of WordNet in both directions, and T is a term frequency metric.

5. The method of claim 1, wherein said semantic decoding is performed by the natural language processing system only when a statistical based processing operation first determines that two or more known queries can correspond to the user question.

6. The method of claim 5, wherein said statistical based processing operation is based on an analysis of noun phrases and other parts of speech.

7. The method of claim 1, wherein during step (d) a word by word comparison is made between each word in the user question and each word in each of said one or more known queries to compute a matrix of values identifying a minimum lexical distance between such words.

8. The method of claim 1, wherein the semantic decoding is performed by a natural language engine operating in part on an online server connected to the Internet and in part on a client device.