CN111694955A - Early dispute message detection method and system for social platform - Google Patents

Early dispute message detection method and system for social platform Download PDF

Info

Publication number
CN111694955A
CN111694955A CN202010382894.0A CN202010382894A CN111694955A CN 111694955 A CN111694955 A CN 111694955A CN 202010382894 A CN202010382894 A CN 202010382894A CN 111694955 A CN111694955 A CN 111694955A
Authority
CN
China
Prior art keywords
message
dispute
proportion
social platform
messages
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010382894.0A
Other languages
Chinese (zh)
Other versions
CN111694955B (en
Inventor
曹娟
卢名彦
谢添
刘浩远
郭俊波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Computing Technology of CAS
Original Assignee
Institute of Computing Technology of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Institute of Computing Technology of CAS filed Critical Institute of Computing Technology of CAS
Priority to CN202010382894.0A priority Critical patent/CN111694955B/en
Publication of CN111694955A publication Critical patent/CN111694955A/en
Application granted granted Critical
Publication of CN111694955B publication Critical patent/CN111694955B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • G06F16/353—Clustering; Classification into predefined classes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • G06F16/355—Class or cluster creation or modification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention provides a method and a system for detecting early dispute messages of a social platform, wherein the method comprises the following steps: collecting all messages under a preset topic in a social platform, marking the disputeness of each message according to the comment information of each message, extracting the multi-dimensional disputeness characteristics of the marked messages as training data, training a gradient lifting tree model according to the training data, and obtaining a disputeness message detection model; and acquiring the message to be published from the social platform as a message to be detected, and inputting the multi-dimensional disputed characteristics of the message to be detected into the disputed message detection model to obtain a disputed message detection result of the message to be detected. The invention can obtain the dispute of the published message to be audited in the social platform.

