User-to-user communication method, program, and apparatus

• US 20080189367A1
• Filed: 12/18/2007
• Published: 08/07/2008
• Est. Priority Date: 02/01/2007
• Status: Abandoned Application

• Current Assignee: OKI Electric Industry Company Limited
• Original Assignee: OKI Electric Industry Company Limited
• Inventors: Okumura, Koji
• Application Number:
  US12/000,808
• Publication Number:
  US 20080189367A1
• Time in Patent Office: Days
• US Class Current: 709/204
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
• CPC Class Codes: H04L 51/216 Handling conversation history, e.g. grouping of messages in sessions or threads
• CPC Class Codes: H04L 67/306 User profiles

Abstract

A user-to-user communication system allows users to chat with each other by sending messages via a computer. The computer also includes a chatbot with which the users can chat; that chatbot generates response messages automatically. The computer keeps a history of messages sent by the users, infers users'"'"' interests from the content of the messages they send, matches users with similar interests, and presents the matched users as suitable chat partners. In inferring users'"'"' interests, the computer also considers message sent to the chatbot, so that a user can be matched with suitable partners even before the user has accumulated an extensive history of communication with other users.

Claims

1. A method of facilitating exchanges of messages between users using user terminals interconnected by a network to which a computing apparatus is also connected, the computing apparatus including an automatic response unit for generating automatic response messages in response to the messages sent by the users, the method comprising:

storing a message history of the messages sent by the users, including at least messages sent by the users to the automatic response unit, in a message history table in a storage unit in the computing apparatus;

identifying users with similar interests from the message history stored in the message history table, taking the messages sent by the users to the automatic response unit into account; and

presenting information enabling the users identified as having similar interests to exchange further messages with each other.

2. The method of claim 1, wherein identifying users with similar interests further comprises:

extracting each user'"'"'s interests from the message history stored in the message history table;

calculating each user'"'"'s interest levels in the extracted interests;

storing information indicating each user'"'"'s interests and the calculated interest levels in a user interest level table in the storage unit; and

calculating similarities of the users'"'"' interest levels in shared interests by using the information stored in the user interest level table.

3. The method of claim 2, wherein:

storing the message history also includes storing the messages sent by the users to other users in the message history table; and

calculating each user'"'"'s interest levels further includes weighting the messages stored in the message history table, giving greater weight to the messages sent to the automatic response unit than to the messages sent to the other users.

4. The method of claim 2, wherein calculating each user'"'"'s interest levels further comprises:

counting occurrences of first key words directly related to the user'"'"'s interests in the messages sent by the user; and

counting occurrences of second key words related to the first key words in the messages sent by the user.

5. The method of claim 4, wherein calculating each user'"'"'s interest levels further comprises:

storing coefficients indicating degrees of relatedness of the second key words to the first key words in a key word relation table in the storage unit; and

multiplying counts of occurrences of the second key words by said coefficients.

6. The method of claim 4, wherein the automatic response messages generated by the automatic response unit include a nonspecific message urging continuation of an ongoing conversation, and calculating each user'"'"'s interest levels further comprises:

identifying pairs of messages sent by users before and after receiving the nonspecific message from the automatic response unit; and

identifying relations between the first key words and the second key words from said pairs of messages.

7. The method of claim 2, wherein storing the message history includes storing times of sending of the messages, and calculating each user'"'"'s interest levels includes weighting the messages stored in the message history table according to their times of sending, greater weights being assigned to more recently sent messages.

8. The method of claim 2, wherein calculating each user'"'"'s interest levels includes weighting the user'"'"'s interests according to the number of users with which the interests are shared, greater weight being assigned to interests shared with fewer users.

9. The method of claim 2, further comprising:

searching the user interest level table to find an interest shared by a plurality of users currently exchanging messages;

having the automatic response unit generate an interjectable message related to the interest shared by the plurality of users; and

interjecting the interjectable message into the exchange of messages between the plurality of users.

10. The method of claim 1, further comprising:

training the automatic response unit, from the message history stored in the message history table, to generate imitative messages that imitate the messages sent by a first user among said users by mentioning the first user'"'"'s interests;

and sending the imitative messages to a second user among said users, thereby enabling the second user to exchange messages with the automatic response unit instead of exchanging messages with the first user.

11. A machine-readable recording medium storing a user-to-user communication program including instructions for causing a computer to carry out the user-to-user communication method of claim 1.

12. An apparatus for facilitating exchanges of messages between users using user terminals, comprising:

a storage unit for storing a message history table;

an automatic response unit for generating automatic response messages in response to the messages sent by the users;

a history keeping unit for storing the messages sent by the users, including at least the messages sent by the users to the automatic response unit, in the message history table;

an identification unit for identifying users having similar interests from the message history stored in the message history table, taking the messages sent by the users to the automatic response unit into account; and

a presentation unit for presenting information enabling the users identified as having similar interests to exchange further messages with each other.

13. The apparatus of claim 12, wherein the storage unit also stores a user interest level table indicating the users'"'"' interests and interest levels, the apparatus further comprising:

a user interest level calculation unit for extracting the users'"'"' interests from the message history stored in the message history table, calculating the users'"'"' interest levels in the extracted interests, and storing information indicating each user'"'"'s interests and the calculated interest levels in the user interest level table; and

a user interest level similarity calculation unit for calculating similarities of the users'"'"' interest levels in shared interests by using the information stored in the user interest level table.

14. The apparatus of claim 13, wherein:

the history keeping unit also stores messages sent by the users to other users in the message history table; and

the user interest level calculation unit, in calculating the users'"'"' interest levels, weights the messages stored in the message history table, giving greater weight to the messages sent to the automatic response unit than to the messages sent to the other users.

15. The apparatus of claim 13, wherein the user interest level calculation unit calculates each user'"'"'s interest levels by counting occurrences of first key words directly related to the user'"'"'s interests in the messages sent by the user, and counting occurrences of second key words related to the first key words in the messages sent by the user.

16. The apparatus of claim 13, wherein:

the storage unit stores a key word relation table giving coefficients indicating degrees of relatedness of the second key words to the first key words; and

the user interest level calculation unit multiplies counts of occurrences of the second key words by said coefficients.

17. The apparatus of claim 13, wherein:

the automatic response messages generated by the automatic response unit include a nonspecific message urging continuation of an ongoing conversation, and;

the user interest level calculation unit identifies pairs of messages sent by users before and after receiving the nonspecific message from the automatic response unit, and identifies relations between the first key words and the second key words from said pairs of messages.

18. The apparatus of claim 13, wherein the automatic response unit searches the user interest level table to find an interest shared by a plurality of users currently exchanging messages and generates an interjectable message related to the interest shared by the plurality of users, and the apparatus interjects the interjectable message into the exchange of messages between the plurality of users.

19. The apparatus of claim 12, wherein the automatic response unit automatically learns the users'"'"' interests from the message history table, learns to generate imitative messages that imitate the messages sent by a first user among said users by mentioning the first user'"'"'s interests, and sends the imitative messages to a second user among said users, thereby enabling the second user to exchange messages with the automatic response unit instead of exchanging messages with the first user.