CN107193796A - A kind of public sentiment event detecting method and device - Google Patents
A kind of public sentiment event detecting method and device Download PDFInfo
- Publication number
- CN107193796A CN107193796A CN201610197073.3A CN201610197073A CN107193796A CN 107193796 A CN107193796 A CN 107193796A CN 201610197073 A CN201610197073 A CN 201610197073A CN 107193796 A CN107193796 A CN 107193796A
- Authority
- CN
- China
- Prior art keywords
- vector
- sensitive
- text
- feature words
- senses
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Machine Translation (AREA)
- Document Processing Apparatus (AREA)
Abstract
The invention discloses a kind of public sentiment event detecting method and device, method includes：Obtain the feature term vector of text to be detected；The corresponding vector of all Feature Words is obtained, and obtains sensitive senses of a dictionary entry vector；Calculate the similarity of the corresponding feature term vector of all Feature Words of Feature Words vector sum of text to be detected；Corresponding first sensitive senses of a dictionary entry when obtaining similarity maximum, and obtain the quantity of Feature Words in the quantity and text to be detected of the first sensitive senses of a dictionary entry in text to be detected, according to the first preset weights and the second preset weights, the weighted sum of the quantity of the first sensitive senses of a dictionary entry and the quantity of Feature Words is calculated, it is public sentiment event that the event described in text to be detected is determined when weighted sum is more than threshold value.The present invention is by that to text vector to be detected, can reach effective semantic constraint；Simultaneously by the similarity for the corresponding feature term vector of all Feature Words of Feature Words vector sum for calculating text to be detected, the problem of can accurately detecting the public sentiment event for needing to be paid close attention to.
Description
Technical field
The present invention relates to field of computer technology, and in particular to a kind of public sentiment event detecting method and
Device.
Background technology
With the fast development of internet, network public-opinion, which is turning into ordinary people expression interests, to be told
Ask, it is just to advocate social equity, pass on the common heartfelt wishes of the common people to governments at all levels of China incessantly
One piece of thought position.Increasing people is ready the viewpoint for wanting to express and showing for being seen
As being published on network, more people are allowed to participate in by the propagation of network, so as to netizen
Mood and social stability generate significant impact.Therefore, using modern science and technology, accurate inspection
Public sentiment event tool is surveyed to be of great significance.
The detection on public sentiment event is found at present, is also rested on and is utilized the sensitive vocabulary of some public sentiments
It is such as name, outer to carry out semantic matches, and due to the name entity word with public sentiment event correlation
Scholar's name translated name and mechanism name referred to as, carriage are just embodied in the linguistic context for only appearing in associated event
Feelings.And for there is the name entity born the same name, it is necessary to analyze it with reference to current public sentiment event context
Implication, for such ambiguous Feature Words of tool, in traditional static corpus may not containing pair
Its newest explanatory senses of a dictionary entry.It is this traditional based on public sentiment Feature Words (sensitive word, name reality
Body etc.) filter method, be still a kind of important because its realization mechanism is simple, execution efficiency is high
Preprocessing means；However, in face of internet mass text, especially fragmentation, lack of standardization
Social media content, the pretreatment filtration mechanism deposits due to lacking effective semantic constraint
In certain false positive, easily cause misjudgement, fail to judge, it is impossible to which accurately identifying needs to be paid close attention to
Public sentiment event.In the network public-opinion early warning application environment of big data phase is brought to subsequent treatment
When considerable noise data input, therefore urgently need to have the data prediction of semantic understanding ability
Mechanism.
The content of the invention
Because traditional Feature Words filter method faces internet mass text, lack effective language
Justice constraint, easily causes misjudgement, fails to judge, it is impossible to the accurate public sentiment for detecting to need to be paid close attention to
The problem of event, the present invention proposes a kind of public sentiment event detecting method and device.
