User activity estimation system and a user activity estimating method

• US 8,200,517 B2
• Filed: 08/02/2006
• Issued: 06/12/2012
• Est. Priority Date: 08/04/2005
• Status: Expired due to Fees

• Current Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Original Assignee: NTT Docomo Incorporated (Nippon Telegraph and Telephone Corporation)
• Inventors: Sakamoto, Kenji, Yamada, Naoharu, Kunito, Goro, Tanaka, Satoshi, Yamazaki, Kenichi
• Primary Examiner(s): Jeanty, Romain
• Assistant Examiner(s): FEENEY, BRETT A
• Application Number:
  US11/497,272
• Publication Number:
  US 20070033083A1
• Time in Patent Office: 2,141 Days
• Field of Search: None
• US Class Current: 705/7.11
• CPC Class Codes: G06F 3/002 Specific input/output arrangements not covered by G06F3/01 - G06F3/16 other optical apparatus G02B27/00
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

A noun and a verb is mapped as the information about an article to which an article identifier such as a radio tag, etc. is attached and a user activity, thereby estimating a user activity from the article to which the article identifier is attached and providing a user with appropriate information. An article identifier for identification of an article is associated with article information and stored (step S202), and the relationship between a noun relating to the article information and a verb relating to the article are associated with a conditional probability and held in the probability information holding unit (step S207). Upon detection of the article identifier, the probability of the verb to the noun is calculated based on the contents held in the probability information holding unit, thereby estimating a user activity (step S212). Thus, based on the detection result of the article identifier attached to the article, the activity of the user on the article can be estimated. As a result, an appropriate service can be provided for the activity of the user.

Claims

1. A computer readable medium storing a program for estimating a user activity from an article to which an article identifier for identification of the article is attached, the program executing instructions to perform the steps comprising:

extracting a noun and a verb in a sentence from document data that have been input;

calculating a conditional probability of the verb to the noun in accordance with P (verb v|noun n)∝

1/(a number of types of verbs)P(noun n|verb v);

storing an article identifier and in association with article information corresponding to the noun;

storing probability information in association with the noun and the verb that have been extracted, the conditional probability that has been calculated, and the number of the types of the verbs;

detecting an article identifier; and

estimating the user activity based on the probability information, in response to detection of the article identifier;

wherein when a plurality of the verbs are associated with the noun, a plurality of the conditional probabilities of the plurality of the verbs are compared with each other, the estimating step estimates that there is a higher possibility that the user will perform the user activity corresponding to the verb having a higher conditional probability.

2. The computer readable medium according to claim 1, wherein:

in the estimating the user activity, the noun is replaced with the article information corresponding to the article identifier, and the verb is replaced with the user activity based on an association between the noun and the verb.

3. The computer readable medium according to claim 1, further comprising:

interacting with a user to present the estimated user activity to the user;

wherein when the user selects the estimated user activity that is close to a current user activity of the user, the conditional probability is changed based upon the estimated user activity selected by the user.

4. The computer readable medium according to claim 1, wherein the estimating the user activity based on the probability information includes:

acquiring the article information stored in association with the article identifier;

acquiring the noun corresponding to the article information and the conditional probabilities of all the verbs associated with the noun; and

estimating the verb to the noun acquired.

5. A user activity estimating method for estimating a user activity from an article to which an article identifier for identification of the article is attached, the method comprising:

extracting a noun and a verb in a sentence from document data that have been input;

calculating a conditional probability of the verb to the noun in accordance with P (verb v|noun n)∝

1/(a number of types of verbs)P(noun n|verb v);

storing an article identifier and in association with article information corresponding to the noun;

storing probability information in association with the noun and the verb that have been extracted, the conditional probability that has been calculated, and the number of the types of the verbs;

detecting an article identifier; and

estimating the user activity based on the probability information, in response to detection of the article identifier;

wherein when a plurality of the verbs are associated with the noun, a plurality of the conditional probabilities of the plurality of the verbs are compared with each other, the estimating step estimates that there is a higher possibility that the user will perform the user activity corresponding to the verb having a higher conditional probability;

wherein the above steps are performed by a computer.