Description

Early dispute message detection method and system for social platform
Technical Field
The invention relates to the field of big data analysis and the technical field of information mining, in particular to an early dispute message detection method and system for a social platform.
Background
With the rapid development of the internet and the wide use of various network communication tools, people's social contact ways have been deeply changed. Social media represented by microblogs and Twitter become important channels for public to acquire information by virtue of characteristics of openness, instantaneity and the like. The rich content of the social media enables users to prefer to acquire information from the social media, and meanwhile, the participation of the users enriches the message content of the social media, so that a good cyclic process is formed.
The rapid development of social media and the virtuous circle embodied by the social media provide convenience for people to acquire information and develop social contact, but the social media still have a lot of problems. At present, disputed messages on social media are endless and may appear in various fields. Often, there are multiple pairs of cubes holding different perspectives in these controversial messages, each of which forms a group that opens up a strong debate against the controversial messages. The spread and fermentation of controversial topics such as DeEurope in the United kingdom, the great election in the United states, etc. on the Internet has led to isolation and misunderstanding between different levels of society. Some serious disputed messages even jeopardize the national conscious morphological safety, and urgently need to be supervised. Therefore, timely detection is required before the fermentation of the controversial message to prevent further deterioration of the situation. The invention provides an early dispute message detection method based on a microblog platform, and aims to predict whether a message causes dispute discussion or not when the message is just issued and a comment is not received.
Through investigation, no mature early dispute message detection method exists at present.
Disclosure of Invention
The present invention aims to address the detection of early dispute messages. Specifically, the invention provides a method for detecting an early dispute message of a social platform, which comprises the following steps:
step 1, collecting all messages under a preset topic in a social platform, marking the disputeness of each message according to the comment information of each message, extracting the multi-dimensional disputeness characteristics of the marked messages as training data, training a gradient lifting tree model by using the training data, and obtaining a disputeness message detection model;
and 2, acquiring the message to be published from the social platform as a message to be detected, and inputting the multi-dimensional disputed characteristics of the message to be detected into the disputed message detection model to obtain a disputed message detection result of the message to be detected.
The method for detecting the early dispute message of the social platform comprises the following steps of 1:
step 11, collecting and publishing hot topics within a preset time period, collecting all messages and comments under the hot topics by using a web crawler, labeling each message label according to disputeness of viewpoints contained in the comments of each message, extracting multi-dimensional disputeness characteristics of each message, and combining the labels of each message to obtain training data of a training gradient lifting tree model.
The method for detecting the early disputed messages of the social platform comprises the following steps:
the number of micro-blogs of the user publishing the message, and/or the number of fans of the user publishing the message, and/or the number of interests of the user publishing the message, and/or the number of characters of the message, and/or the number of words of the message, and/or the number of commas of the message, and/or the number of periods of the message, and/or the number of questions of the message, and/or the number of ellipses of the message, and/or the number of exclamations of the message in proportion to the number of characters of the message, and/or the number of periods of the message in proportion to the number of characters of the message, and/or the number of ellipses of the message in proportion to the number of characters of the message, and/or the average word length of the message, and/or the longest exclamations of the message, and/or the number of longest question marks of the message, and/or the number of longest commas of the message, and/or the number of longest sentence marks of the message, and/or the number of longest escape marks of the message, and/or the number of pronouns of the message, and/or the number of quantifiers of the message, and/or the number of negatives of the message, and the proportion of the number of pronouns of the message, and/or the number of the weakly pronounced words of the message, and the proportion of the pronouns of the message, and/or the number of the unsure pronounced words of the message, and the proportion of the number of inflected words of the message, and the number of pronouns of the message, and the proportion of the pronouns of the first, second, and third pronouns of the message, and the proportion of the total words of the message, and/or the number of the names of people, places and organizations of the information and the proportion of the names of all the words of the information, and/or the emotional polarity and the emotional value of the information.
The method for detecting the early dispute message of the social platform comprises the following steps of 2:
and step 21, the dispute message detection model scores the message to be detected according to the multi-dimensional dispute characteristics of the message to be detected, and selects the message to be detected with the score higher than a threshold value as the dispute message.
The method for detecting the early dispute message of the social platform is characterized in that the social platform is a microblog platform.
The invention also provides a system for detecting early dispute messages of a social platform, which comprises the following steps:
the method comprises the following steps that a module 1 collects all messages in a social platform about a preset topic, marks disputeness of each message according to comment information of each message, extracts multi-dimensional disputeness characteristics of the marked messages as training data, trains a gradient lifting tree model according to the training data, and obtains a disputeness message detection model;