In a first aspect, the present invention proposes a kind of public sentiment event detecting method, including：
The feature term vector of text to be detected is obtained, the element representation of the feature term vector is to be checked
Survey whether corresponding Feature Words in text occur；
The corresponding vector of all Feature Words is obtained from semantic knowledge-base, and is obtained from sensitive dictionary
Sensitive senses of a dictionary entry vector, the corresponding vectorial element of the Feature Words includes current signature word, current
Whether the current senses of a dictionary entry comprising the sensitive senses of a dictionary entry, current signature word is corresponding with current signature word for Feature Words
Feature term vector, the justice in the corresponding vector of the sensitive senses of a dictionary entry vector representation current signature word
Item is the current sensitive senses of a dictionary entry；
Calculate the corresponding feature term vector of all Feature Words of Feature Words vector sum of text to be detected
Similarity, wherein, the corresponding feature term vector of all Feature Words includes all sensitive justice
Item vector；
Corresponding first sensitive senses of a dictionary entry when obtaining similarity maximum, and obtain institute in text to be detected
The quantity of Feature Words in the quantity and text to be detected of the first sensitive senses of a dictionary entry is stated, it is default according to first
Weights and the second preset weights, calculate the quantity and the Feature Words of the described first sensitive senses of a dictionary entry
The weighted sum of quantity, the thing described in text to be detected is determined when the weighted sum is more than threshold value
Part is public sentiment event.
Preferably, include before the feature term vector for obtaining text to be detected：
The semantic knowledge-base is built according to web page contents.
Preferably, the web page contents are stored in xml formatted files.
Preferably, the web page contents are wikipedia.
Preferably, include after the semantic knowledge-base according to web page contents structure：
Sensitive dictionary is set up according to the sensitive senses of a dictionary entry of the semantic knowledge-base and default Feature Words.
Second aspect, the present invention also proposes a kind of public sentiment event detection device, including：
Feature term vector acquisition module, the feature term vector for obtaining text to be detected is described
Whether corresponding Feature Words occur in the element representation text to be detected of feature term vector；
Correspondence vector acquisition module, it is corresponding for obtaining all Feature Words from semantic knowledge-base
Vector, and sensitive senses of a dictionary entry vector is obtained from sensitive dictionary, the corresponding vectorial member of the Feature Words
Whether element works as including current signature word, current signature word comprising the sensitive senses of a dictionary entry, current signature word
The corresponding feature term vector of the preceding senses of a dictionary entry and current signature word, the sensitive senses of a dictionary entry vector representation is current
The senses of a dictionary entry in the corresponding vector of Feature Words is the current sensitive senses of a dictionary entry；
Similarity calculation module, all features of Feature Words vector sum for calculating text to be detected
The similarity of the corresponding feature term vector of word, wherein, the corresponding Feature Words of all Feature Words
Vector includes all sensitive senses of a dictionary entry vectors；
Event checking module, corresponding first sensitive senses of a dictionary entry during for obtaining similarity maximum, and
Obtain Feature Words in the quantity and text to be detected of the first sensitive senses of a dictionary entry described in text to be detected
Quantity；According to the first preset weights and the second preset weights, the described first sensitive senses of a dictionary entry is calculated
Quantity and the Feature Words quantity weighted sum, when the weighted sum be more than threshold value when determine
Event described in text to be detected is public sentiment event.
Preferably, in addition to：
Semantic knowledge-base builds module, for building the semantic knowledge-base according to web page contents.
Preferably, the web page contents are stored in xml formatted files.
Preferably, the web page contents are wikipedia.
Preferably, in addition to：
Sensitive dictionary sets up module, for according to the quick of the semantic knowledge-base and default Feature Words
Feel the senses of a dictionary entry and set up sensitive dictionary.
As shown from the above technical solution, the present invention is by text vector to be detected, Neng Gouda
To effective semantic constraint；While all spies of Feature Words vector sum by calculating text to be detected
The similarity of the corresponding feature term vector of word is levied, the carriage for needing to be paid close attention to can be accurately detected
The problem of facts part, substantially reduce misjudgement and the probability failed to judge.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below
The accompanying drawing used required in embodiment or description of the prior art will be briefly described, show and
Easy insight, drawings in the following description are only some embodiments of the present invention, for this area
, on the premise of not paying creative work, can also be according to these for those of ordinary skill
Figure obtains other accompanying drawings.
