CN110297986A - A kind of Sentiment orientation analysis method of hot microblog topic - Google Patents
A kind of Sentiment orientation analysis method of hot microblog topic Download PDFInfo
- Publication number
- CN110297986A CN110297986A CN201910540279.5A CN201910540279A CN110297986A CN 110297986 A CN110297986 A CN 110297986A CN 201910540279 A CN201910540279 A CN 201910540279A CN 110297986 A CN110297986 A CN 110297986A
- Authority
- CN
- China
- Prior art keywords
- text
- topic
- word
- microblog
- hot
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 26
- 230000008451 emotion Effects 0.000 claims abstract description 26
- 230000002996 emotional effect Effects 0.000 claims abstract description 22
- 238000013145 classification model Methods 0.000 claims abstract description 11
- 239000000284 extract Substances 0.000 claims abstract description 8
- 238000000605 extraction Methods 0.000 claims description 15
- 238000000034 method Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 239000000523 sample Substances 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 2
- 239000012497 inhomogeneous sample Substances 0.000 claims description 2
- 238000010801 machine learning Methods 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000005259 measurement Methods 0.000 claims description 2
- 230000004044 response Effects 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 2
- 230000015654 memory Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- 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
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of Sentiment orientation analysis methods of hot microblog topic to acquire the text information of topic according to specified topic;Extract subjective microblogging evaluates word emotion information relevant to hot microblog topic;During emotional semantic classification, for the accuracy for improving microblog text affective multivariate classification, the microblog text affective multivariate classification model based on SVM-BILSTM of proposition;Emotional orientation analysis is finally made according to the analysis of front and result.The beneficial effects of the invention are as follows text is acquired by specified topic, extract emotion information, using emotion multivariate classification model can real-time response public sentiment event Sentiment orientation, improve the response speed of public sentiment event, it is faster efficiently.
Description
Technical field
The invention belongs to technical field of data processing, are related to a kind of Sentiment orientation analysis method of hot microblog topic.
Background technique
The Sentiment orientation analysis model of hot microblog topic mainly includes network information gathering technology, data prediction mistake
Chinese analysis used in journey and part-of-speech tagging method, character representation, feature extraction and the file classification method of text are finally
Shot and long term Memory Neural Networks algorithm in deep learning.Network information gathering technology (Network Information
Collection Technology), it is a kind of computer skill of automatic collection internet data information according to certain rules
Art.Usually using one or more initial URL as starting point, the fetching instruction according to http protocol format is sent by each generic port
Acquire the information [24] in webpage.Such repetitive cycling carries out traversal search to internet information, until collecting all numbers
Until.For different medium types, the acquisition mode of public feelings information also difference is most explosive from public feelings information
From the point of view of angle, it is concentrated mainly on news, microblogging, the big media of forum three.Chinese word segmentation (Chinese Word Segmentation)
One chinese character sequence must be cut into the process of individual word one by one, part-of-speech tagging exactly before analyzing Chinese text
(Part-of-Speech Tagging) refers to the grammatical roles for judging that each word is played the part of in a sentence.Text representation
(Text representation) refer to by certain form by text-string be expressed as numerical value that computer can be handled to
Amount.Because computer cannot directly be handled text-string, it is therefore desirable to by the Feature Words extracted herein into
Line number value or vectorization are enabled a computer to identify and be handled.Text classification (Text Categorization) refers to
It is that each text is automatically assigned in the class pre-defined under given classification system, the key data of text classification is come
Source is non-structured text, it can given text is assigned to the process of respective classes by a classifier.LSTM
(Long Short Term Memory) is the one of Recognition with Recurrent Neural Network (Recurrent Neural Network, RNN) structure
Kind, it is made of input layer, hidden layer, output layer, the input layer of traditional RNN and hidden layer are implanted to memory by LSTM network model
Include special door in memory unit in unit, i.e., input gate, forget door and out gate to control the circulation of information, only
There is the information for meeting algorithm certification that can just be left, the information not being inconsistent then passes through forgetting door and passes into silence, the meter that LSTM model is related to
Calculation is relatively more, also more complicated, so the processing to information is more flexible, it is also more powerful, it is suitable for processing and predicted time sequence
The critical event relatively long with delay of the relatively long event in middle interval.In practical applications, because language has long-term dependence
Relationship, RNN model is bad at all information before capture and reservation, there are problems that long-term dependence, and LSTM can solve
It solves the above problems.LSTM has a variety of answer in sciemtifec and technical spheres such as speech recognition, image recognition, control chat robots
With.Existing microblogging hot spot Sentiment orientation analysis generally has hysteresis quality.