and the module 2 acquires the message to be published from the social platform as the message to be detected, and inputs the multidimensional controversy characteristics of the message to be detected into the controversy message detection model to obtain the controversy message detection result of the message to be detected.
The early dispute message detection system of the social platform, wherein the module 1 comprises:
the module 11 collects and issues hot topics within a preset time period, collects all messages and comments under the hot topics by using a web crawler, marks a label for each message label according to disputeness of viewpoints contained in the comments of each message, extracts multi-dimensional disputeness characteristics of each message, and obtains training data of a training gradient lifting tree model by combining the label of each message.
The system for detecting the early disputed messages of the social platform comprises the following components:
the number of micro-blogs of the user publishing the message, and/or the number of fans of the user publishing the message, and/or the number of interests of the user publishing the message, and/or the number of characters of the message, and/or the number of words of the message, and/or the number of commas of the message, and/or the number of periods of the message, and/or the number of questions of the message, and/or the number of ellipses of the message, and/or the number of exclamations of the message in proportion to the number of characters of the message, and/or the number of periods of the message in proportion to the number of characters of the message, and/or the number of ellipses of the message in proportion to the number of characters of the message, and/or the average word length of the message, and/or the longest exclamations of the message, and/or the number of longest question marks of the message, and/or the number of longest commas of the message, and/or the number of longest sentence marks of the message, and/or the number of longest escape marks of the message, and/or the number of pronouns of the message, and/or the number of quantifiers of the message, and/or the number of negatives of the message, and the proportion of the number of pronouns of the message, and/or the number of the weakly pronounced words of the message, and the proportion of the pronouns of the message, and/or the number of the unsure pronounced words of the message, and the proportion of the number of inflected words of the message, and the number of pronouns of the message, and the proportion of the pronouns of the first, second, and third pronouns of the message, and the proportion of the total words of the message, and/or the number of the names of people, places and organizations of the information and the proportion of the names of all the words of the information, and/or the emotional polarity and the emotional value of the information.
The early dispute message detection system of the social platform, wherein the module 2 comprises:
the module 21 and the dispute message detection model score the message to be detected according to the multidimensional dispute characteristic of the message to be detected, and select the message to be detected with the score higher than the threshold value as the dispute message.
The early dispute message detection system of the social platform is characterized in that the social platform is a microblog platform.
According to the scheme, the invention has the advantage that the disputeness of published messages to be audited in the social platform can be obtained.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The invention comprises the following steps: and collecting candidate messages. Finding some hot topics in a manually defined mode, and collecting microblog contents under the topics by using a web crawler to serve as candidate messages for detection; and extracting multi-dimensional disputed characteristics. For each collected microblog text, extracting a plurality of features from two dimensions of a user and the text for dispute message detection; early dispute message detection. And classifying the extracted multi-dimensional features by using a supervised learning method. And (4) scoring each message by using a pre-trained scoring model, and selecting the message with the score value higher than a certain threshold value, namely the dispute message.
In order to make the aforementioned features and effects of the present invention more comprehensible, embodiments accompanied with figures are described in detail below.
First, candidate message collection
Therefore, all microblogs under the hot topics appearing on the social media are collected and monitored to serve as candidate microblogs for early dispute message detection.
Extracting multi-dimensional dispute characteristics
After collecting the candidate messages, extracting multidimensional disputed characteristics for each candidate message from both the user and the message text, which is described in detail as follows:
1. the microblog number of the user who published the message
2. Number of fans of user who published the message
3. Number of interest of user who published the message
4. Number of characters of the message
5. Number of words of the message
6. Comma number of the message
7. Number of exclamation marks of the message
8. Number of periods of the message
9. Number of question marks of the message
10. Number of ellipses of the message
11. The proportion of exclamation mark number in the message to character number
12. The ratio of the number of periods in the message to the number of characters in the message
13. The number of question marks in the message is in proportion to the number of characters in the message
14. The proportion of the number of ellipses in the message to the number of characters in the message
15. Average word length of the message
16. Longest exclamation mark number of the message
17. The longest number of questions of the message
18. The longest comma number of the message
19. The longest period number of the message
20. Longest number of omitted digits of the message
21. Pronoun number of the message
22. Number of quantifier of the message
23. Number of Arabic numerals of the message
24. The number of negative words of the message and the proportion of the negative words in the message
25. The number of strong words of the message and the ratio of the strong words to the number of words of the message
26. The number of weak words of the message and the proportion of the weak words to the number of the message words
27. The number of words of the message with uncertain degree and the proportion of the number of words of the message