A kind of flow for public sentiment event detecting method that Fig. 1 provides for one embodiment of the invention is shown
It is intended to；
A kind of flow chart for public sentiment event detecting method that Fig. 2 provides for one embodiment of the invention；
A kind of structure for public sentiment event detection device that Fig. 3 provides for one embodiment of the invention is shown
It is intended to.
Embodiment
Below in conjunction with the accompanying drawings, the embodiment to invention is further described.Implement below
Example is only used for clearly illustrating technical scheme, and can not limit this hair with this
Bright protection domain.
Fig. 1 shows a kind of stream for public sentiment event detecting method that one embodiment of the invention is provided
Journey schematic diagram, including：
S101, the feature term vector for obtaining text to be detected, the list of elements of the feature term vector
Show whether corresponding Feature Words occur in text to be detected；
S102, obtain the corresponding vector of all Feature Words from semantic knowledge-base, and from sensitive word
Storehouse obtains sensitive senses of a dictionary entry vector, and the corresponding vectorial element of the Feature Words includes current signature
Whether word, current signature word include the sensitive senses of a dictionary entry, the current senses of a dictionary entry of current signature word and current spy
Levy the corresponding feature term vector of word, the sensitive senses of a dictionary entry vector representation current signature word it is corresponding to
The senses of a dictionary entry in amount is the current sensitive senses of a dictionary entry；
S103, the corresponding Feature Words of all Feature Words of Feature Words vector sum for calculating text to be detected
The similarity of vector, wherein, the corresponding feature term vector of all Feature Words includes all quick
Feel senses of a dictionary entry vector；
Corresponding first sensitive senses of a dictionary entry when S104, acquisition similarity are maximum, and obtain text to be detected
The quantity of Feature Words in the quantity and text to be detected of the first sensitive senses of a dictionary entry described in this, according to the
One preset weights and the second preset weights, calculate the quantity of the described first sensitive senses of a dictionary entry and the spy
The weighted sum of the quantity of word is levied, determines to retouch in text to be detected when the weighted sum is more than threshold value
The event stated is public sentiment event.
Wherein, can be by when the corresponding Feature Words of element of the feature term vector are sensitive word
Corresponding element is set to 0.
The present embodiment is by that to text vector to be detected, can reach effective semantic constraint；
While the corresponding Feature Words of all Feature Words of Feature Words vector sum by calculating text to be detected
The similarity of vector, the problem of can accurately detecting the public sentiment event for needing to be paid close attention to, greatly
Big reduction misjudgement and the probability failed to judge.
As the alternative of the present embodiment, include before step S101：
S100, according to web page contents build the semantic knowledge-base.
By building semantic knowledge-base, ambiguity tagging is carried out to public sentiment sensitive word, for analysis detection
Public sentiment event provides semantic support, is that sensitive word in text to be detected finds correct implication and carried
For foundation.Because public sentiment Feature Words are often the direct embodiment to public sentiment, but public sentiment Feature Words
Different implications can be but represented in different linguistic context, therefore, it is special that such has ambiguous public sentiment
Levy word and often bring false positive issue to text filtering pretreatment.Therefore, by by the semanteme
Knowledge base accurately provides its description and may recognize that its expressed meaning in specific linguistic context.
Wherein, it is by dividing for the corresponding vector of the Feature Words stored in semantic knowledge-base
The pretreated text of word is trained what is obtained using deep learning instrument word2vec.It is right
Each participle (being the Feature Words in text to be detected), can use the vector of certain dimension
It is effectively represented.It is as shown in the table
Specifically, the web page contents are stored in xml formatted files.
For example, the web page contents are wikipedia.
