INFORMATION RETRIEVAL DEVICE AND INFORMATION RETRIEVAL METHOD

• US 20140258302A1
• Filed: 01/25/2013
• Published: 09/11/2014
• Est. Priority Date: 02/08/2012
• Status: Abandoned Application

• Current Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Original Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Inventors: Iizuka, Shinya
• Application Number:
  US14/001,601
• Publication Number:
  US 20140258302A1
• Time in Patent Office: Days
• US Class Current: 707/739
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/3349 Reuse of stored results of previous queries
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 16/685 using automatically derived transcript of audio data, e.g. lyrics speech recognition G10L15/00
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 2015/088 Word spotting

Abstract

An information retrieval device for retrieving information related to a word includes: an input unit for inputting a word; a pattern generation unit which, upon input of a new word after input of a given number of words, generates a word group in a case of adding the new word to a previously input word and a word group in a case of replacing a previously input word with the new word; an occurrence information derivation unit which, for each of the word groups generated, derives occurrence information corresponding to a probability of occurrence of the word group; and a determination unit which determines a word group to be used in new retrieval, based on the derived occurrence information.

Claims

1. An information retrieval device for retrieving information related to a word, the information retrieval device comprising:

an input unit for inputting a word;

a pattern generation unit which, upon input of a new word after input of a given number of words, generates a word group in a case of adding the new word to a previously input word and a word group in a case of replacing a previously input word with the new word;

an occurrence information derivation unit which, for each of the word groups generated, derives occurrence information corresponding to a probability of occurrence of the word group; and

a determination unit which determines a word group to be used in new retrieval, based on the derived occurrence information.

2. The information retrieval device according to claim 1, further comprising:

a class information database storing class information of various words,

wherein the pattern generation unit reads the class information of each of the new word and the previously input word out of the class information database, and generates the word groups, excluding a case of replacing the previously input word having the class information different form the class information that the new word has, with the new word.

3. The information retrieval device according to claim 1,

wherein the determination unit outputs the word groups in a selectable ranking format in descending order of the derived occurrence information, and determines a word group selected by a user as the word group to be used in the new retrieval.

4. The information retrieval device according to claim 1,

wherein the determination unit determines a word group with maximum occurrence information derived, as the word group to be used in the new retrieval.

5. The information retrieval device according to claim 1,

wherein the input unit includes:

a speech input unit for inputting a speech; and

a recognition unit for recognizing a word from the input speech.

6. The information retrieval device according to claim 1,

wherein the occurrence information derivation unit includes an occurrence information database storing occurrence information of various word groups, and reads the occurrence information of each of the generated word groups out of the occurrence information database, to derive the respective occurrence information of the word groups.

7. The information retrieval device according to claim 1,

wherein the given number of words are a plurality of words, and the occurrence information derivation unit derives co-occurrence information as the occurrence information.

8. The information retrieval device according to claim 1,

wherein adding the new word to a previously input word includes:

a case of adding the new word as an additional word;

a case of binding the new word to the head of one previously input word; and

a case of binding the new word to the tail of one previously input word.

9. An information retrieval method for retrieving information related to a word, which is executed by an information retrieval device, the information retrieval method comprising:

an input step for inputting a word;

a pattern generation step of, upon input of a new word after input of a given number of words, generating a word group in a case of adding the new word to a previously input word and a word group in a case of replacing a previously input word with the new word;

an occurrence information derivation step of, for each of the word groups generated, deriving occurrence information corresponding to a probability of occurrence of the word group; and

a determination step of determining a word group to be used in new retrieval, based on the derived occurrence information.