Method of finding answers to questions

• US 7,058,564 B2
• Filed: 04/03/2001
• Issued: 06/06/2006
• Est. Priority Date: 03/30/2001
• Status: Expired due to Term

• Current Assignee: Essencient Ltd
• Original Assignee: Hapax Ltd.
• Inventors: Ejerhed, Eva Ingegerd
• Primary Examiner(s): HARPER, V PAUL
• Application Number:
  US09/824,064
• Publication Number:
  US 20020177991A1
• Time in Patent Office: 1,890 Days
• Field of Search: 707/5, 707/4, 707/3, 707/102, 704/9, 704/8, 704/257, 704/2, 704/1, 434/362
• US Class Current: 704/1
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A method and a system for automatically finding one or more answers to a natural language question in a computer stored natural language text database is disclosed. The natural language text database has been analyzed with respect to syntactic functions of constituents, lexical meaning of word tokens and clause boundaries, and the natural language question comprises a question clause. A computer readable representation of the question clause is analyzed with respect to syntactic functions of its constituents and the lexical meaning of its word tokens. In response to the analysis a set of conditions for a clause in the natural language text database to constitute an answer to the question clause is defined. The conditions relate to the syntactic functions of constituents and the lexical meaning of word tokens in the clause. Furthermore, clauses that satisfy said conditions are identified in the natural language text database, and answers to the question clause is returned by means of the identified clauses that matches the conditions.

Claims

1. A method of automatically finding one or more answers to a natural language question in a computer stored natural language text database, wherein said natural language text database has been analyzed with respect to clause boundaries, constituents, syntactic functions of the constituents, word tokens of the constituents, and lemmas of the word tokens, and wherein said natural language question comprises a question clause having an interrogative pronoun, comprising the steps of:

analyzing a computer readable representation of said question clause with respect to constituents, syntactic functions of the constituents, word tokens of the constituents, and lemmas of the word tokens;

identifying the interrogative pronoun in said computer readable representation of said question clause;

determining a syntactic function of interest based on the analyzing step and the identified interrogative pronoun;

identifying a constituent having the determined syntactic function in the question clause;

selecting a word token being a lexical head of the identified constituent in the question clause;

defining a set of conditions comprising a condition stipulating that, for a clause in said natural language text database to constitute an answer to said question clause, the clause comprises a word token having the same lemma as the selected word token of the question clause and being comprised in a constituent having the same syntactic function as the identified constituent of said question clause;

identifying clauses in said natural language text database that satisfy said conditions; and

returning answers to said question clause by means of the identified clauses that satisfy said conditions.

2. The method of claim 1, further comprising the step of:

determining a syntactic function of a queried constituent based on the analyzing step and the identified interrogative pronoun,

wherein said set of conditions comprises a further condition stipulating that, for a clause in said natural language text database to constitute an answer to said question clause, the clause comprises a constituent having the same syntactic function as the queried constituent.

3. The method according to claim 1, wherein the identified constituent of the question clause has the syntactic function main verb, the selected word token is a lexical head of the constituent having the syntactic function main verb of the question clause, and said set of conditions in the defining step comprises:

a verb condition stipulating that, for a clause to constitute an answer to said question clause, the clause comprises a word token being a lexical head of a constituent having the syntactic function main verb, and

having the same lemma as the lexical head of the constituent having the syntactic function of main verb of said question clause.

4. The method according to claim 1, wherein the identified constituent of the question clause has the syntactic function subject, the selected word token is a lexical head of the constituent having the syntactic function subject of the question clause, and said set of conditions in the defining step comprises:

a subject condition stipulating that, for a clause to constitute an answer to said question clause, the clause comprises a word token being a lexical head of a constituent having the syntactic function subject, and having the same lemma as the lexical head of the constituent having the syntactic function of subject of said question clause.

5. The method according to claim 1, wherein the identified constituent of the question clause has the syntactic function object, the selected word token is a lexical head of the constituent having the syntactic function object of the question clause, and said set of conditions in the defining step comprises:

an object condition stipulating that, for a clause to constitute an answer to said question clause, the clause comprises a word token being a lexical head of a constituent having the syntactic function object, and having the same lemma as the lexical head of the constituent having the syntactic function of object of said question clause.