Wikipedia (Wikipedia) is one of largest online network encyclopedia, is used
The Wiki mechanism of colony online cooperation editor, with quality is high, covering is wide, develop in real time and
Semi-structured the features such as, originated for building the high-quality language material of semantic knowledge-base.Particular for
Ambiguity word in wikipedia, the senses of a dictionary entry of artificial mark reflection public sentiment feature, is follow-up early warning point
Analysis provides support.Using the wikipedia language material of xml forms as input, retouching for word is therefrom extracted
Content is stated, analyses whether as ambiguity word and redirection word, whether need complicated and simple conversion, reservation is plucked
Introductory section is wanted, while being labeled to sensitive features word.
By the powerful semantic knowledge of wikipedia, public sentiment sensitive word can be increased automatically, expand carriage
The sign scope of facts part, so as to aid in user preferably to hold public sentiment trend, formulates related right
Plan is tackled.
Further, include after step S100：
S1001, according to the sensitive senses of a dictionary entry of the semantic knowledge-base and default Feature Words set up sensitive
Dictionary.
Wherein, can be using subordinate sentence as processing unit, to quick when handling text to be detected
Sense word is handled.During specific processing, by the spy in the feature term vector of text subordinate sentence to be detected
Levy word vector corresponding with Feature Words in semantic knowledge-base to match, by calculating different characteristic word
The senses of a dictionary entry between similarity and similarity with text to be detected, the higher explanation of similarity should
The senses of a dictionary entry more presses close to its real meaning in the text, then chooses the senses of a dictionary entry and match with sensitive word, profit
The accurate meaning of each ambiguity word in the text when obtaining object function maximum with optimal method.
Calculation formula is as follows：
maxf(wi)
f(wi)=f (wi+1)+Sim(wi,wi+1)+Sim(wi,doci)
s.t.
wi∈{v1,v2…,vm}
doci=(w1,w2,…,wn),wi=0
Wherein：wiRepresent the Feature Words in text to be detected, f (wi) represent word wiTo sentence knot
The semantic similarity value of tail word, dociThat text removes the vector representation after sensitive word, i.e., it is corresponding
The element of position is set to 0；v1, v2... it is the corresponding vector of Feature Words, if the word is non-discrimination
Adopted word, then have a vector representation, conversely, there is multiple vector representations；Sim(wi,wi+1) it is meter
Calculate the function of adjacent sensitive Word similarity, Sim (wi,doci) it is that calculating sensitive word is similar to text
The function of degree.Because word with text represents that Similarity Measure function can be used with term vector
Cosine similarity computational methods.
When for example, according to text detection public sentiment event to be detected, as shown in Fig. 2 can be first
Participle is carried out to text to be detected and goes stop words to operate, wherein, participle refers to text to be detected
Sentence in this is divided into multiple Feature Words, goes stop words to refer to leave out the deactivation in text to be detected
Word, such as " simultaneously ", " in addition ".
Then, text to be detected is obtained from semantic knowledge-base and sensitive dictionary using word2vec
Vector of sensitive senses of a dictionary entry, is easy to the adjacent word being subsequently directed in the sentence of text to be detected to enter in this
Row Similarity Measure；
Then, the sensitive senses of a dictionary entry vector vector corresponding with other Feature Words of each Feature Words is utilized
It is each quick when taking similarity maximum and the feature term vector of text to be detected carries out Similarity Measure
Feel the senses of a dictionary entry implication so that obtain with other words and text to be detected can be reasonably combined sensitivity
The senses of a dictionary entry, determines concrete meaning of this feature word in text to be detected；
Finally, weight summation is carried out to the name entity in text and the sensitive senses of a dictionary entry, more than certain
Threshold value is then judged to needing the public sentiment event of early warning.Wherein, name entity refers to text to be detected
The quantity of middle Feature Words.
The present embodiment utilizes all Feature Words in the not synonymity and text to be detected of Feature Words
Information labeling carries out the semantics recognition of supervised learning.It can avoid relying solely on Keywords matching
The drawbacks of error detection is carried out to public sentiment event, so that public sentiment event is accurately identified, it is pre- to needing
Alert public sentiment event carries out early warning.
