Method for detecting negative opinions in social media, computer program product and computer

Patent metadata

Field	Value
Publication number	US-9268747-B2
Application number	US-201313792307-A
Country	US
Kind code	B2
Filing date	Mar 11, 2013
Priority date	Mar 12, 2012
Publication date	Feb 23, 2016
Grant date	Feb 23, 2016

Abstract

A method, device, and computer program product for detecting negative opinions in social media, computer program product, and computer. Negative opinions in social media can be precisely detected at an early stage. A method for processing, with a computer, a plurality of messages sent by a plurality of users over time includes the following steps: obtaining a plurality of messages, each including a specific proper noun; determining a politeness level of each of the plurality of messages, each including the specific proper noun; and calculating a proportion of messages having a politeness level lower than a certain threshold with respect to the plurality of messages, each including the specific proper noun.

First claim

We claim:

1. A computer implemented method for processing a plurality of messages sent by a plurality of users over time, wherein the computer includes a processor communicatively coupled to a memory, the method comprising the steps of:
obtaining the plurality of messages each including a specific proper noun;
determining a politeness level of each of the plurality of messages each including the specific proper noun; and
calculating a proportion of messages having a politeness level lower than a certain threshold with respect to the plurality of messages each including the specific proper noun,
wherein the message includes a citation portion cited from another message sent by another user and a comment portion including a comment given regarding the citation portion by a user as a sender of the message and the step of determining a politeness level determines the politeness level on the basis of both the comment portion and the citation portion, the comment portion have a different effect than the citation portion on the determined politeness level, and
wherein the determined politeness level is a function of a value of alpha corresponding to one plus a ratio of number of characters in the comment portion divided by number of characters in the entire message including the comment portion and the citation portion.

2. The method according to claim 1 , wherein the specific proper noun is any name of a specific organization, specific product, specific service, and/or specific person.

3. The method according to claim 1 , wherein the step of determining a politeness level determines the politeness level on the basis of whether or not the message includes a specific word representative of a politeness level.

4. The method according to claim 3 , wherein the specific word representative of a politeness level is an ancillary word including “yagaru” or “chimau” or a conjugational form of the ancillary word and the step of determining a politeness level determines that the politeness level is low when the message includes the specific word.

5. The method according to claim 3 , wherein the specific word representative of a politeness level is an independent word including “oki” and the step of determining a politeness level determines that the politeness level is high when the message includes the specific word.

6. The method according to claim 1 , wherein, when the number of characters in the comment portion is large compared to when the number of characters in the comment portion is small, the step of determining a politeness level determines that the politeness level is higher when determining the politeness level to be high and determines that the politeness level is lower when determining the politeness level to be low.

7. The method according to claim 1 , wherein the step of determining a politeness level determines the politeness level in view of a trend of politeness of messages of the individual users.

8. The method according to claim 1 , wherein the step of determining a politeness level determines a politeness level of one message on the basis of politeness levels of other messages sent by a user who has sent the one message.

9. The method according to claim 8 , wherein the step of determining a politeness level determines the politeness level of the one message on the basis of a difference between the politeness level of the one message and the politeness levels of the other messages.

10. The method according to claim 1 , further comprising the steps of:
recording temporal change in the proportion of the messages having a politeness level lower than the certain threshold; and
outputting a certain alarm when a value of the temporal change exceeds a certain threshold.

11. The method according to claim 1 , wherein, when the proportion exceeds a certain threshold, messages having a politeness level lower than a certain lower-limit threshold are output.

12. The method according to claim 11 , wherein the output messages are selected in ascending order of the politeness level.

13. The method according to claim 11 , wherein the step of determining a politeness level determines whether the output messages include a specific word representative of a politeness level and highlights the specific word in the output messages.

14. The method according to claim 11 , wherein, when the proportion exceeds the certain threshold, messages having a politeness level higher than a certain upper-limit threshold are output.

15. The method according to claim 1 , wherein the step of obtaining receives and obtains, from a social media server, a plurality of messages sampled under a condition of including the specific proper noun, among messages sent to the social media server in a certain period.

16. The method according to claim 1 wherein the determined politeness level is a function of a ratio of the number of characters in the comment portion over the number of characters in the entire message including the comment portion and the citation portion.

17. The method according to claim 1 wherein the value of alpha corresponds to one in the absence of the citation portion.

18. A non-transitory computer readable storage medium tangibly embodying a computer readable non-transitory program code having computer readable instructions which, when implemented, cause a computer to carry out the steps of a method for processing a plurality of messages sent by a plurality of users over time, the method comprising the steps of:
obtaining the plurality of messages each including a specific proper noun;
determining a politeness level of each of the plurality of messages each including the specific proper noun; and
calculating a proportion of messages having a politeness level lower than a certain threshold with respect to the plurality of messages each including the specific proper noun,
wherein a first message of the plurality of messages includes a citation portion cited from another message sent by another user and a comment portion having a comment given regarding the citation portion by a user as a sender of the message, the first message determined to have a greater politeness level magnitude than a second message of the plurality of message including only a comment portion having a comment similar to the comment portion of the first message, and
wherein the determined politeness level is a function of a value of alpha corresponding to one plus a ratio of number of characters in the comment portion divided by number of characters in the entire message including the comment portion and the citation portion.

19. The non-transitory computer readable storage medium according to claim 18 wherein the value of alpha corresponds to one in absence of the citation portion.

20. A computer for processing a plurality of messages sent by a plurality of users over time, comprising:
means for obtaining the plurality of messages each including a specific proper noun;
means for determining a politeness level of each of the plurality of messages each including the specific proper noun; and
means for calculating a proportion of messages having a politeness level lower than a certain threshold with respect to the plurality of messages each including the specific proper noun,
wherein a first message of the plurality of messages includes a citation portion cited from another message sent by another user and a comment portion having a comment given regarding the citation portion by a user as a sender of the message, the first message determined to have a greater politene

Assignees

• IBM

Inventors

• Enoki Miki
• Kanayama Hiroshi
• Nasukawa Tetsuya
• Negishi Shiho

Classifications

• G06Q30/02

  Marketing; Price estimation or determination; Fundraising · CPC title

• G06F40/284 Primary

  Lexical analysis, e.g. tokenisation or collocates · CPC title

• G06F40/216

  using statistical methods · CPC title

• G06F40/10 Primary

  Text processing (natural language analysis G06F40/20; semantic analysis G06F40/30; processing or translation of natural language G06F40/40) · CPC title

• G06F40/30

  Semantic analysis · CPC title