Summary of the invention
The purpose of the present invention is to provide a kind of Sentiment orientation analysis method of hot microblog topic, beneficial effect of the invention
Fruit is to acquire text by specified topic, extracts emotion information, being capable of real-time response public sentiment thing using emotion multivariate classification model
Part Sentiment orientation improves the response speed of public sentiment event, faster efficient.
The technical scheme adopted by the invention is that following the steps below:
A, the data acquisition of hot microblog topic and pretreatment;According to specified topic, the text information of topic is acquired;
B, extraction subjective microblogging evaluates word emotion information relevant to hot microblog topic;Emotion information extraction process
In, for the quality for improving the emotion information obtained, micro-blog emotion information is improved in conjunction with TF-IDF-COS and SVM algorithm and extracts mould
Type, to extract the extraction of subjective microblogging evaluates word emotion information relevant to hot microblog topic;
C, during emotional semantic classification, for improve microblog text affective multivariate classification accuracy, proposition based on SVM-
The microblog text affective multivariate classification model of BILSTM;
D, emotional orientation analysis is finally made according to the analysis of front and result.
Further, in step A, the data acquisition of hot microblog topic and pretreatment refer to the specific words in selected microblogging
Topic, the text information of the topic is crawled using Python tool to microblog, then to collected semi-structured information into
Row pretreatment, and then obtain plain text corpus and stored.
Further, TF-IDF-COS and SVM algorithm are as follows in step B:
TF-IDF algorithm combination cosine coefficient similarity calculating method is chosen to calculate the similarity of text and topic, is passed through
The cosine coefficient between the TF-IDF weight and word i and hot topic word T (w) of word i is calculated, and then is extracted and hot topic word
The biggish word of similarity, then text relevant to topic and the text unrelated with topic are classified by SVM algorithm, in turn
Obtain microblogging text relevant to topic;
Word frequency reflects the number that a word occurs in a document, and calculation formula is as follows:
Wherein, wiIt is expressed as i-th of vocabulary, pjIt is expressed as jth piece text, nijI-th of vocabulary is expressed as in jth piece text
The number of middle appearance, njIt is expressed as the summation of jth piece text vocabulary.
Inverse document frequency is the measurement to a word importance, describes the use scope of the word, calculation formula is such as
Under:
Wherein, m is the total number of documents of corpus, miFor in corpus include word wiNumber of documents.Meanwhile to prevent
Some uncommon word does not make the denominator of the formula be 0 in corpus, so having carried out smoothing processing to IDF, i.e. denominator is done
Add 1 processing, make the word for not having to occur in corpus also available one suitable IDF value,
TF-IDF=word frequency (TF) × inverse document frequency (IDF)
In Text Representation, every microblogging text can be indicated with the feature of word in microblogging, the spy of these words
Sign and its weight constitute the vector (W in space1,j,W2,j,W3,j,···,Wn,j), wherein Wi,jIt is entry i in microblogging text Dj
In weight, calculate it is as follows:
Wi,j=TFi,j×IDFi×COSi。
Further, SVM-BILSTM algorithm is exactly a kind of algorithm that SVM and BILSTM is combined in step C, utilizes SVM-
The microblog text affective multivariate classification model of BILSTM, output stage just, calibration, forward direction, negative sense, relatively negative, extremely minus 6 emotion classes
Not.
The calculation method of BILSTM is represented by
st=f (Uxt+Wst-1)
s′t=f (U ' xt+W′s′t+1)
Wherein, weight U and U ', W and W ', V and V ' are different weight matrix when BILSTM is calculated respectively, and (W, U) is positive
To the weight for being input to hidden layer when calculating, to the weight of hidden layer, (V, V ') is hidden layer when (U ', W ') is retrospectively calculate
Weight of the BILSTM hidden layer to output layer.