Fig. 3 shows a kind of knot for public sentiment event detection device that one embodiment of the invention is provided
Structure schematic diagram, including：
Feature term vector acquisition module 31, the feature term vector for obtaining text to be detected,
Whether corresponding Feature Words occur in the element representation text to be detected of the feature term vector；
Correspondence vector acquisition module 32, for obtaining all Feature Words pair from semantic knowledge-base
The vector answered, and sensitive senses of a dictionary entry vector is obtained from sensitive dictionary, the corresponding vector of the Feature Words
Element whether include current signature word, current signature word comprising the sensitive senses of a dictionary entry, current signature word
The current senses of a dictionary entry and the corresponding feature term vector of current signature word, the sensitive senses of a dictionary entry vector representation
The senses of a dictionary entry in the corresponding vector of current signature word is the current sensitive senses of a dictionary entry；
Similarity calculation module 33, the Feature Words vector sum for calculating text to be detected owns
The similarity of the corresponding feature term vector of Feature Words, wherein, the corresponding spy of all Feature Words
Levying term vector includes all sensitive senses of a dictionary entry vectors；
Event checking module 34, corresponding first sensitive senses of a dictionary entry during for obtaining similarity maximum,
And obtain feature in the quantity and text to be detected of the first sensitive senses of a dictionary entry described in text to be detected
The quantity of word；According to the first preset weights and the second preset weights, the described first sensitive justice is calculated
The weighted sum of the quantity of item and the quantity of the Feature Words, it is true when the weighted sum is more than threshold value
Event described in fixed text to be detected is public sentiment event.
The present embodiment is by that to text vector to be detected, can reach effective semantic constraint；
While the corresponding Feature Words of all Feature Words of Feature Words vector sum by calculating text to be detected
The similarity of vector, the problem of can accurately detecting the public sentiment event for needing to be paid close attention to, greatly
Big reduction misjudgement and the probability failed to judge.
As the alternative of the present embodiment, in addition to：
Semantic knowledge-base builds module, for building the semantic knowledge-base according to web page contents.
Specifically, the web page contents are stored in xml formatted files.
For example, the web page contents are wikipedia.
Further, in addition to：
Sensitive dictionary sets up module, for according to the quick of the semantic knowledge-base and default Feature Words
Feel the senses of a dictionary entry and set up sensitive dictionary.
In the specification of the present invention, numerous specific details are set forth.It is to be appreciated, however, that this
The embodiment of invention can be put into practice in the case of these no details.In some examples
In, known method, structure and technology is not been shown in detail, so as not to fuzzy to this specification
Understanding.
Claims (10)
1. a kind of public sentiment event detecting method, it is characterised in that including：
The feature term vector of text to be detected is obtained, the element representation of the feature term vector is to be checked
Survey whether corresponding Feature Words in text occur；
The corresponding vector of all Feature Words is obtained from semantic knowledge-base, and is obtained from sensitive dictionary
Sensitive senses of a dictionary entry vector, the corresponding vectorial element of the Feature Words includes current signature word, current
Whether the current senses of a dictionary entry comprising the sensitive senses of a dictionary entry, current signature word is corresponding with current signature word for Feature Words
Feature term vector, the justice in the corresponding vector of the sensitive senses of a dictionary entry vector representation current signature word
Item is the current sensitive senses of a dictionary entry；
Calculate the corresponding feature term vector of all Feature Words of Feature Words vector sum of text to be detected
Similarity, wherein, the corresponding feature term vector of all Feature Words includes all sensitive justice
Item vector；
Corresponding first sensitive senses of a dictionary entry when obtaining similarity maximum, and obtain institute in text to be detected
The quantity of Feature Words in the quantity and text to be detected of the first sensitive senses of a dictionary entry is stated, it is default according to first
Weights and the second preset weights, calculate the quantity and the Feature Words of the described first sensitive senses of a dictionary entry
The weighted sum of quantity, the thing described in text to be detected is determined when the weighted sum is more than threshold value
Part is public sentiment event.
2. according to the method described in claim 1, it is characterised in that described to obtain to be detected
Include before the feature term vector of text：
The semantic knowledge-base is built according to web page contents.
3. method according to claim 2, it is characterised in that the web page contents are deposited
Storage is in xml formatted files.
4. method according to claim 3, it is characterised in that the web page contents are
Wikipedia.
5. method according to claim 4, it is characterised in that described according in webpage
Appearance includes after building the semantic knowledge-base：
Sensitive dictionary is set up according to the sensitive senses of a dictionary entry of the semantic knowledge-base and default Feature Words.
6. a kind of public sentiment event detection device, it is characterised in that including：
Feature term vector acquisition module, the feature term vector for obtaining text to be detected is described
Whether corresponding Feature Words occur in the element representation text to be detected of feature term vector；
Correspondence vector acquisition module, it is corresponding for obtaining all Feature Words from semantic knowledge-base
Vector, and sensitive senses of a dictionary entry vector is obtained from sensitive dictionary, the corresponding vectorial member of the Feature Words
Whether element works as including current signature word, current signature word comprising the sensitive senses of a dictionary entry, current signature word
The corresponding feature term vector of the preceding senses of a dictionary entry and current signature word, the sensitive senses of a dictionary entry vector representation is current
The senses of a dictionary entry in the corresponding vector of Feature Words is the current sensitive senses of a dictionary entry；
Similarity calculation module, all features of Feature Words vector sum for calculating text to be detected
The similarity of the corresponding feature term vector of word, wherein, the corresponding Feature Words of all Feature Words
Vector includes all sensitive senses of a dictionary entry vectors；
Event checking module, corresponding first sensitive senses of a dictionary entry during for obtaining similarity maximum, and
Obtain Feature Words in the quantity and text to be detected of the first sensitive senses of a dictionary entry described in text to be detected
Quantity；According to the first preset weights and the second preset weights, the described first sensitive senses of a dictionary entry is calculated
Quantity and the Feature Words quantity weighted sum, when the weighted sum be more than threshold value when determine
Event described in text to be detected is public sentiment event.
7. device according to claim 6, it is characterised in that also include：
Semantic knowledge-base builds module, for building the semantic knowledge-base according to web page contents.
8. device according to claim 7, it is characterised in that the web page contents are deposited
Storage is in xml formatted files.
9. device according to claim 8, it is characterised in that the web page contents are
Wikipedia.
10. device according to claim 9, it is characterised in that also include：
Sensitive dictionary sets up module, for according to the quick of the semantic knowledge-base and default Feature Words
Feel the senses of a dictionary entry and set up sensitive dictionary.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610144761 | 2016-03-14 | ||
| CN2016101447613 | 2016-03-14 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107193796A true CN107193796A (en) | 2017-09-22 |
| CN107193796B CN107193796B (en) | 2021-12-24 |
Family
ID=59870838
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610197073.3A Active CN107193796B (en) | 2016-03-14 | 2016-03-31 | Public opinion event detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107193796B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107992471A (en) * | 2017-11-10 | 2018-05-04 | 北京光年无限科技有限公司 | Information filtering method and device in a kind of interactive process |
| CN108647335A (en) * | 2018-05-12 | 2018-10-12 | 苏州华必讯信息科技有限公司 | Internet public opinion analysis method and apparatus |
| CN109214407A (en) * | 2018-07-06 | 2019-01-15 | 阿里巴巴集团控股有限公司 | Event detection model, calculates equipment and storage medium at method, apparatus |
