CN108647257A - A kind of microblog emotional determines method - Google Patents
A kind of microblog emotional determines method Download PDFInfo
- Publication number
- CN108647257A CN108647257A CN201810372663.4A CN201810372663A CN108647257A CN 108647257 A CN108647257 A CN 108647257A CN 201810372663 A CN201810372663 A CN 201810372663A CN 108647257 A CN108647257 A CN 108647257A
- Authority
- CN
- China
- Prior art keywords
- sentence
- pending
- microblogging
- dictionary
- microblogging text
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Machine Translation (AREA)
Abstract
A kind of microblog emotional of present invention offer determines method, can improve the accuracy of sentiment analysis result.The method includes：Build microblog emotional dictionary；Obtain pending microblogging text；According to pre-set clause classifying rules collection, determine that the clause type and the clause type of each subordinate sentence in the pending microblogging text obtained influence weights to Sentiment orientation；Weights are influenced on Sentiment orientation according to the clause type and the clause type of each subordinate sentence in the microblog emotional dictionary of structure, and determining pending microblogging text, determine the Sentiment orientation value of the pending microblogging text.The present invention is suitable for obtaining the emotion of user's expression.
Description
Technical field
The present invention relates to processing data information technical field, particularly relates to a kind of microblog emotional and determine method.
Background technology
Microblogging is that one kind of information manufacture, exchange, propagation and an acquisition based on relationship is integrated, Opening is social
Service platform, huge microblog users group and mass data information, under cover huge commercial value and social value, how
Effectively obtaining the emotion of user's expression becomes the hot spot of nowadays each area research.
Current microblog emotional analysis method using it is relatively broad be the sentiment analysis method based on machine learning.Based on machine
The sentiment analysis method of device study is learned by converting text to digital model, with the training data for having marked classification to train
Then Sentiment orientation classification that a preferable disaggregated model recycles the disaggregated model prediction unknown text for learning is practised out, often
Machine learning algorithm has naive Bayesian (NB), support vector machines (SVM), arest neighbors (KNN), maximum entropy (ME) and decision
Tree method etc..
The sentiment analysis method based on machine learning used in the prior art ignores microblogging text particularity and inclines to emotion
Tropism influences, and causes sentiment analysis result not accurate enough.
Invention content
The technical problem to be solved in the present invention is to provide a kind of microblog emotionals to determine method, to solve present in the prior art
Sentiment analysis method based on machine learning ignore microblogging text particularity emotion tendency influenced, lead to sentiment analysis knot
The not accurate enough problem of fruit.
In order to solve the above technical problems, a kind of microblog emotional of offer of the embodiment of the present invention determines method, including：
Build microblog emotional dictionary；
Obtain pending microblogging text；
According to pre-set clause classifying rules collection, the sentence of each subordinate sentence in the pending microblogging text obtained is determined
Formula type and the clause type influence weights to Sentiment orientation；
According to the clause type of each subordinate sentence and institute in the microblog emotional dictionary of structure, and determining pending microblogging text
State clause type influences weights to Sentiment orientation, determines the Sentiment orientation value of the pending microblogging text.
Further, the structure microblog emotional dictionary includes：
Microblogging vocabulary, microblogging emoticon are extracted, according to the microblogging vocabulary of extraction, microblogging emoticon to basic emotion word
Allusion quotation is extended, and obtains microblog emotional dictionary；
Wherein, the microblog emotional dictionary includes：Microblogging basis sentiment dictionary, network sentiment dictionary, microblogging emoticon
Dictionary, modification dictionary.
Further, the modification dictionary includes：Negative word dictionary and degree adverb dictionary.
Further, after obtaining pending microblogging text, the method further includes：
The pending microblogging text of acquisition is pre-processed and carries out Emotion tagging；
Wherein, described pre-process includes：Noise reduction, participle and stop words filter operation, and user is added during participle
Custom Dictionaries.