28. The turning word number of the message and the proportion of the turning word number to the word number of the message
29. The number of the first, second and third person named pronouns of the message and the proportion of the first, second and third person named pronouns to all the pronouns of the message
30. The number of names of people, places and organizations of the message, and the proportion of the names to all the words of the message
31. The emotional polarity and the emotional value of the message are calculated by utilizing the TengSen emotional value calculation API
Third, early dispute message detection
Early dispute detection refers to determining whether a message causes dispute discussions when the message is just released and a comment is not received, wherein the message causing dispute discussions is a dispute message, and otherwise, the message is a non-dispute message. After extracting the multi-dimensional dispute characteristics from the candidate messages, the invention adopts the gradient lifting tree model of Light GBT to score, so as to realize the early detection of the dispute messages. Therefore, the gradient lifting tree model needs to be trained in advance.
Training the gradient lifting tree model needs positive and negative samples, and the patent collects the positive and negative samples through the following mode.
1. The method is characterized in that hot topics of more than one month are collected and published firstly, and the topics are fermented in sufficient time to obtain enough comments.
2. All messages and comments on these trending topics are collected using a web crawler.
3. And judging whether each message is a dispute message or not by manually checking the collected messages and comments. If the comment of a message contains both supporting and anti-opinions, and the supporting and anti-opinions are in equal proportion, the message is a dispute message, otherwise, the message is a non-dispute message. The result of the manual examination is the true tag of the message.
4. And extracting multi-dimensional dispute characteristics of each message, and combining the real labels obtained by manual examination to obtain positive and negative samples of the training gradient lifting tree.
Training a gradient lifting tree model by using the obtained positive and negative samples, then scoring all candidate messages by using the trained model, and selecting the messages with scores higher than a certain threshold value as dispute messages.
The following are system examples corresponding to the above method examples, and this embodiment can be implemented in cooperation with the above embodiments. The related technical details mentioned in the above embodiments are still valid in this embodiment, and are not described herein again in order to reduce repetition. Accordingly, the related-art details mentioned in the present embodiment can also be applied to the above-described embodiments.
The invention also provides a system for detecting early dispute messages of a social platform, which comprises the following steps:
the method comprises the following steps that a module 1 collects all messages in a social platform about a preset topic, marks disputeness of each message according to comment information of each message, extracts multi-dimensional disputeness characteristics of the marked messages as training data, trains a gradient lifting tree model according to the training data, and obtains a disputeness message detection model;
and the module 2 acquires the message to be published from the social platform as the message to be detected, and inputs the multidimensional controversy characteristics of the message to be detected into the controversy message detection model to obtain the controversy message detection result of the message to be detected.
The early dispute message detection system of the social platform, wherein the module 1 comprises:
the module 11 collects and issues hot topics within a preset time period, collects all messages and comments under the hot topics by using a web crawler, marks a label for each message label according to disputeness of viewpoints contained in the comments of each message, extracts multi-dimensional disputeness characteristics of each message, and obtains training data of a training gradient lifting tree model by combining the label of each message.
The system for detecting the early disputed messages of the social platform comprises the following components:
the number of micro-blogs of the user publishing the message, and/or the number of fans of the user publishing the message, and/or the number of interests of the user publishing the message, and/or the number of characters of the message, and/or the number of words of the message, and/or the number of commas of the message, and/or the number of periods of the message, and/or the number of questions of the message, and/or the number of ellipses of the message, and/or the number of exclamations of the message in proportion to the number of characters of the message, and/or the number of periods of the message in proportion to the number of characters of the message, and/or the number of ellipses of the message in proportion to the number of characters of the message, and/or the average word length of the message, and/or the longest exclamations of the message, and/or the number of longest question marks of the message, and/or the number of longest commas of the message, and/or the number of longest sentence marks of the message, and/or the number of longest escape marks of the message, and/or the number of pronouns of the message, and/or the number of quantifiers of the message, and/or the number of negatives of the message, and the proportion of the number of pronouns of the message, and/or the number of the weakly pronounced words of the message, and the proportion of the pronouns of the message, and/or the number of the unsure pronounced words of the message, and the proportion of the number of inflected words of the message, and the number of pronouns of the message, and the proportion of the pronouns of the first, second, and third pronouns of the message, and the proportion of the total words of the message, and/or the number of the names of people, places and organizations of the information and the proportion of the names of all the words of the information, and/or the emotional polarity and the emotional value of the information.
The early dispute message detection system of the social platform, wherein the module 2 comprises:
the module 21 and the dispute message detection model score the message to be detected according to the multidimensional dispute characteristic of the message to be detected, and select the message to be detected with the score higher than the threshold value as the dispute message.

