Information processing device information processing method, and recording medium

• US 7,010,477 B1
• Filed: 08/31/2000
• Issued: 03/07/2006
• Est. Priority Date: 08/31/1999
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Iwahashi, Naoto
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Shortledge, Thomas E.
• Application Number:
  US09/830,532
• Time in Patent Office: 2,014 Days
• Field of Search: 704/7, 704/8, 704/9, 704/4, 704/1, 704/10
• US Class Current: 704/7
• CPC Class Codes: G06F 40/216 using statistical methods
• CPC Class Codes: G06F 40/242 Dictionaries
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A word is input, along with a concept notation function that is a function for expressing what are described by that word, from a word set output unit to an optimal word string output unit. The optimal word string output unit computes a similarity between a word from the word set output unit and a registered word registered in a dictionary. The optimal word string output unit computes, even when a word not registered in the dictionary is input, a similarity between that word and a word registered in the dictionary.

Claims

1. An information processing apparatus for determining analogousness between input words and registered words registered in a dictionary, the information processing apparatus comprising:

a sensor section configured to sense an object related to said input words, and generating a sensed output;

a pre-processing section configured to receive said sensed output, and extract feature parameters of said object related to said input words from said sensed output;

a plurality of discriminators configured to receive said feature parameters of said object, and process said feature parameters to generate notation functions using a plurality of discrimination functions;

a selector configured to select a plurality of discriminated words representing said object;

a word train generating section configured to receive said plurality of discriminated words along with said notation functions, said word train generating section operating to generate a set of word trains by using all permutations of said plurality of discriminated words; and

word analogousness calculating means configured to calculate word analogousness between each word train of said set of word trains and a registered word train using said notation functions,

wherein said feature parameters include parameters indicating color, shape, size, position, direction, and/or velocity of said object.

2. The information processing apparatus as set forth in claim 1, wherein said notation functions include

a concept notation function.

3. The information processing apparatus as set forth in claim 1, wherein said plurality of discrimination functions is generated in a learning process.

4. The information processing apparatus as set forth in claim 3, wherein said learning process uses said feature parameters obtained by observing said object.

5. The information processing apparatus as set forth in claim 1, wherein said word analogousness calculating means also calculates word analogousness between a word in said each word train and a word in said registered word train using a concept notation function.

6. The information processing apparatus as set forth in claim 5, wherein the concept notation function includes a probability density function.

7. The information processing apparatus as set forth in claim 5, wherein the word analogousness calculating means calculates the word analogousness between the word in said each word train and the word in said registered word train on the basis of Bhattacharyya distance or Kullback divergence between the concept notation functions of the word in said word train and the word in said registered word train.

8. An information processing method for determining analogousness between input words and registered words registered in a dictionary, the information processing method comprising:

sensing an object related to said input words, and generating a sensed output;

receiving said sensed output and extracting feature parameters of said object related to said input words from said sensed output;

processing said feature parameters to generate notation functions using a plurality of discrimination functions;

selecting a plurality of discriminated words representing said object;

generating a set of word trains by using all permutations of said plurality of discriminated words; and

calculating word analogousness between each word train of said set of word trains and a registered word train using said notation functions.

9. A recording medium on which a program for determining analogousness between input words and registered words registered in a dictionary is recorded, the program comprising the executable instructions to:

sense an object related to said input words, and generating a sensed output;

receive said sensed output and extract feature parameters of said object related to said input words from said sensed output;

process said feature parameters to generate notation functions using a plurality of discrimination functions;

select a plurality of discriminated words representing said object;

generate a set of word trains by using all permutations of said plurality of discriminated words; and

calculate word analogousness between each word train of said set of word trains and a registered word train using said notation functions.