| CN109344258A (en) * | 2018-11-28 | 2019-02-15 | 中国电子科技网络信息安全有限公司 | A kind of intelligent self-adaptive sensitive data identifying system and method |
| CN109472018A (en) * | 2018-09-26 | 2019-03-15 | 深圳壹账通智能科技有限公司 | Enterprise's public sentiment monitoring method, device, computer equipment and storage medium |
| CN110516166A (en) * | 2019-08-30 | 2019-11-29 | 北京明略软件系统有限公司 | Public sentiment event-handling method, device, processing equipment and storage medium |
| CN110674251A (en) * | 2019-08-21 | 2020-01-10 | 杭州电子科技大学 | Computer-assisted secret point annotation method based on semantic information |
| CN110727880A (en) * | 2019-10-18 | 2020-01-24 | 西安电子科技大学 | Sensitive corpus detection method based on word bank and word vector model |
| CN110807319A (en) * | 2019-10-31 | 2020-02-18 | 北京奇艺世纪科技有限公司 | Text content detection method and device, electronic equipment and storage medium |
| CN113505221A (en) * | 2020-03-24 | 2021-10-15 | 国家计算机网络与信息安全管理中心 | Enterprise false propaganda risk identification method, device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096680A (en) * | 2009-12-15 | 2011-06-15 | 北京大学 | Method and device for analyzing information validity |
| CN103605692A (en) * | 2013-11-04 | 2014-02-26 | 北京奇虎科技有限公司 | Device and method used for shielding advertisement contents in ask-and-answer community |
| CN103605691A (en) * | 2013-11-04 | 2014-02-26 | 北京奇虎科技有限公司 | Device and method used for processing issued contents in social network |
| CN104820629A (en) * | 2015-05-14 | 2015-08-05 | 中国电子科技集团公司第五十四研究所 | Intelligent system and method for emergently processing public sentiment emergency |
| CN104899230A (en) * | 2014-03-07 | 2015-09-09 | 上海市玻森数据科技有限公司 | Public opinion hotspot automatic monitoring system |
-
2016
- 2016-03-31 CN CN201610197073.3A patent/CN107193796B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096680A (en) * | 2009-12-15 | 2011-06-15 | 北京大学 | Method and device for analyzing information validity |
| CN103605692A (en) * | 2013-11-04 | 2014-02-26 | 北京奇虎科技有限公司 | Device and method used for shielding advertisement contents in ask-and-answer community |
| CN103605691A (en) * | 2013-11-04 | 2014-02-26 | 北京奇虎科技有限公司 | Device and method used for processing issued contents in social network |
| CN104899230A (en) * | 2014-03-07 | 2015-09-09 | 上海市玻森数据科技有限公司 | Public opinion hotspot automatic monitoring system |
| CN104820629A (en) * | 2015-05-14 | 2015-08-05 | 中国电子科技集团公司第五十四研究所 | Intelligent system and method for emergently processing public sentiment emergency |
Non-Patent Citations (2)
| Title |
|---|
| HASSAN SAYYADI ET AL.: "A Graph Analytical Approach for Topic Detection", 《ACM TRANSACTIONS ON INTERNET TECHNOLOGY》 * |
| 曹坚峰: "面向公共危机预警的网络舆情分析研究", 《中国博士学位论文全文数据库-信息科技辑》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107992471A (en) * | 2017-11-10 | 2018-05-04 | 北京光年无限科技有限公司 | Information filtering method and device in a kind of interactive process |
| CN108647335A (en) * | 2018-05-12 | 2018-10-12 | 苏州华必讯信息科技有限公司 | Internet public opinion analysis method and apparatus |
| CN109214407A (en) * | 2018-07-06 | 2019-01-15 | 阿里巴巴集团控股有限公司 | Event detection model, calculates equipment and storage medium at method, apparatus |
| CN109214407B (en) * | 2018-07-06 | 2022-04-19 | 创新先进技术有限公司 | Event detection model, method and device, computing equipment and storage medium |
| CN109472018A (en) * | 2018-09-26 | 2019-03-15 | 深圳壹账通智能科技有限公司 | Enterprise's public sentiment monitoring method, device, computer equipment and storage medium |
| CN109344258B (en) * | 2018-11-28 | 2021-11-12 | 中国电子科技网络信息安全有限公司 | Intelligent self-adaptive sensitive data identification system and method |
| CN109344258A (en) * | 2018-11-28 | 2019-02-15 | 中国电子科技网络信息安全有限公司 | A kind of intelligent self-adaptive sensitive data identifying system and method |
| CN110674251A (en) * | 2019-08-21 | 2020-01-10 | 杭州电子科技大学 | Computer-assisted secret point annotation method based on semantic information |
| CN110516166A (en) * | 2019-08-30 | 2019-11-29 | 北京明略软件系统有限公司 | Public sentiment event-handling method, device, processing equipment and storage medium |