Claims (10)

1. A method for detecting early dispute messages of a social platform is characterized by comprising the following steps:
step 1, collecting all messages under a preset topic in a social platform, marking the disputeness of each message according to the comment information of each message, extracting the multi-dimensional disputeness characteristics of the marked messages as training data, training a gradient lifting tree model by using the training data, and obtaining a disputeness message detection model;
and 2, acquiring the message to be published from the social platform as a message to be detected, and inputting the multi-dimensional disputed characteristics of the message to be detected into the disputed message detection model to obtain a disputed message detection result of the message to be detected.
2. The method for detecting early dispute messages of a social platform as claimed in claim 1, wherein the step 1 comprises:
step 11, collecting and publishing hot topics within a preset time period, collecting all messages and comments under the hot topics by using a web crawler, labeling each message label according to disputeness of viewpoints contained in the comments of each message, extracting multi-dimensional disputeness characteristics of each message, and combining the labels of each message to obtain training data of a training gradient lifting tree model.
3. The method for detecting early disputed messages on a social platform as claimed in claim 1 or 2 wherein the multidimensional disputed characteristics comprise:
the number of micro-blogs of the user publishing the message, and/or the number of fans of the user publishing the message, and/or the number of interests of the user publishing the message, and/or the number of characters of the message, and/or the number of words of the message, and/or the number of commas of the message, and/or the number of periods of the message, and/or the number of questions of the message, and/or the number of ellipses of the message, and/or the number of exclamations of the message in proportion to the number of characters of the message, and/or the number of periods of the message in proportion to the number of characters of the message, and/or the number of ellipses of the message in proportion to the number of characters of the message, and/or the average word length of the message, and/or the longest exclamations of the message, and/or the number of longest question marks of the message, and/or the number of longest commas of the message, and/or the number of longest sentence marks of the message, and/or the number of longest escape marks of the message, and/or the number of pronouns of the message, and/or the number of quantifiers of the message, and/or the number of negatives of the message, and the proportion of the number of pronouns of the message, and/or the number of the weakly pronounced words of the message, and the proportion of the pronouns of the message, and/or the number of the unsure pronounced words of the message, and the proportion of the number of inflected words of the message, and the number of pronouns of the message, and the proportion of the pronouns of the first, second, and third pronouns of the message, and the proportion of the total words of the message, and/or the number of the names of people, places and organizations of the information and the proportion of the names of all the words of the information, and/or the emotional polarity and the emotional value of the information.
4. The method for detecting early dispute messages of a social platform according to claim 1 or 2, wherein the step 2 comprises:
and step 21, the dispute message detection model scores the message to be detected according to the multi-dimensional dispute characteristics of the message to be detected, and selects the message to be detected with the score higher than a threshold value as the dispute message.
5. The method for detecting early dispute messages of a social platform as claimed in claim 1 or 2, wherein the social platform is a micro blog platform.
6. An early dispute message detection system of a social platform, comprising:
the method comprises the following steps that a module 1 collects all messages in a social platform about a preset topic, marks disputeness of each message according to comment information of each message, extracts multi-dimensional disputeness characteristics of the marked messages as training data, trains a gradient lifting tree model according to the training data, and obtains a disputeness message detection model;
and the module 2 acquires the message to be published from the social platform as the message to be detected, and inputs the multidimensional controversy characteristics of the message to be detected into the controversy message detection model to obtain the controversy message detection result of the message to be detected.
7. The system for early dispute message detection of a social platform as claimed in claim 1, wherein the module 1 comprises:
the module 11 collects and issues hot topics within a preset time period, collects all messages and comments under the hot topics by using a web crawler, marks a label for each message label according to disputeness of viewpoints contained in the comments of each message, extracts multi-dimensional disputeness characteristics of each message, and obtains training data of a training gradient lifting tree model by combining the label of each message.
8. An early dispute message detection system for a social platform as claimed in claim 6 or 7 wherein the multi-dimensional dispute feature comprises:
the number of micro-blogs of the user publishing the message, and/or the number of fans of the user publishing the message, and/or the number of interests of the user publishing the message, and/or the number of characters of the message, and/or the number of words of the message, and/or the number of commas of the message, and/or the number of periods of the message, and/or the number of questions of the message, and/or the number of ellipses of the message, and/or the number of exclamations of the message in proportion to the number of characters of the message, and/or the number of periods of the message in proportion to the number of characters of the message, and/or the number of ellipses of the message in proportion to the number of characters of the message, and/or the average word length of the message, and/or the longest exclamations of the message, and/or the number of longest question marks of the message, and/or the number of longest commas of the message, and/or the number of longest sentence marks of the message, and/or the number of longest escape marks of the message, and/or the number of pronouns of the message, and/or the number of quantifiers of the message, and/or the number of negatives of the message, and the proportion of the number of pronouns of the message, and/or the number of the weakly pronounced words of the message, and the proportion of the pronouns of the message, and/or the number of the unsure pronounced words of the message, and the proportion of the number of inflected words of the message, and the number of pronouns of the message, and the proportion of the pronouns of the first, second, and third pronouns of the message, and the proportion of the total words of the message, and/or the number of the names of people, places and organizations of the information and the proportion of the names of all the words of the information, and/or the emotional polarity and the emotional value of the information.
9. An early dispute message detection system for a social platform as claimed in claim 6 or 7 wherein the module 2 comprises:
the module 21 and the dispute message detection model score the message to be detected according to the multidimensional dispute characteristic of the message to be detected, and select the message to be detected with the score higher than the threshold value as the dispute message.
10. The system for early dispute message detection of a social platform of claim 6 or 7, wherein the social platform is a micro blog platform.
CN202010382894.0A 2020-05-08 2020-05-08 Early dispute message detection method and system for social platform Active CN111694955B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010382894.0A CN111694955B (en) 2020-05-08 2020-05-08 Early dispute message detection method and system for social platform