SVM refers to finds out the optimal separating hyper plane for meeting classificating requirement in the vector space where sample point,
It can separate inhomogeneous sample, maximize class interval, it is a kind of by supervised learning in machine learning
(SupervisedLearning) mode carries out the generalized linear classifier of binary classification to data, depends on different core
Function, common kernel function are as follows:
K(xi,yi)=(xi*yi)
Choose training sample set T=(xi,yi), i=1,2, n;X is input vector;Y={ 1, -1 }, yiFor xi
To class label, hyperplane equation is as follows: ω xi+ b=0
Wherein ω is normal vector, determines the direction of hyperplane, and b is displacement item, determines the distance between hyperplane and origin.
Finally obtain training sample kernel function expansion are as follows:
I=1,2, n;X is input vector;Y={ 1, -1 }, yiFor xiTo class label, k is kernel function, and b is displacement
, ɑ is Lagrange multiplier.
Detailed description of the invention
Fig. 1 is that the Sentiment orientation of hot microblog topic of the invention analyzes overall framework figure;
Fig. 2 is hot microblog topic Sentiment orientation analysis model figure of the invention;
Fig. 3 is improved emotion information extraction model figure of the invention;
Fig. 4 is the sentiment classification model figure figure based on SVM-BILSTM.
Specific embodiment
The present invention is described in detail With reference to embodiment.
The Sentiment orientation analysis method of hot microblog topic of the present invention is as follows:
A, the data acquisition of hot microblog topic and pretreatment;According to specified topic, the text information of topic is acquired;It is micro-
The data acquisition of rich hot topic and pretreatment refer to the specific topics in selected microblogging, utilize Python tool to microblog
The text information of the topic is crawled, then collected semi-structured information is pre-processed, and then obtains plain text corpus
It is stored.
B, extraction subjective microblogging evaluates word emotion information relevant to hot microblog topic;Emotion information extraction process
In, for the quality for improving the emotion information obtained, micro-blog emotion information is improved in conjunction with TF-IDF-COS and SVM algorithm and extracts mould
Type, to extract the extraction of subjective microblogging evaluates word emotion information relevant to hot microblog topic;TF-IDF-COS and SVM is calculated
Method is as follows:
TF-IDF algorithm combination cosine coefficient similarity calculating method is chosen to calculate the similarity of text and topic, is passed through
The cosine coefficient between the TF-IDF weight and word i and hot topic word T (w) of word i is calculated, and then is extracted and hot topic word
The biggish word of similarity, then text relevant to topic and the text unrelated with topic are classified by SVM algorithm, in turn
Obtain microblogging text relevant to topic;
TF-IDF=word frequency (TF) × inverse document frequency (IDF)
In Text Representation, every microblogging text can be indicated with the feature of word in microblogging, the spy of these words
Sign and its weight constitute the vector (W in space1,j,W2,j,W3,j,···,Wn,j), wherein Wi,jIt is entry i in microblogging text Dj
In weight, calculate it is as follows:
Wi,j=TFi,j×IDFi×COSi。
TF indicates the frequency that a word occurs in a document, and IDF indicates the inverse for the document of specific word occur.
C, during emotional semantic classification, for improve microblog text affective multivariate classification accuracy, proposition based on SVM-
The microblog text affective multivariate classification model of BILSTM;SVM-BILSTM algorithm is exactly a kind of calculation that SVM and BILSTM is combined
Method, using the microblog text affective multivariate classification model of SVM-BILSTM, output stage just, it is calibration, forward direction, negative sense, relatively negative, extremely negative
6 emotional categories.
D, emotional orientation analysis is finally made according to the analysis of front and result.
The result of comprehensive two steps of C and D obtains last Sentiment orientation analysis conclusion.
Fig. 1 is that the Sentiment orientation of hot microblog topic of the invention analyzes overall framework figure;Fig. 2 is microblogging heat of the invention
Point topic Sentiment orientation analysis model figure;Fig. 3 is improved emotion information extraction model figure of the invention;Fig. 4 be based on
The sentiment classification model figure figure of SVM-BILSTM.It can be seen that the present embodiment by the Sentiment orientation of microblogging text from overall framework figure
It is divided into data acquisition and pretreatment, emotion information extraction, emotional semantic classification, Sentiment orientation 4 modules of analysis.