| CN110516166B (en) * | 2019-08-30 | 2022-10-25 | 北京明略软件系统有限公司 | Public opinion event processing method, device, processing equipment and storage medium |
| CN110727880B (en) * | 2019-10-18 | 2022-06-17 | 西安电子科技大学 | Sensitive corpus detection method based on word bank and word vector model |
| CN110727880A (en) * | 2019-10-18 | 2020-01-24 | 西安电子科技大学 | Sensitive corpus detection method based on word bank and word vector model |
| CN110807319A (en) * | 2019-10-31 | 2020-02-18 | 北京奇艺世纪科技有限公司 | Text content detection method and device, electronic equipment and storage medium |
| CN110807319B (en) * | 2019-10-31 | 2023-07-25 | 北京奇艺世纪科技有限公司 | Text content detection method, detection device, electronic equipment and storage medium |
| CN113505221A (en) * | 2020-03-24 | 2021-10-15 | 国家计算机网络与信息安全管理中心 | Enterprise false propaganda risk identification method, device and storage medium |
| CN113505221B (en) * | 2020-03-24 | 2024-03-12 | 国家计算机网络与信息安全管理中心 | Enterprise false propaganda risk identification method, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107193796B (en) | 2021-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107193796A (en) | A kind of public sentiment event detecting method and device | |
| CN106557462A (en) | Name entity recognition method and system | |
| CN109086357A (en) | Sensibility classification method, device, equipment and medium based on variation autocoder | |
| Wang et al. | A Neural Model for Joint Event Detection and Summarization. | |
| CN106570180A (en) | Artificial intelligence based voice searching method and device | |
| Liu et al. | Deep contextual language understanding in spoken dialogue systems. | |
| JP6558863B2 (en) | Model creation device, estimation device, method, and program | |
| CN109472022A (en) | New word identification method and terminal device based on machine learning | |
| Elkahky et al. | A challenge set and methods for noun-verb ambiguity | |
| Shrikhande et al. | Sarcasm detection in newspaper headlines | |
| CN111209373A (en) | Sensitive text recognition method and device based on natural semantics | |
| Bogale Gereme et al. | Fighting fake news using deep learning: Pre-trained word embeddings and the embedding layer investigated | |
| Sundararajan et al. | Textual feature ensemble-based sarcasm detection in Twitter data | |
| CN117278675A (en) | Outbound method, device, equipment and medium based on intention classification | |
| EP3835994A1 (en) | System and method for identification and profiling adverse events | |
| CN108519993A (en) | The social networks focus incident detection method calculated based on multiple data stream | |
| Ajees et al. | A named entity recognition system for Malayalam using conditional random fields | |
| CN114417881B (en) | Sensitive word detection method and device, electronic equipment and storage medium | |
| Mukherjee | Extracting aspect specific sentiment expressions implying negative opinions | |
| Gonzalez et al. | Retrieval-based goal-oriented dialogue generation | |
| Türkmen et al. | A novel method for extracting feature opinion pairs for Turkish | |
| Aliane et al. | Annotating events, time and place expressions in arabic texts | |
| Xie et al. | New word detection in ancient Chinese literature | |
| Mamatha et al. | Supervised aspect category detection of co-occurrence data using conditional random fields | |
| Brito et al. | KPCA Embeddings: An Unsupervised Approach to Learn Vector Representations of Finite Domain Sequences. |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230619 Address after: 3007, Hengqin international financial center building, No. 58, Huajin street, Hengqin new area, Zhuhai, Guangdong 519031 Patentee after: New founder holdings development Co.,Ltd. Patentee after: Peking University Patentee after: BEIJING FOUNDER ELECTRONICS Co.,Ltd. Address before: 100871, fangzheng building, 298 Fu Cheng Road, Beijing, Haidian District Patentee before: PEKING UNIVERSITY FOUNDER GROUP Co.,Ltd. Patentee before: Peking University Patentee before: BEIJING FOUNDER ELECTRONICS Co.,Ltd. |
|
| TR01 | Transfer of patent right |