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010382894.0A CN111694955B (en) 2020-05-08 2020-05-08 Early dispute message detection method and system for social platform

Publications (2)

Publication Number Publication Date
CN111694955A true CN111694955A (en) 2020-09-22
CN111694955B CN111694955B (en) 2023-09-12

Family

ID=72477360

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010382894.0A Active CN111694955B (en) 2020-05-08 2020-05-08 Early dispute message detection method and system for social platform

Country Status (1)

Country Link
CN (1) CN111694955B (en)

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103279483A (en) * 2013-04-23 2013-09-04 中国科学院计算技术研究所 Topic prevalence range assessment method and system facing micro-blogs
CN106372083A (en) * 2015-07-22 2017-02-01 中国科学院计算技术研究所 Controversial news clue automatic discovery method and system
CN107180077A (en) * 2017-04-18 2017-09-19 北京交通大学 A kind of social networks rumour detection method based on deep learning
US20180189559A1 (en) * 2016-12-29 2018-07-05 Ncsoft Corporation Apparatus and method for detecting debatable document
CN108897784A (en) * 2018-06-08 2018-11-27 北京航空航天大学 One emergency event dimensional analytic system based on social media
CN109492097A (en) * 2018-10-23 2019-03-19 重庆誉存大数据科技有限公司 A kind of corporate news data classification of risks method
CN109977393A (en) * 2017-12-28 2019-07-05 中国科学院计算技术研究所 A kind of popular news prediction technique and system based on content controversial
CN110032733A (en) * 2019-03-12 2019-07-19 中国科学院计算技术研究所 A kind of rumour detection method and system for news long text
CN110134785A (en) * 2019-04-15 2019-08-16 平安普惠企业管理有限公司 Management method, device, storage medium and the equipment of forum's article
CN110209953A (en) * 2019-02-25 2019-09-06 哈尔滨工业大学 A kind of calculation method towards uncertain social computing problem
CN110263238A (en) * 2019-06-21 2019-09-20 浙江华坤道威数据科技有限公司 A kind of public sentiment listening system based on big data
CN110909125A (en) * 2019-10-30 2020-03-24 中山大学 Media rumor detection method for shoji society