Further, according to pre-set clause classifying rules collection, the pending microblogging text obtained is determined
In each subordinate sentence clause type and the clause type on Sentiment orientation influence weights before, the method further includes：
Subordinate sentence algorithm is constructed, subordinate sentence is carried out to the pending microblogging text of acquisition, wherein the subordinate sentence algorithm is needle
Microblogging text expression lack of standard is configured.Further, the clause type includes：Simple sentence and complex sentence；
Wherein, the simple sentence includes：Exclamative sentence, interrogative sentence, assertive sentence, the complex sentence include：Confirmative question, turnover
Sentence, selection sentence, progressive sentence, concession sentence, summarizes sentence at hypothetical sentence.
Further, Sentiment orientation value E (p)=E (s of the pending microblogging text1)+E(s2)+…+E(sn)；
Wherein, E (p) indicates the Sentiment orientation value of pending microblogging text p, E (si) indicate in pending microblogging text p the
I subordinate sentence siSentiment orientation value；E(si)=Wseni·Epri(si), WseniFor i-th of subordinate sentence siClause type to emotion
The influence weights of tendency, WseniIt is determined by clause classifying rules, Epri (si) it is i-th of subordinate sentence siBasic emotion value.
Further, the basic emotion value Epri (si) be expressed as：
Wherein, Wdeg is degree adverb weights, and Wno is negative word weights, E (wt) indicate i-th of subordinate sentence siIn t-th of feelings
Feel the emotional value of word, E (emjj) indicate i-th of subordinate sentence siIn j-th of emoticon emotional value, n indicate emotion word number, m
Indicate the number of emoticon.
Further, each subordinate sentence in the microblog emotional dictionary according to structure, and determining pending microblogging text
Clause type and the clause type influence weights to Sentiment orientation, determine the pending microblogging text Sentiment orientation value it
Afterwards, the method further includes：
The Sentiment orientation value of the pending microblogging text of output is compared with the Sentiment orientation value marked in advance
Verification.
Further, each subordinate sentence in the microblog emotional dictionary according to structure, and determining pending microblogging text
Clause type and the clause type influence weights to Sentiment orientation, determine the pending microblogging text Sentiment orientation value it
Afterwards, the method further includes：
Build user interface.
The above-mentioned technical proposal of the present invention has the beneficial effect that：
In said program, microblog emotional dictionary is built；Obtain pending microblogging text；Classify according to pre-set clause
Rule set determines the clause type and the clause type of each subordinate sentence in the pending microblogging text obtained to Sentiment orientation
Influence weights；According to the microblog emotional dictionary of structure, and in the pending microblogging text that determines the clause type of each subordinate sentence and
The clause type influences weights to Sentiment orientation, determines the Sentiment orientation value of the pending microblogging text.In this way, construction sentence
Each subordinate sentence of formula classifying rules set pair microblogging text is classified and adds clause type influences weights to Sentiment orientation, utilizes emotion
Tendency influences right-value optimization sentiment analysis as a result, to realize in fine granularity to the sentiment analysis of microblogging text so that obtains
Sentiment analysis result it is more accurate.
Description of the drawings
Fig. 1 is the flow diagram one that microblog emotional provided in an embodiment of the present invention determines method；
Fig. 2 is the flow diagram two that microblog emotional provided in an embodiment of the present invention determines method；
Fig. 3 is user interface schematic diagram provided in an embodiment of the present invention.
Specific implementation mode
To keep the technical problem to be solved in the present invention, technical solution and advantage clearer, below in conjunction with attached drawing and tool
Body embodiment is described in detail.
The present invention ignores microblogging text particularity for the existing sentiment analysis method based on machine learning and inclines to emotion
Tropism influences, the problem for causing sentiment analysis result not accurate enough, provides a kind of microblog emotional and determines method.
As shown in Figure 1, microblog emotional provided in an embodiment of the present invention determines method, including：
S101 builds microblog emotional dictionary；
S102 obtains pending microblogging text；
S103 determines in the pending microblogging text obtained each point according to pre-set clause classifying rules collection
The clause type and the clause type of sentence influence weights to Sentiment orientation；
S104, according to the clause class of each subordinate sentence in the microblog emotional dictionary of structure, and determining pending microblogging text
Type and the clause type influence weights to Sentiment orientation, determine the Sentiment orientation value of the pending microblogging text.
Microblog emotional described in the embodiment of the present invention determines method, builds microblog emotional dictionary；Obtain pending microblogging text
This；According to pre-set clause classifying rules collection, the clause class of each subordinate sentence in the pending microblogging text obtained is determined
Type and the clause type influence weights to Sentiment orientation；According to the microblog emotional dictionary of structure, and determine pending micro-
The clause type of each subordinate sentence and the clause type influence weights to Sentiment orientation in blog article sheet, determine the pending microblogging text
This Sentiment orientation value.In this way, construction each subordinate sentence of clause classifying rules set pair microblogging text is classified and adds clause type
Weights are influenced on Sentiment orientation, influence right-value optimization sentiment analysis using Sentiment orientation as a result, right in fine granularity to realize
The sentiment analysis of microblogging text so that obtained sentiment analysis result is more accurate.
In aforementioned microblog emotional determines the specific implementation mode of method, further, the structure microblog emotional dictionary
Including：
Microblogging vocabulary, microblogging emoticon are extracted, according to the microblogging vocabulary of extraction, microblogging emoticon to basic emotion word
Allusion quotation is extended, and obtains microblog emotional dictionary；
Wherein, the microblog emotional dictionary includes：Microblogging basis sentiment dictionary, network sentiment dictionary, microblogging emoticon
Dictionary, modification dictionary.
In microblogging text, when user gives opinion with regard to something and carries apparent Sentiment orientation, these Sentiment orientations
Often embodied by emotion word.When building microblog emotional dictionary, on the one hand will summarize to current existing sentiment dictionary remittance
Always, network vocabulary, emoticon, the microblogging neologisms etc. that consider to occur in microblogging text are on the other hand also needed to.
In the present embodiment, extraction micro blog network vocabulary, microblogging neologisms (can be the microblogging words occurred in preset time period
Language) and microblogging emoticon, using the micro blog network vocabulary, microblogging neologisms and microblogging emoticon of extraction to obtaining in advance
Existing basic sentiment dictionary be extended, obtain microblog emotional dictionary.
In the present embodiment, the microblog emotional dictionary includes：Microblogging basis sentiment dictionary, network sentiment dictionary, microblogging table
Feelings symbol dictionary, modification dictionary, the modification dictionary include：Negative word dictionary and degree adverb dictionary.
In aforementioned microblog emotional determines the specific implementation mode of method, further, pending microblogging text is being obtained
Later, the method further includes：
The pending microblogging text of acquisition is pre-processed and carries out Emotion tagging；
Wherein, described pre-process includes：Noise reduction, participle and stop words filter operation, and user is added during participle
Custom Dictionaries.
In the present embodiment, as shown in Fig. 2, after obtaining pending microblogging text, in order to preferably to being waited for described in acquisition
It handles microblogging text and carries out sentiment analysis, first it can be pre-processed, wherein the pretreatment includes：Noise reduction, participle and
The operations such as stop words filtering, and User Defined dictionary is added during participle, to improve the standard that it segments microblogging text
True property.
In the present embodiment, in order to verify the correctness that the microblog emotional described in the present embodiment determines method, also need to obtaining
The pending microblogging text carry out Emotion tagging, the Sentiment orientation value of mark is divided into 1, -1,0, indicates feeling polarities respectively：
Positive, negative sense and neutrality.
In aforementioned microblog emotional determines the specific implementation mode of method, further, according to pre-set clause
Classifying rules collection determines the clause type and the clause type of each subordinate sentence in the pending microblogging text obtained to emotion
Before tendency influences weights, the method further includes：
Subordinate sentence algorithm is constructed, subordinate sentence is carried out to the pending microblogging text of acquisition, wherein the subordinate sentence algorithm is needle
Microblogging text expression lack of standard is configured.
In the present embodiment, the particularity of microblogging text is considered, for pretreated pending microblogging text, add to micro-
The considerations of this particularity of blog article, realizes subordinate sentence optimization, for example, punctuation mark reuses or the nonstandard statements such as missing.For example, this
In embodiment, it is contemplated that certain customers have when editing text use "...”、“！！！", the nonstandard statement such as " " when, point
The punctuate of special statement is identified when sentence according to the subordinate sentence algorithm of construction so that subordinate sentence is stated closer to user to be accustomed to, to solve
The certainly nonstandard problem of form of presentation.
In the present embodiment, after subordinate sentence optimization, it is contemplated that the semantic meaning representation rule of microblogging text, the i.e. clause of text expression
Relationship introduces clause classifying rules collection, and more acurrate classification is carried out to text clause, obtain more accurate clause type and each
Clause type influences weights to Sentiment orientation, in this way, microblogging text is sub-divided into subordinate sentence rank from whole sentence rank, realizes more particulate
The sentiment analysis of degree.
In the present embodiment, the clause relationship includes simple sentence and complex sentence, simple sentence such as exclamative sentence, interrogative sentence, affirmative
Sentence etc., complex sentence such as confirmative question, turnover sentence, hypothetical sentence, selection sentence, progressive sentence, concession sentence, summary sentence etc., simple sentence and complexity
The type that sentence includes can be extended according to practical application scene.
In the present embodiment, each clause type influences weights to Sentiment orientation and is determined by clause classifying rules, by continuous
Training verification, can adjust clause type on Sentiment orientation influence weighting parameter, finally obtain more accurate classification results.
In the present embodiment, one section of pending microblogging text p is inputted, the microblog emotional described in the present embodiment determines the defeated of method
Go out for：The Sentiment orientation value (pos, neg, flag) of the pending microblogging text p and each subordinate sentence siSentiment orientation value
Gather [[S1pos,S1neg],[S2pos,S2neg],...,[Snpos,SnNeg]], wherein the positive feeling polarities value of pos expressions,
Neg indicates that negative sense feeling polarities value, flag indicate Sentiment orientation value.
In the present embodiment, Sentiment orientation value E (p)=E (s of the pending microblogging text p1)+E(s2)+…+E(sn)；
Wherein, E (p) indicates the Sentiment orientation value of pending microblogging text p, E (si) indicate in pending microblogging text p the
I subordinate sentence siSentiment orientation value；E(si)=Wseni·Epri(si), WseniFor i-th of subordinate sentence siClause type to emotion
The influence weights of tendency, WseniIt is determined by clause classifying rules, Epri (si) it is i-th of subordinate sentence siBasic emotion value.
In the present embodiment, the basic emotion value Epri (si) be expressed as：
Wherein, Wdeg is degree adverb weights, and Wno is negative word weights, E (wt) indicate i-th of subordinate sentence siIn t-th of feelings
Feel the emotional value of word, E (emjj) indicate i-th of subordinate sentence siIn j-th of emoticon emotional value, n indicate emotion word number, m
Indicate the number of emoticon.
In aforementioned microblog emotional determines the specific implementation mode of method, further, in the microblog emotional according to structure
The clause type and the clause type of each subordinate sentence influence to weigh on Sentiment orientation in dictionary, and determining pending microblogging text
It is worth, after the Sentiment orientation value for determining the pending microblogging text, the method further includes：
The Sentiment orientation value of the pending microblogging text of output is compared with the Sentiment orientation value marked in advance
Verification.
In the present embodiment, the microblog emotional described in the present embodiment is determined to the pending microblogging text of method output
Sentiment orientation value and the Sentiment orientation value marked in advance carry out contrast verification, for example, traditional standard may be used：Accuracy rate,
The evaluation criterion of recall rate and F values as algorithm, specifically：
The accuracy rate and recall rate of all correct result quantity and standard results quantity are calculated, and according to accuracy rate and is recalled
Rate calculates F values.
In the present embodiment, in order to verify the validity that the microblog emotional described in the present embodiment determines method, one group has been carried out
With the contrast experiment of the sentiment analysis method based on machine learning, experimental result is as shown in table 1.
1 microblog emotional of table determines method and the existing sentiment analysis methods experiment Comparative result based on machine learning
As can be seen from Table 1, the side that the embodiment of the present invention passes through consideration microblogging text subordinate sentence and clause classifying rules collection
Method has greatly improved for microblog text affective polarity classification accuracy compared to conventional sorting methods.
In aforementioned microblog emotional determines the specific implementation mode of method, further, in the microblog emotional according to structure
The clause type and the clause type of each subordinate sentence influence to weigh on Sentiment orientation in dictionary, and determining pending microblogging text
It is worth, after the Sentiment orientation value for determining the pending microblogging text, the method further includes：
Build user interface.
As shown in Table 1, the microblog emotional described in the present embodiment determines that method can complete the Sentiment orientation point of microblogging text
Analysis builds user interface, the interactive experience of friendly is provided to the user, but also it can be better on this basis
It determines that method carries out emotional orientation analysis with the microblog emotional described in the present embodiment, realizes the practicality and realistic meaning.
In the present embodiment, the user interface of structure is as shown in figure 3, user inputs need in the pending data frame of left side
Microblogging text to be processed selects the operation to be carried out in middle column, handling result that will be shown in right side exports results box,
Daily record frame is used for recording user's operation information.
To sum up, the microblog emotional described in the present embodiment determines method, improves sentiment dictionary by extension and combines special sentence
Influence of the type to Text Orientation judges microblog emotional tendentiousness, can effectively analyze the Sentiment orientation of microblogging text, and
It supports user friendly interaction, easy to operate, practicality and high efficiency, there is very strong practical value and realistic meaning.
The microblog emotional provided in an embodiment of the present invention determines that method has the advantages that：
A) embodiment of the present invention proposes a kind of microblog emotional analysis method based on classifying rules collection, and there have to be very high accurate
Rate.
B) embodiment of the present invention adds clause when handling microblogging text particularity influences weights, Yi Jiyong to Sentiment orientation
The expression way influence lack of standardization to subordinate sentence in family can be good at analyzing microblogging text in fine granularity.
C) embodiment of the present invention provides a friendly user interface, and easy to operate, practicality and high efficiency has very strong
Practical value and realistic meaning.
It should be noted that herein, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, without departing from the principles of the present invention, several improvements and modifications can also be made, these improvements and modifications
It should be regarded as protection scope of the present invention.
Claims (10)
1. a kind of microblog emotional determines method, which is characterized in that including：
Build microblog emotional dictionary；
Obtain pending microblogging text；
According to pre-set clause classifying rules collection, the clause class of each subordinate sentence in the pending microblogging text obtained is determined
Type and the clause type influence weights to Sentiment orientation；
According to the clause type of each subordinate sentence and the sentence in the microblog emotional dictionary of structure, and determining pending microblogging text
Formula type influences weights to Sentiment orientation, determines the Sentiment orientation value of the pending microblogging text.
2. microblog emotional according to claim 1 determines method, which is characterized in that the structure microblog emotional dictionary packet
It includes：
Extract microblogging vocabulary, microblogging emoticon, according to the microblogging vocabulary of extraction, microblogging emoticon to basic sentiment dictionary into
Row extension, obtains microblog emotional dictionary；
Wherein, the microblog emotional dictionary includes：Microblogging basis sentiment dictionary, network sentiment dictionary, microblogging emoticon dictionary,
Modify dictionary.
3. microblog emotional according to claim 2 determines method, which is characterized in that the modification dictionary includes：Negative word
Dictionary and degree adverb dictionary.
4. microblog emotional according to claim 1 determines method, which is characterized in that obtain pending microblogging text it
Afterwards, the method further includes：
The pending microblogging text of acquisition is pre-processed and carries out Emotion tagging；
Wherein, described pre-process includes：Noise reduction, participle and stop words filter operation, and addition user makes by oneself during participle
Adopted dictionary.
5. microblog emotional according to claim 1 determines method, which is characterized in that classify according to pre-set clause
Rule set determines the clause type and the clause type of each subordinate sentence in the pending microblogging text obtained to Sentiment orientation
Before influencing weights, the method further includes：
Subordinate sentence algorithm is constructed, subordinate sentence is carried out to the pending microblogging text of acquisition, wherein the subordinate sentence algorithm is for micro-
What rich text expression lack of standard was configured.
6. microblog emotional according to claim 1 determines method, which is characterized in that the clause type includes：Simple sentence
And complex sentence；
Wherein, the simple sentence includes：Exclamative sentence, interrogative sentence, assertive sentence, the complex sentence include：Confirmative question, turnover sentence, vacation
If sentence, progressive sentence, concession sentence, summarizes sentence at selection sentence.
7. microblog emotional according to claim 1 determines method, which is characterized in that the emotion of the pending microblogging text
Propensity value E (p)=E (s1)+E(s2)+…+E(sn)；
Wherein, E (p) indicates the Sentiment orientation value of pending microblogging text p, E (si) i-th point is indicated in pending microblogging text p
Sentence siSentiment orientation value；E(si)=Wseni·Epri(si), WseniFor i-th of subordinate sentence siClause type to Sentiment orientation
Influence weights, WseniIt is determined by clause classifying rules, Epri (si) it is i-th of subordinate sentence siBasic emotion value.
8. microblog emotional according to claim 7 determines method, which is characterized in that the basic emotion value Epri (si) table
It is shown as：
Wherein, Wdeg is degree adverb weights, and Wno is negative word weights, E (wt) indicate i-th of subordinate sentence siIn t-th of emotion word
Emotional value, E (emjj) indicate i-th of subordinate sentence siIn j-th of emoticon emotional value, n indicate emotion word number, m indicate
The number of emoticon.
9. microblog emotional according to claim 4 determines method, which is characterized in that in the microblog emotional word according to structure
The clause type and the clause type of each subordinate sentence influence to weigh on Sentiment orientation in allusion quotation, and determining pending microblogging text
It is worth, after the Sentiment orientation value for determining the pending microblogging text, the method further includes：
The Sentiment orientation value of the pending microblogging text of output is subjected to contrast verification with the Sentiment orientation value marked in advance.
10. microblog emotional according to claim 1 determines method, which is characterized in that in the microblog emotional word according to structure
The clause type and the clause type of each subordinate sentence influence to weigh on Sentiment orientation in allusion quotation, and determining pending microblogging text
It is worth, after the Sentiment orientation value for determining the pending microblogging text, the method further includes：
Build user interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810372663.4A CN108647257A (en) | 2018-04-24 | 2018-04-24 | A kind of microblog emotional determines method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810372663.4A CN108647257A (en) | 2018-04-24 | 2018-04-24 | A kind of microblog emotional determines method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108647257A true CN108647257A (en) | 2018-10-12 |
Family
ID=63747035
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810372663.4A Pending CN108647257A (en) | 2018-04-24 | 2018-04-24 | A kind of microblog emotional determines method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108647257A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113378576A (en) * | 2021-05-08 | 2021-09-10 | 重庆航天信息有限公司 | Food safety data mining method |
| CN113378578A (en) * | 2021-05-08 | 2021-09-10 | 重庆航天信息有限公司 | Food and medicine public opinion analysis method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104881402A (en) * | 2015-06-02 | 2015-09-02 | 北京京东尚科信息技术有限公司 | Method and device for analyzing semantic orientation of Chinese network topic comment text |
| KR20150107178A (en) * | 2014-03-13 | 2015-09-23 | 울산대학교 산학협력단 | Method for analysis of korean dependency relation by homograph disambiguation |
| CN105843796A (en) * | 2016-03-28 | 2016-08-10 | 北京邮电大学 | Microblog emotional tendency analysis method and device |
-
2018
- 2018-04-24 CN CN201810372663.4A patent/CN108647257A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150107178A (en) * | 2014-03-13 | 2015-09-23 | 울산대학교 산학협력단 | Method for analysis of korean dependency relation by homograph disambiguation |
| CN104881402A (en) * | 2015-06-02 | 2015-09-02 | 北京京东尚科信息技术有限公司 | Method and device for analyzing semantic orientation of Chinese network topic comment text |
| CN105843796A (en) * | 2016-03-28 | 2016-08-10 | 北京邮电大学 | Microblog emotional tendency analysis method and device |
Non-Patent Citations (1)
| Title |
|---|
| 王洪伟，郑丽娟，尹裴，何绍义: "基于句子级情感的中文网络评论的情感极性分类", 《管理科学学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113378576A (en) * | 2021-05-08 | 2021-09-10 | 重庆航天信息有限公司 | Food safety data mining method |
| CN113378578A (en) * | 2021-05-08 | 2021-09-10 | 重庆航天信息有限公司 | Food and medicine public opinion analysis method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107092596B (en) | Text emotion analysis method based on attention CNNs and CCR | |
| CN109446404B (en) | Method and device for analyzing emotion polarity of network public sentiment | |
| CN105183833B (en) | Microblog text recommendation method and device based on user model | |
| CN102929861B (en) | Method and system for calculating text emotion index | |
| CN105183717B (en) | A kind of OSN user feeling analysis methods based on random forest and customer relationship | |
| CN104268160A (en) | Evaluation object extraction method based on domain dictionary and semantic roles | |
| CN106202584A (en) | A kind of microblog emotional based on standard dictionary and semantic rule analyzes method | |
| CN107609132A (en) | One kind is based on Ontology storehouse Chinese text sentiment analysis method | |
| Peng et al. | Human–machine dialogue modelling with the fusion of word-and sentence-level emotions | |
| CN107688576B (en) | Construction and tendency classification method of CNN-SVM model | |
| CN104199845B (en) | Line Evaluation based on agent model discusses sensibility classification method | |
| CN110750648A (en) | Text emotion classification method based on deep learning and feature fusion | |
| CN109271634A (en) | A kind of microblog text affective polarity check method based on user feeling tendency perception | |
| CN105740382A (en) | Aspect classification method for short comment texts | |
| CN107305545A (en) | A kind of recognition methods of the network opinion leader based on text tendency analysis | |
| CN110472115B (en) | Social network text emotion fine-grained classification method based on deep learning | |
| CN104778240A (en) | Micro blog text data classification method on basis of multi-feature fusion | |
| CN103744838B (en) | A kind of Chinese emotion digest system and method for measuring main flow emotion information | |
| CN110119443A (en) | A kind of sentiment analysis method towards recommendation service | |
| CN113673239A (en) | Hotel comment emotion polarity classification method based on emotion dictionary weighting | |
| Zhang et al. | Exploring deep recurrent convolution neural networks for subjectivity classification | |
| Arif et al. | A machine learning based approach for opinion mining on social network data | |
| KR101326313B1 (en) | Method of classifying emotion from multi sentence using context information | |
| Suresh et al. | Mining of customer review feedback using sentiment analysis for smart phone product | |
| Liu et al. | A reliable cross-site user generated content modeling method based on topic model |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181012 |