It when carrying out data acquisition, needs first to select a topic, is then crawled using Python tool to microblog
Text information of the topic, including extract text punctuation mark, emoticon etc., then to collected semi-structured information into
Row pretreatment, extraction, data cleansing, participle and part-of-speech tagging including punctuation mark and emoticon remove stop words etc., in turn
Plain text corpus is obtained to be stored.Valuable evaluation pair is extracted using certain technical method in emotion information abstraction module
As and evaluates word, with obtain it is not only related to topic, but also be subjectivity express microblogging text as this paper emotional semantic classification task
Experimental data.In the emotion multivariate classification task that emotional semantic classification module is mainly microblogging text, firstly, subjective and objective before
It is slightly modified on the basis of microblog text affective characteristic of division, it combines to open and thinks improved octuple Feature Words and He Yue, Xiao Min, opens
The 10 class feeling polarities features that the moon proposes increase punctuation mark as one of microblog text affective feature, will commonly use in microblogging
Punctuation mark, positive and negative two class is classified as by the way of handmarking, is stored as " [+] " " [-] " respectively, and then be equipped with
12 kinds of emotional semantic classification features, and using this 12 kinds of emotional semantic classification features as emotional semantic classification Feature Words, to improve the standard of emotional semantic classification
True property.It is mainly to study the tendentiousness classification problem of microblogging text in Sentiment orientation analysis module, that is, is carrying out microblogging text
After emotional semantic classification, emotional intensity is added and calculates, Sentiment orientation is included in different classification states.According to emotional semantic classification module
Output as a result, i.e. extremely just, calibration, forward direction, negative sense, relatively negative, extremely minus 6 classifications, respectively each emotional category assigns corresponding
Weight, with facilitate calculate hot microblog topic emotional intensity, and then determine hot microblog topic Sentiment orientation state.
The above is only not to make limit in any form to the present invention to better embodiment of the invention
System, any simple modification that embodiment of above is made according to the technical essence of the invention, equivalent variations and modification,
Belong in the range of technical solution of the present invention.
Claims (4)
1. a kind of Sentiment orientation analysis method of hot microblog topic, it is characterised in that follow the steps below:
A, the data acquisition of hot microblog topic and pretreatment;According to specified topic, the text information of topic is acquired;
B, extraction subjective microblogging evaluates word emotion information relevant to hot microblog topic;In emotion information extraction process, it is
The quality for improving the emotion information obtained improves micro-blog emotion information extraction model in conjunction with TF-IDF-COS and SVM algorithm, comes
Extract the extraction of subjective microblogging evaluates word emotion information relevant to hot microblog topic;
C, during emotional semantic classification, for improve microblog text affective multivariate classification accuracy, proposition based on SVM-
The microblog text affective multivariate classification model of BILSTM;
D, emotional orientation analysis is finally made according to the analysis of front and result.
2. according to a kind of Sentiment orientation analysis method of hot microblog topic described in claim 1, it is characterised in that: the step
In A, the data acquisition of hot microblog topic and pretreatment refer to the specific topics in selected microblogging, using Python tool to micro-
Rich platform crawls the text information of the topic, then pre-processes to collected semi-structured information, and then obtains pure text
This corpus is stored.
3. according to a kind of Sentiment orientation analysis method of hot microblog topic described in claim 1, it is characterised in that: the step
TF-IDF-COS and SVM algorithm are as follows in B:
TF-IDF algorithm combination cosine coefficient similarity calculating method is chosen to calculate the similarity of text and topic, passes through calculating
Cosine coefficient between the TF-IDF weight and word i and hot topic word T (w) of word i, and then extract similar to hot topic word
Biggish word is spent, then text relevant to topic and the text unrelated with topic are classified by SVM algorithm, and then obtains
Microblogging text relevant to topic, word frequency reflect the number that a word occurs in a document, and calculation formula is as follows:
Wherein, wiIt is expressed as i-th of vocabulary, pjIt is expressed as jth piece text, nijI-th of vocabulary is expressed as to go out in jth piece text
Existing number, njIt is expressed as the summation of jth piece text vocabulary;Inverse document frequency is the measurement to a word importance, description
The use scope of the word, calculation formula are as follows:
Wherein, m is the total number of documents of corpus, miFor in corpus include word wiNumber of documents, meanwhile, it is a certain to prevent
A uncommon word does not make the denominator of the formula be 0 in corpus, so having carried out smoothing processing to IDF, i.e., denominator, which is done, adds at 1
Reason makes the word for not having to occur in corpus also available one suitable IDF value,
TF-IDF=word frequency (TF) × inverse document frequency (IDF)
In Text Representation, every microblogging text can be indicated with the feature of word in microblogging, the feature of these words and
Its weight constitutes the vector (W in space1,j,W2,j,W3,j,…,Wn,j), wherein Wi,jIt is entry i in microblogging text DjIn power
Weight calculates as follows:
Wi,j=TFi,j×IDFi×COSi。
4. according to a kind of Sentiment orientation analysis method of hot microblog topic described in claim 1, it is characterised in that: the step
SVM-BILSTM algorithm is exactly a kind of algorithm that SVM and BILSTM is combined in C, utilizes the microblog text affective of SVM-BILSTM
Multivariate classification model, output stage just, calibration, forward direction, negative sense, relatively negative, extremely minus 6 emotional categories, the calculation method of BILSTM is
st=f (Uxt+Wst-1)
s′t=f (U ' xt+W′s′t+1)
Wherein, weight U and U ', W and W ', V and V ' are different weight matrix when BILSTM is calculated respectively, and W, U are positive calculate
When be input to the weight of hidden layer, hidden layer arrives the weight of hidden layer when U ', W ' are retrospectively calculate, and V, V ' are BILSTM hidden layer
To the weight of output layer, SVM refers to finds out the optimal classification for meeting classificating requirement in the vector space where sample point
Hyperplane, it can separate inhomogeneous sample, maximize class interval, it is a kind of by supervision in machine learning
Habit mode carries out the generalized linear classifier of binary classification to data, depends on different kernel functions:
K(xi,yi)=(xi*yi)
Choose training sample set T=(xi,yi), i=1,2 ..., n;X is input vector;Y={ 1, -1 }, yiFor xiTo class label,
Hyperplane equation is as follows: ω xi+ b=0
Wherein ω is normal vector, determines the direction of hyperplane, and b is displacement item, determines the distance between hyperplane and origin, finally
Obtain training sample kernel function expansion are as follows:
I=1,2 ..., n;X is input vector;Y={ 1, -1 }, yiFor xiTo class label, k is kernel function, and b is displacement item, and ɑ is to draw
Ge Lang multiplier.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910540279.5A CN110297986A (en) | 2019-06-21 | 2019-06-21 | A kind of Sentiment orientation analysis method of hot microblog topic |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910540279.5A CN110297986A (en) | 2019-06-21 | 2019-06-21 | A kind of Sentiment orientation analysis method of hot microblog topic |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110297986A true CN110297986A (en) | 2019-10-01 |
Family
ID=68028413
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910540279.5A Pending CN110297986A (en) | 2019-06-21 | 2019-06-21 | A kind of Sentiment orientation analysis method of hot microblog topic |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110297986A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111199158A (en) * | 2019-12-30 | 2020-05-26 | 沈阳民航东北凯亚有限公司 | Method and device for scoring civil aviation customer service |
CN111986702A (en) * | 2020-07-31 | 2020-11-24 | 中国地质大学(武汉) | Speaker mental impedance phenomenon recognition method based on voice signal processing |
CN112214991A (en) * | 2020-10-10 | 2021-01-12 | 上海海事大学 | Microblog text position detection method based on multi-feature fusion weighting |
CN115934951A (en) * | 2023-02-16 | 2023-04-07 | 湖南大学 | Network hot topic user emotion prediction method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107315778A (en) * | 2017-05-31 | 2017-11-03 | 温州市鹿城区中津先进科技研究院 | A kind of natural language the analysis of public opinion method based on big data sentiment analysis |
CN107943800A (en) * | 2016-10-09 | 2018-04-20 | 郑州大学 | A kind of microblog topic public sentiment calculates the method with analysis |
CN108959383A (en) * | 2018-05-31 | 2018-12-07 | 平安科技(深圳)有限公司 | Analysis method, device and the computer readable storage medium of network public-opinion |
CN109271634A (en) * | 2018-09-17 | 2019-01-25 | 重庆理工大学 | A kind of microblog text affective polarity check method based on user feeling tendency perception |
CN109446404A (en) * | 2018-08-30 | 2019-03-08 | 中国电子进出口有限公司 | A kind of the feeling polarities analysis method and device of network public-opinion |
CN109684646A (en) * | 2019-01-15 | 2019-04-26 | 江苏大学 | A kind of microblog topic sentiment analysis method based on topic influence |
-
2019
- 2019-06-21 CN CN201910540279.5A patent/CN110297986A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107943800A (en) * | 2016-10-09 | 2018-04-20 | 郑州大学 | A kind of microblog topic public sentiment calculates the method with analysis |
CN107315778A (en) * | 2017-05-31 | 2017-11-03 | 温州市鹿城区中津先进科技研究院 | A kind of natural language the analysis of public opinion method based on big data sentiment analysis |
CN108959383A (en) * | 2018-05-31 | 2018-12-07 | 平安科技(深圳)有限公司 | Analysis method, device and the computer readable storage medium of network public-opinion |
CN109446404A (en) * | 2018-08-30 | 2019-03-08 | 中国电子进出口有限公司 | A kind of the feeling polarities analysis method and device of network public-opinion |
CN109271634A (en) * | 2018-09-17 | 2019-01-25 | 重庆理工大学 | A kind of microblog text affective polarity check method based on user feeling tendency perception |
CN109684646A (en) * | 2019-01-15 | 2019-04-26 | 江苏大学 | A kind of microblog topic sentiment analysis method based on topic influence |
Non-Patent Citations (1)
Title |
---|
司马海峰等: "《遥感图像分类中的智能计算方法》", 31 January 2018 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111199158A (en) * | 2019-12-30 | 2020-05-26 | 沈阳民航东北凯亚有限公司 | Method and device for scoring civil aviation customer service |
CN111986702A (en) * | 2020-07-31 | 2020-11-24 | 中国地质大学(武汉) | Speaker mental impedance phenomenon recognition method based on voice signal processing |
CN111986702B (en) * | 2020-07-31 | 2022-11-04 | 中国地质大学(武汉) | Speaker psychological impedance phenomenon identification method based on voice signal processing |
CN112214991A (en) * | 2020-10-10 | 2021-01-12 | 上海海事大学 | Microblog text position detection method based on multi-feature fusion weighting |
CN112214991B (en) * | 2020-10-10 | 2024-05-14 | 上海海事大学 | Microblog text standing detection method based on multi-feature fusion weighting |
CN115934951A (en) * | 2023-02-16 | 2023-04-07 | 湖南大学 | Network hot topic user emotion prediction method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Abdullah et al. | SEDAT: sentiment and emotion detection in Arabic text using CNN-LSTM deep learning | |
CN110866117B (en) | Short text classification method based on semantic enhancement and multi-level label embedding | |
CN109933664B (en) | Fine-grained emotion analysis improvement method based on emotion word embedding | |
Ishaq et al. | Aspect-based sentiment analysis using a hybridized approach based on CNN and GA | |
CN109753566A (en) | The model training method of cross-cutting sentiment analysis based on convolutional neural networks | |
CN109726745B (en) | Target-based emotion classification method integrating description knowledge | |
CN108563638B (en) | Microblog emotion analysis method based on topic identification and integrated learning | |
CN110297986A (en) | A kind of Sentiment orientation analysis method of hot microblog topic | |
CN111078833B (en) | Text classification method based on neural network | |
CN111753058B (en) | Text viewpoint mining method and system | |
CN110750648A (en) | Text emotion classification method based on deep learning and feature fusion | |
Das et al. | Sarcasm detection on flickr using a cnn | |
CN111522908A (en) | Multi-label text classification method based on BiGRU and attention mechanism | |
Zhao et al. | ZYJ123@ DravidianLangTech-EACL2021: Offensive language identification based on XLM-RoBERTa with DPCNN | |
Xiao et al. | Chinese sentiment analysis using bidirectional LSTM with word embedding | |
CN114417851A (en) | Emotion analysis method based on keyword weighted information | |
Chang et al. | A METHOD OF FINE-GRAINED SHORT TEXT SENTIMENT ANALYSIS BASED ON MACHINE LEARNING. | |
Samih et al. | Enhanced sentiment analysis based on improved word embeddings and XGboost. | |
Farag et al. | Bullying hurts: a survey on non-supervised techniques for cyber-bullying detection | |
CN111159405B (en) | Irony detection method based on background knowledge | |
CN112507717A (en) | Medical field entity classification method fusing entity keyword features | |
Mehendale et al. | Cyber bullying detection for Hindi-English language using machine learning | |
CN115906824A (en) | Text fine-grained emotion analysis method, system, medium and computing equipment | |
CN114817533A (en) | Bullet screen emotion analysis method based on time characteristics | |
Wang et al. | Social sentiment detection of event via microblog |
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 |
Application publication date: 20191001 |
|
RJ01 | Rejection of invention patent application after publication |