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103279483A (en) * 2013-04-23 2013-09-04 中国科学院计算技术研究所 Topic prevalence range assessment method and system facing micro-blogs
CN106372083A (en) * 2015-07-22 2017-02-01 中国科学院计算技术研究所 Controversial news clue automatic discovery method and system
US20180189559A1 (en) * 2016-12-29 2018-07-05 Ncsoft Corporation Apparatus and method for detecting debatable document
CN107180077A (en) * 2017-04-18 2017-09-19 北京交通大学 A kind of social networks rumour detection method based on deep learning
CN109977393A (en) * 2017-12-28 2019-07-05 中国科学院计算技术研究所 A kind of popular news prediction technique and system based on content controversial
CN108897784A (en) * 2018-06-08 2018-11-27 北京航空航天大学 One emergency event dimensional analytic system based on social media
CN109492097A (en) * 2018-10-23 2019-03-19 重庆誉存大数据科技有限公司 A kind of corporate news data classification of risks method
CN110209953A (en) * 2019-02-25 2019-09-06 哈尔滨工业大学 A kind of calculation method towards uncertain social computing problem
CN110032733A (en) * 2019-03-12 2019-07-19 中国科学院计算技术研究所 A kind of rumour detection method and system for news long text
CN110134785A (en) * 2019-04-15 2019-08-16 平安普惠企业管理有限公司 Management method, device, storage medium and the equipment of forum's article
CN110263238A (en) * 2019-06-21 2019-09-20 浙江华坤道威数据科技有限公司 A kind of public sentiment listening system based on big data
CN110909125A (en) * 2019-10-30 2020-03-24 中山大学 Media rumor detection method for shoji society

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
张倩: "网络舆情文本多标签标注系统的研究与实现", 《信息科技辑》 *
金志威: "融合多模态特征的社会多媒体谣言检测技术研究", 《南京信息工程大学学报》 *
马鸣: "微博上谣言的特征提取及识别算法的研究", 《信息科技辑》 *

Also Published As

Publication number Publication date
CN111694955B (en) 2023-09-12

Similar Documents

Publication Publication Date Title
Dahal et al. Topic modeling and sentiment analysis of global climate change tweets
CN101599071B (en) Automatic extraction method of conversation text topic
CN108563638B (en) Microblog emotion analysis method based on topic identification and integrated learning
Abdelrazeq et al. Sentiment analysis of social media for evaluating universities
CN104899335A (en) Method for performing sentiment classification on network public sentiment of information
CN113609103A (en) AI-based interactive reading support database construction method and system
Appling et al. Discriminative models for predicting deception strategies
Ambalavan et al. Unveiling online suicide behavior: What can we learn about mental health from suicide survivors of Reddit?
CN113032557A (en) Microblog hot topic discovery method based on frequent word set and BERT semantics
CN105224920A (en) A kind ofly will proceed to the learning test diagnostic system on line under line
Priya Emoji based sentiment analysis using KNN
CN107992550A (en) A kind of network comment analysis method and system
Shen et al. Information retrieval of a disaster event from cross-platform social media
Chen et al. 4Is of social bully filtering: identity, inference, influence, and intervention
Ramzan et al. Are tweets the real estimators of election results?
CN113468868A (en) NLP-based real-time network hotspot content analysis method
CN106095758A (en) A kind of literary works based on term vector model guess method
CN104199981A (en) Method and system for classifying persons and mechanisms based on microblog texts
CN107992473A (en) Fraud information Feature Words extracting method and system based on point-by-point mutual information technology
CN111694955A (en) Early dispute message detection method and system for social platform
CN116825288A (en) Autism rehabilitation course recording method and device, electronic equipment and storage medium
CN116881395A (en) Public opinion information detection method and device
Abascal-Mena et al. User generated content: an analysis of user behavior by mining political tweets
KR101487871B1 (en) Manual Auto-generating device for Crisis Management Response of Online-based.
CN112507115B (en) Method and device for classifying emotion words in barrage text and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant