CN102663046A - Sentiment analysis method oriented to micro-blog short text - Google Patents
Sentiment analysis method oriented to micro-blog short text Download PDFInfo
- Publication number
- CN102663046A CN102663046A CN201210088366XA CN201210088366A CN102663046A CN 102663046 A CN102663046 A CN 102663046A CN 201210088366X A CN201210088366X A CN 201210088366XA CN 201210088366 A CN201210088366 A CN 201210088366A CN 102663046 A CN102663046 A CN 102663046A
- Authority
- CN
- China
- Prior art keywords
- emotion
- speech
- microblogging
- sentence
- negative
- 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 94
- 238000000034 method Methods 0.000 claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000001914 filtration Methods 0.000 claims abstract description 9
- 230000008451 emotion Effects 0.000 claims description 244
- 238000013459 approach Methods 0.000 claims description 29
- 230000036651 mood Effects 0.000 claims description 27
- 230000007935 neutral effect Effects 0.000 claims description 10
- 230000008878 coupling Effects 0.000 claims description 9
- 238000010168 coupling process Methods 0.000 claims description 9
- 238000005859 coupling reaction Methods 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 244000188472 Ilex paraguariensis Species 0.000 claims description 3
- 238000002203 pretreatment Methods 0.000 claims description 3
- GNFTZDOKVXKIBK-UHFFFAOYSA-N 3-(2-methoxyethoxy)benzohydrazide Chemical compound COCCOC1=CC=CC(C(=O)NN)=C1 GNFTZDOKVXKIBK-UHFFFAOYSA-N 0.000 claims 1
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims 1
- YTAHJIFKAKIKAV-XNMGPUDCSA-N [(1R)-3-morpholin-4-yl-1-phenylpropyl] N-[(3S)-2-oxo-5-phenyl-1,3-dihydro-1,4-benzodiazepin-3-yl]carbamate Chemical compound O=C1[C@H](N=C(C2=C(N1)C=CC=C2)C1=CC=CC=C1)NC(O[C@H](CCN1CCOCC1)C1=CC=CC=C1)=O YTAHJIFKAKIKAV-XNMGPUDCSA-N 0.000 claims 1
- 238000007781 pre-processing Methods 0.000 abstract description 2
- 238000002372 labelling Methods 0.000 abstract 1
- 230000002996 emotional effect Effects 0.000 description 22
- 238000004422 calculation algorithm Methods 0.000 description 17
- 238000011160 research Methods 0.000 description 17
- 230000007246 mechanism Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 4
- 230000007306 turnover Effects 0.000 description 4
- 230000006872 improvement Effects 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 244000097202 Rathbunia alamosensis Species 0.000 description 2
- 235000009776 Rathbunia alamosensis Nutrition 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 230000000750 progressive effect Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008909 emotion recognition Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000010921 in-depth analysis Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Machine Translation (AREA)
Abstract
The invention discloses a sentiment analysis method oriented to a micro-blog short text. The method comprises the following steps: step 1, collecting micro-blog data including keywords so as to store in a database; step 2, pre-processing the micro-blog data; step 3, loading associated dictionaries; step 4, processing sentence division and filtering sentences which do not include user configuration keywords; step 5, processing word division to the sentences including the keywords and labeling parts of speech; step 6, processing dependency sentence structure analysis to the sentences including subjects by a sentence structure analyzing tool; step 7, judging the polarity of each sentence including subject words; and step 8, judging the polarity of a whole micro-blog after judging the polarities of all sentences including the subject words. According to the sentiment analysis method provided by the invention, sentiment analysis is more specific, so that users can know sentiment attitude of concerned aspects from the micro-blog.
Description
Technical field
The invention belongs to technical field of data processing, particularly, relate to a kind of emotion analytical approach towards the microblogging short text.
Background technology
The emotion analysis also claims that suggestion excavates, and refers to from text identification automatically and extraction and has tendentious attitude, suggestion and emotion.Its in recent years, subjectivity text (suggestion) Research on Mining is very active, principal feature is to analyze the subjective viewpoint that comprises in the text and calculate its semantic polarity.Because the emotion classification can solve the mixed and disorderly phenomenon of online various review information to a certain extent; Make things convenient for the user to locate information needed exactly; Therefore, the emotion classification has become a gordian technique with big practical value, is the powerful measure of organization and management data.And microblogging is because its tremendous influence power; Become more and more users and delivered first of viewpoint and emotion and select, such as to some famous person like or abhor, to the comment of some film, to the evaluation of some brand and suggestion, to the view of some current events etc.Microblogging is carried out effective emotion analysis and research can be widely used in public sentiment monitoring, brand building, advertisement marketing, information filtering, suggestion feedback, opinion poll etc.
The research work that the emotion of generally acknowledging is at present analyzed comparison system starts from (Pang et al.; 2002) based on the supervised learning method film comment text is carried out the research that emotion tendency is classified and classify to emotion tendentiousness of text based on unsupervised learning (Turney, 2002).(Pang et al.; 2002) text based N metagrammar (ngram) and part of speech characteristics such as (POS) are used naive Bayesian respectively; Maximum entropy and SVMs are divided into two types of forward and negative senses with emotion tendentiousness of text, the emotion of text is carried out the way that binary divides also use till today always.They use the film comment data set to become the test set that widely used emotion is analyzed at present in experiment simultaneously.The keyword that extracts in the text and seed speech are calculated based on some mutual information in (Turney, 2002), and (excellent, similarity poor) comes the emotion tendency of text is differentiated (SO-PMI algorithm).
Major part after this all is based on the research of (Pang et al., 2002).And comparatively speaking; (Turney et al.; 2002) though the method for the unsupervised learning that proposes is simpler in realization; But since the emotion similarity between the word be difficult to calculate accurately with the seed speech be difficult to confirm that the research that continues in the unsupervised learning direction is not a lot, but utilizes the thought of SO-PMI algorithm computation emotion tendentiousness of text but to be inherited by Many researchers.
At present, remain main flow based on the emotion analysis of supervised learning, except (Li et al.; 2009) decompose based on nonnegative matrix three; Outside (Abbasi et al., 2008) were analyzed based on the emotion of genetic algorithm, maximum supervised learning algorithm of use was a naive Bayesian; The k arest neighbors, maximum entropy and SVMs.And for the improvement of algorithm mainly at pretreatment stage to text.
A place different with text classification is exactly the sentence that really shows emotion that the emotion analysis need be extracted text sometimes.(Pang et al., 2004) based on the analysis of the neutral instance in the text, all are the sentence that really shows emotion in the text in order can to obtain as far as possible based on the selection of the subjective sentence in the text and (Wilson el al., 2009).(Abbasi et al., 2008) propose to select to analyze useful characteristic for emotion in a large amount of feature sets through the method for information gain.
And for feature selecting, except N metagrammar and part of speech characteristic, (Wilson el al., 2009) propose to mix word feature; The negative word characteristic, emotion decorative features, the emotion analysis of all kinds of syntactic features such as transference characteristic; (Abbasi et al., 2008) propose to mix sentence structure (N metagrammar, the part of speech of sentence; Punctuate) and the emotion analysis of architectural feature (length of word, the number of word in the part of speech, the architectural feature of text etc.).
Except pre-service for text; Also carried out (the Melville et al. of the research of following aspect for emotion analysis in the supervised learning; 2009) and (Li et al., 2009) propose to combine the emotion speech priori based on the posterior emotion tendency of judging text based on contextual emotion tendency jointly in the emotion tendency of dictionary and the training text.The characteristic of subject matter of (Taboada et al., 2009) proposition combination text (describing comment, background, explanation etc.) and text itself is judged the emotion tendency of text jointly.(Tsutsumi et al., 2007) propose to utilize the multiple Classifiers Combination technology to come text emotion is classified.(Wan, 2008) and (Wan, 2009) propose to combine emotion abundant in the English to analyze the effect that resource improves Chinese emotion analysis.
Compare with the emotion analysis based on supervised learning, the rule-based and research unsupervised learning aspect is not a lot.Outside (Turney, 2002), (Zhu Yan haze et al., 2002) utilize HowNet that Chinese word language semanteme has been carried out the emotion tendency and calculate.(Lou De becomes et al.; 2006) utilize syntactic structure and the sub-semanteme of dependence centering sentence to carry out the emotion analysis; (Hiroshi et al.; 2004) realize the analysis of Japanese phrase level emotion through transforming a rule-based machine translator; (Zagibalov et al., 2008) in (Turney, 2002) thus the basis of SO-PMI algorithm on through for the in-depth analysis of Chinese text characteristic and introduce iteration mechanism and improved the accuracy rate that the unsupervised learning emotion is analyzed to a great extent.
Cross-cutting emotion analysis is an emerging field in the emotion analysis; Present research in this respect is not a lot; Main cause is that present research does not also have good solution how to seek a kind of mapping relations between two fields, how to seek the equilibrium relation between the characteristic weights between two fields in other words.Research for cross-cutting emotion analysis starts from (Blitzer et al.; 2007) cross-cutting emotion analysis is introduced in the corresponding study of structure; SCL is a kind of cross-domain texts analytical algorithm that is of wide application, and the purpose of SCL is that the characteristic on the training set is corresponded in the test set as far as possible.Has introduced SCL in the Chinese cross-cutting emotion analysis (Tan et al., 2009).(Tan2 et al., 2009) propose a kind of semi-supervised learning method of naive Bayesian and EM algorithm has been applied in the cross-cutting emotion analysis.Ordering (Graph Ranking) algorithm application will will be schemed in cross-cutting emotion analysis based on the thought of EM in (Wu et al., 2009), and the figure sort algorithm can be thought a kind of k-NN algorithm of iteration.Can find out that from present research cross-cutting emotion is analyzed subject matter and is to seek a kind of mapping relations between two fields, but such mapping relations or very difficult the searching perhaps need great mathematical justification.So the method that semi-supervised learning is used in a lot of researchs reduces the difference between training set and the test set gradually through successive iteration.
In the Chinese emotion analysis and research relevant to theme; More mostly current be to a certain specific area; Like automobile, hotel, media event etc.; Mostly be to specific field for the main method of this type research, set up relevant domain body and the dictionary of estimating commonly used thereof, through the analysis of sentence formula, predefine sentence masterplate, extract kernel sentence, judge the positive negativity of comment based on the methods such as machine learning of supervision.But these methods can not directly be used in the emotion analysis to microblogging; Because the microblogging content embraces a wide spectrum of ideas; The comment of delivering from the microblogging user to special entities such as products; Also have suggestion, treat, adopt diverse ways just can better carry out the emotion analysis so will distinguish to different entities to each side such as personage, incidents; In addition; For the method for existing dependence syntactic analysis aspect the emotion analysis relevant, itself bring except the syntactic analysis instrument to Chinese theme inaccurate; The extraction algorithm of its theme and qualifier haves much room for improvement; Simultaneously because of it combines semantic sentence formula information better, microblogging body lack of standard is very big in addition, and the pre-service that standardizes effectively all is the importance that improves the emotion accuracy of analysis to the microblogging content before analyzing.
In carry out the emotion analysis and research towards Chinese microblogging short text, have the scholar to adopt the emotion speech statistical method based on dictionary for the microblogging emotion analysis that theme has nothing to do, basic process is following: at first, a microblogging is carried out subordinate sentence by punctuate.Secondly, in a microblogging subordinate sentence, search the speech that is included in the weights dictionary, with their weights stack.Once more, in this microblogging visitor subordinate sentence, search the speech that is included in the negative dictionary, and statistics numbers, to confirm the positive or negative tone.At last, with the calculated value stack of each subordinate sentence, draw the mood value of a complete microblogging.The result that the microblogging mood weights counter emotion recognition of using C# language to write is tested judges that through intersecting accuracy reaches 80.6%.The advantage of method is that algorithm is simple; Efficient is higher; In that being carried out positive and negative differentiation, microblogging reached certain accuracy; But still have following problem: 1) result relies on its defined emotion dictionary too much, causes coverage rate wide inadequately, can't judge or only think for the sentence that does not appear at the emotion speech place in the emotion dictionary to be neutrality; 2), lean on emotion speech polarity to add merely and can't clear and definite bloger express what emotion to particular topic actually for the microblogging that a plurality of themes and a plurality of emotion speech occur; 3) only negative word being carried out even number is forward, and odd number is that the statistics of negative sense is easy to erroneous judgement, is that many times the bloger representes to negate mood to a plurality of entities in addition to the negating of emotion speech because can't confirm negative word; 4) do not consider degree adverb and sentence formula information, for some confirmative questions, comprise microblogging error in judgement such as turnover; 5) except that little for the using value the basic statistics bloger mood, what often the user more was concerned about is the mood attitude that is directed against concrete a certain entity in the microblogging, but not the general positive and negative judgement of whole piece microblogging.
Summary of the invention
The present invention is directed to the deficiency that prior art Chinese microblogging short text emotion is analyzed, proposed a kind of emotion analytical approach and system towards the microblogging short text.This method and system is to the integral body and the fine-grained microblogging emotional orientation analysis of particular topic and association attributes or part; Use is based on interdependent syntactic analysis; Method in conjunction with contents such as semantic information, domain bodies has improved analytical accuracy, helps the user to understand the emotional attitude of holding about special entity in the main flow microblogging effectively.Thereby the emotion situation through analyzing bloger's microblogging draws the mood index of bloger in a certain period.Comment content to a certain microblogging is carried out positive and negative emotional orientation analysis, and the user can be understood for specific blog article reviewer's the support or the comment and the ratio thereof of opposition viewpoint attitude.
A kind of emotion analytical approach towards the microblogging short text, wherein emotion analyze to as if the theme of entity, the method comprising the steps of: step 1, gather the microblogging data that comprise the designated key words and deposit database in; Step 2 reads the microblogging of special key words from database, filters out itself not comprise the configuration key word is expressed an opinion or the microblogging of message, and the microblogging data through filtration treatment are carried out denoising, removes the data lack of standardization in the microblogging; Step 3 loads relevant dictionary, according to user configured key word classification, loads outer, the corresponding field of general positive negative affect dictionary positive and negative evaluation dictionary commonly used, negates dictionary, degree dictionary, sentence formula dictionary; Step 4 is carried out subordinate sentence, filters out not comprise the sentence that the user disposes key word; Step 5 is carried out participle to the sentence that comprises key word, and part-of-speech tagging extracts adjective, noun, verb, adverbial word in the sentence, and uses corresponding field dictionary to search for, as appears at and then carry out mark in the dictionary; Speech for remaining matees in general emotion dictionary, and is same for appearing at the speech mark in the emotion vocabulary and adding the emotion set of words, if the emotion word set is combined into sky; Think that then this sentence does not have obvious emotion tendency; Be defaulted as neutrality, carry out next processing, otherwise carry out next step; Step 6 utilizes the syntactic analysis instrument that the sentence that comprises theme is carried out interdependent syntactic analysis; Step 7 is judged the polarity of each sentence of comprising descriptor; Step 8; After having judged the polarity of the sentence that all comprise descriptor; Front sentence polarity sum is designated as PositiveSum and negative sentence polarity sum is NegativeSum in the result of calculation set, according to not counting the emotion tendency that PosSenNum value calculating whole piece microblogging counted in NegSenNum and forward sentence in the sentence result set to sentence:
The present invention also provides a kind of emotion analytical approach towards the microblogging short text, wherein emotion analyze to as if bloger's mood index, then the method comprising the steps of: step 1, bloger's microblogging is carried out pre-service; Step 2, relevant dictionary loads, and comprises general positive negative affect dictionary, negates dictionary, degree dictionary, general positive and negative emoticon dictionary; Step 3, according to this microblogging whether be purely share, the emotion tendency that microblogging confirmed in emoticon, emotion speech, negative word, degree speech in the pure forwarding, microblogging; Step 4 is filed according to the date all microbloggings, and the emotion tendency according to all microbloggings of issuing on the same day draws bloger's microblogging mood index of this day.
The present invention also provides a kind of emotion analytical approach towards the microblogging short text, and what wherein emotion was analyzed comments on tendentiousness to liking microblogging, and the method comprising the steps of: step 1, to microblogging comment carrying out pre-service; Step 2, relevant dictionary loads, and comprises general positive negative affect dictionary, negates dictionary, degree dictionary, general positive and negative emoticon dictionary; Step 3, the emoticon in the statistics comment is saved in respectively among GoodEmotions and the BadEmotions according to general positive and negative emoticon dictionary; Step 4, participle is carried out in comment to the whole piece microblogging, and part-of-speech tagging carries out emotion dictionary coupling to adjective, noun, adverbial word, verb, and the positive negative affect speech of appearance is saved in respectively among PositiveWords and the NegativeWords; Step 5 if GoodEmotions, BadEmotions, PositiveWords and NegativeWords are sky, thinks that then this comment is neutral comment, establishes its emotion tendency CommentOrientation=0; Step 6, the search negative word, as comprise negative word, check that then whether it modifies a certain emotion speech, then gets negative to emotion speech polarity in this way; Step 7, search degree speech, as comprise the degree speech, and check then whether it modifies a certain emotion speech, be that the current polarity of emotion speech multiply by degree speech intensive parameter then in this way to adjustment emotion speech polarity; Step 8 is calculated the positive negative sense result of comment, and computing formula is following: forward is the polarity of all speech among the expression number+PositiveWords among the PositiveSum=GoodEmotions as a result; Negative sense is the polarity of all speech among the expression number+NegativeWords among the NegativeSum=BadEmotions as a result; Step 9, the comment emotion tendency:
Step 10, according to the interjection of mark, if in the comment interjection is arranged, then end value multiply by certain parameter as final emotion score value: CommentOrientation=1.5*CommentOrientation;
Use emotion analytical approach and the system towards the microblogging short text of the present invention; Select microblogging source (several big main flow microblogging system) according to user configured key word; At first carry out the collection of the relevant microblogging of key word, key word can be a certain personage, product, service, mechanism, incident etc.Can dispose the microblogging content of gathering specific personage and mechanism's issue in addition, the comment content of gathering relevant microblogging.
In the emotion analytical applications; The user can carry out the microblogging emotional orientation analysis to the microblogging that comprises nominal key, designated person and the microblogging of mechanism's issue, the related commentary of certain bar microblogging that are disposed, and the analytical approach of taking to different contents is different.
When carrying out emotional orientation analysis to the microblogging of specifying particular topic; The user can carry out whole positive and negative trend analysis to relevant microblogging; Can also more fine-grained configuration key word; Carry out positive and negative trend analysis like the attribute of the personage in the incident, product, personage's particular aspects etc., make the emotion analysis have more specific aim, make the user can understand the emotional attitude of in the microblogging its aspect of being concerned about being held.
Analysis result is carried out omnibearing visual show, comprise that positive and negative neutral microblogging is by the displaying of emotion degree, positive and negative neutral microblogging scale map, positive and negative neutral microblogging trend graph etc.
Description of drawings
Fig. 1 is the application system schematic diagram of the present invention towards the emotion analytical approach of microblogging short text;
Fig. 2 is the simple use process flow diagram of the present invention towards the emotion analytical approach of microblogging short text;
Fig. 3 is the frame diagram of the present invention towards the application system of the emotion analytical approach of microblogging short text;
Fig. 4 user's configuration flow figure that is the present invention in the emotion analytical approach of microblogging short text.
Fig. 5 is that the present invention analyzes the detailed algorithm process flow diagram to the relevant emotion of theme in the emotion analytical approach of microblogging short text;
Fig. 6 is that the present invention analyzes the detailed algorithm process flow diagram to bloger's mood in the emotion analytical approach of microblogging short text;
Fig. 7 is the present invention in the emotion analytical approach of microblogging short text, and the detailed algorithm process flow diagram is analysed in the microblogging scoring;
Fig. 8 utilizes the present invention towards the emotion analytical approach of microblogging short text a certain personage's microblogging emotion analysis result to be showed (emotion ratio and tendency) sectional drawing;
Fig. 9 utilizes the present invention to be directed against a certain personage's microblogging emotion analysis result forward microblogging sectional drawing towards the emotion analytical approach of microblogging short text;
Figure 10 utilizes the present invention to show sectional drawing towards the emotion analytical approach of microblogging short text to a certain bloger's mood analysis result;
Figure 11 utilizes the present invention to show sectional drawing towards the emotion analytical approach of microblogging short text to a certain microblogging comment emotion analysis result.
Embodiment
For making the object of the invention, technical scheme and advantage clearer, below in conjunction with specific embodiment, and with reference to accompanying drawing, to further explain of the present invention.
Emotion analytical approach towards the microblogging short text of the present invention is primarily aimed at three types microblogging short essay and carries out the emotion analysis.
First kind is to carry out integral body and fine-grained microblogging emotional orientation analysis to user's designated key; Use is based on interdependent syntactic analysis; Method in conjunction with contents such as semantic information, domain bodies has improved analytical accuracy, helps the user to understand the emotional attitude of holding about special entity in the main flow microblogging effectively.
Thereby second kind is to draw the mood index of bloger in a certain period through the emotion situation of analyzing bloger's microblogging.
The third is to carry out positive and negative emotional orientation analysis to the comment content of a certain microblogging, and the user can be understood for specific blog article reviewer's the support or the comment and the ratio thereof of opposition viewpoint attitude.
For first kind of situation, to the integral body and the fine-grained microblogging emotional orientation analysis of particular topic and association attributes or part, the emotion analytical approach towards the microblogging short text that the present invention proposes mainly comprises the steps:
Step 1 is at first carried out the collection analysis relevant configuration, and configuration item comprises topic title, the affiliated classification of topic, topic key word, microblogging website and gathers content.Configuration flow is following:
A) user imports the title of a certain topic;
B) select the affiliated field classification of this topic;
C) import crucial words relevant under this topic;
D) select the Source Site of coming of microblogging data, can multiselect;
E) content type of microblogging is gathered in selection, comprises the microblogging text, picture etc.
Step 2, data acquisition step, the microblogging data that comprise the designated key words through the microblogging data collecting module collected deposit database in;
Step 3; The data pre-treatment step; At first according to user configured topic keyword speech; Read the microblogging that comprises key word from database, standardize through the microblogging data preprocessing module then and filter pre-service, mainly comprise two parts: the one, filter out itself not comprise the configuration key word is expressed an opinion or microbloggings such as the answer of message, forwarding; The 2nd, to carrying out denoising through the microblogging data of filtration treatment, remove the data lack of standardization in the microblogging through a last step, comprise that unnecessary punctuation mark, link etc. are useless or cause the information of interference to syntactic analysis.
This data pre-treatment step specifically may further comprise the steps: filter out non-original microblogging, promptly transmit, reply others etc. microblogging; Filter out contents such as sharing picture video merely and do not have the microblogging of comment, characteristic is that beginning of the sentence is " sharing ", " uploading pictures " " uploaded videos " etc.; Filter out beginning of the sentence sentence tail " # " and middle content thereof, right for " # " in the sentence, only remove symbol, keep content; Filter out the content in beginning of the sentence " [] " and the bracket thereof; Filter out bloger's name of beginning of the sentence sentence tail " " symbol and its back, filter out " " symbol in the sentence; "～" changes fullstop into; Replacing with Chinese character to "+" "-" "=" in the sentence " adds " " subtracting " and " equals "; Remove unnecessary punctuation mark, then only keep one like a plurality of fullstops or comma; Remove all links in the microblogging; Remove that the sentence tail " sees for details ", " play-by-play ", " little interview " wait the sentence that belongs to.
Step 4 loads relevant dictionary, according to user configured key word classification, except loading general positive negative affect dictionary, loads corresponding field positive and negative evaluation dictionary commonly used.Load negates dictionary, degree dictionary, sentence formula dictionary.Set up following emotion dictionary:
General positive negative affect dictionary: the word collection is used in the Chinese emotion analysis based on Hownet provides; It provides positive emotion word, negative emotion word, positive word, the negative evaluation word estimated; Filter and adjustment through artificial; Obtain positive emotion and estimate 3743 of speech, negative emotion is estimated 3737 of speech.
The field dictionary of estimating commonly used: because there is different emotion dictionaries in different fields; The foundation of field emotion dictionary needs a large amount of resources; System only comprises hotel's speech of estimating commonly used at present, progressively sets up association area structural system in the future, improves the corresponding dictionary of estimating.
The negative word of negative dictionary: this paper extracts and comprises the former notion of negative justice in the knowledge net, and the manual work filtration obtains 18 negative words.Specifically be respectively: not, not, nothing, non-, not, not, not, not, not, do not have, do not have, lose, exempt from, lack, prohibit, avoid, guard against, anti-.These negative words not only comprise the adopted former definition to basic negative word, also include the adopted former of expansion back negative word.No matter be with the expansion of arranging in pairs or groups of basic negative word, negate that the former vocabulary of justice carries out other collocation of degree level still to include, these characteristic key words with negative meaning have all been carried out effective processing to sentence.
Degree dictionary: the degree rank word lists that the Chinese emotion analysis that provides based on Hownet is concentrated with word; It comprises totally 219 of other degree speech of 6 degree levels; Filter and adjustment through artificial, keep 4 original grade classifications, reduced uncommon words; Only keep 114 of the most frequently used degree speech, degree speech rank and self-defined intensity thereof are following:
Sentence formula dictionary: in complex sentence, have some conjunctions or adverbial word sometimes subordinate sentence linked together, and they also contained must logical organization, therefore, the conjunction that occurs in multiple ten days is called conjunctive word.In ten days formula structure; Different conjunctive words can make that also the semantic tendency of sentence changes, therefore, and according to the requirement of emotion analysis; We will carry out concrete emotion value valuation analysis to progressive relationship, coordination and turnover relation, promptly following five groups of conjunctive words carried out quantitatively.1. arranged side by side: with, also, simultaneously ... Simultaneously, on one side ... On one side, again ... Again, both ... Again; 2. go forward one by one: and even, more, not only ... Also, not only ... And, not only ... Also; The turnover: yet but, still, and, but; The hypothesis: if if if if, suppose; 5. condition: only if no matter need only, have only, no matter.For preceding two groups, all be the overlapping of emotion side by side with progressive relationship, before and after from emotion tendency intensity, can being expressed as the polarity of subordinate sentence and; And for the turnover conjunctive word; The transfer or the transformation of the emotion often that it is expressed; Therefore, the processing to adversative is exactly to obtain the polarity number of whole complex sentence again according to the emotion tendency computing formula of follow-up sentence to the opposite processing of the do of the emotion propensity value in the subordinate sentence of adversative; And the conjunctive word notion of hypothesis and condition is under the prerequisite that first wife's sentence situation satisfies, and the emotion value research in the follow-up sentence is just meaningful.Therefore, when these two types of conjunctive words occurring, the emotional expression of this complex sentence is no practical significance, and promptly polarity is 0.
Step 5 is carried out subordinate sentence, filters out not comprise the sentence that the user disposes key word.
Step 6 is carried out participle to the sentence that comprises key word, and part-of-speech tagging extracts adjective, noun, verb, adverbial word in the sentence, and uses corresponding field dictionary to search for, as appears at and then carry out mark in the dictionary; Speech for remaining matees in general emotion dictionary, and is same for appearing at the speech mark in the emotion vocabulary and adding the emotion set of words.If the emotion word set is combined into sky, think that then this sentence does not have obvious emotion tendency, be defaulted as neutrality, carry out next processing, otherwise carry out next step.
Step 7 utilizes the syntactic analysis instrument that the sentence that comprises theme is carried out interdependent syntactic analysis.
Step 8 is searched for negative word in the sentence, degree speech, sentence formula speech and VOB structure and is write down the relevant position and syntactic information.
Step 9, mark topic keyword position and syntactic information thereof add pending subject information tabulation.Step 10 is taken out a descriptor from the subject information set.
Step 11 is taken out from the emotion word set and is waited to judge the emotion speech, and it is right to begin to travel through successively its grammatical relation from this emotion speech, if in traversal, find this descriptor, thinks that then this emotion speech modifies this descriptor, and this emotion speech of mark is for use, and matched indicia is true; Then do not carry out next step as having.
Step 13 is carried out the predicate part of speech and is judged, if predicate is a verb, carries out next step; As otherwise return step 10.
Step 15 is searched the VOB of coupling descriptor SBV structure in the VOB structure, if object be emotion speech then matched indicia for true, this emotion speech and VOB are labeled as and use; Like object is not the emotion speech, and then inquiry closes on the ADV structure and sees whether has before the object emotion speech to modify, if any same this emotion speech of mark and VOB for using; As not having, then return step 10.
Step 16 if matched indicia is true, is then carried out negative match, and define negative word and modify this emotion speech,
Then the dynamic polarity of this emotion speech is got negative; Degree speech coupling defines the degree speech and modifies this emotion speech, and then the dynamic polarity of this emotion speech equals the intensive parameter that existing polarity multiply by the degree speech.
Step 17 deposits this descriptor and context polarity in the interim result set in, carries out next descriptor and handles, and turns back to for the 9th step.
Step 18, this sentence disposes, and calculates this polarity, calculates this polarity according to not face polarity logarithm NegativeNum and positive polarity logarithm PositiveNum in the interim result set according to following formula.
Step 19, handle all sentences that comprise descriptor according to above-mentioned steps after, front sentence polarity sum is designated as PositiveSum and negative sentence polarity sum is NegativeSum in the result of calculation set.According to not counting the emotion tendency that PosSenNum value calculating whole piece microblogging counted in NegSenNum and forward sentence in the sentence result set to sentence;
Step 20, next bar microblogging is analyzed, and handles all microblogging saving result collection to database and return to the user and check.
The treatment step that above-described emotion analytical approach towards the microblogging short text of the present invention is adopted when being directed against first kind of situation; Just to particular topic and association attributes or integral body and fine-grained microblogging emotional orientation analysis partly; Promptly, effective during like special entities such as the concrete personage in personage, product and service, mechanism or the incident, mechanisms to the key word of a certain user's appointment.
Incident itself to be carried out whole emotion different with the treatment step of top description in analyzing but handling; If mainly be because the blog article publisher carries out open comment to the personage who relates in the incident, mechanism etc. or to result that incident caused etc. often; If event name be used as can cause when entity is stated method in the use analyzing inaccurate; So above-described process is not supported not carry out whole emotional orientation analysis to incident itself, just can adopt said process analysis when only giving in the outgoing event a certain concrete entity such as specific personage, tissue to the user.
Thereby the second kind of situation that is directed against towards microblogging short text emotion analytical approach of the present invention is to draw the mood index of bloger in a certain period through the emotion situation of analyzing bloger's microblogging; The third situation is to carry out positive and negative emotional orientation analysis to the comment content of a certain microblogging, and the user can be understood for specific blog article reviewer's the support or the comment and the ratio thereof of opposition viewpoint attitude.The processing procedure that is adopted to second kind of situation and the third situation and above-mentioned first kind of situation exist some different be because: 1) microblogging of certain bloger's issue has very big randomness; Content comprises various aspects; There is not regularity; Even configuration custom entities key word, but the microblogging number that comprises this key word is very little, and analysis result is too big practical value not; 2) if the entity and the corresponding emotion speech practicality thereof that adopt the Automatic Extraction bloger of system microblogging to be comprised are also bad, and the technical difficulty increasing, the entity speech of extraction also need pass through artificial the filtration; 3) microblogging integral body is carried out general positive and negative evaluation; Final analysis result can be found out this bloger's mood situation; The microblogging of issuing every day according to the bloger has how much comprise positive mood, has how much to have to comprise negative emotions, and then draws bloger's microblogging mood index of some day.
In addition, on content-length, a microblogging comprises one or several sentence usually, and comment is general relatively more brief, and most of comment only comprises in short usually; When handling the microblogging emotion analysis of specifying topic, do not consider emoticon; Be because be not sure of the specific aim of emoticon; And because comment all is to deliver viewpoint to specific blog article, the ratio that emoticon occurs is bigger, so when analyzing the microblogging comment, consider emoticon; Microblogging comment content is often omitted descriptor, and sentence element is imperfect, so the syntactic analysis instrument is inapplicable.
Based on above consideration, mainly adopt same algorithm when the microblogging of a certain bloger issue is carried out the mood Index for Calculation and positive and negative emotional orientation analysis is carried out in comment to microblogging, but the method that when data are carried out pre-service, adopts is different.Thereby the second kind of situation that is directed against towards microblogging short text emotion analytical approach of the present invention is to draw the bloger at the mood index of a certain period through the emotion situation of analyzing bloger's microblogging, may further comprise the steps:
Step 1, the pre-service of bloger's microblogging, detailed step is following:
A) filter out beginning of the sentence sentence tail " # " and middle content thereof, right for " # " in the sentence, only remove symbol, keep content;
B) filter out the interior content of beginning of the sentence " [] " and bracket thereof;
C) filter out bloger's name of beginning of the sentence sentence tail " " symbol and its back, filter out " " symbol in the sentence;
D) "～" changes fullstop into;
E) remove unnecessary punctuation mark, then only keep one like a plurality of fullstops or comma;
F) remove all links in the microblogging;
G) remove that the sentence tail " sees for details ", " play-by-play ", " little interview " wait the sentence that belongs to.
Step 2, relevant dictionary loads, and comprises general positive negative affect dictionary, negates dictionary, degree dictionary, general positive and negative emoticon dictionary.Wherein negative affect dictionary, negative dictionary, degree dictionary are same resource with the topic designated entities, see above-mentioned illustrate dictionary.General positive and negative emoticon dictionary is to set up according to the corresponding meaning of the expression of expression mood commonly used in the main flow microbloggings such as Sina, Tengxun, Netease, Sohu.Wherein comprise 39 of forward emoticons commonly used, 33 of negative sense emoticons.See the following form in detail.
Step 3 judges whether this microblogging is pure sharing, and promptly the bloger has shared a width of cloth picture, a video etc., thinks that then this microblogging emotion is positive, establishes its emotion tendency SentimentOrientation=1, carries out next bar microblogging analysis.
Step 4 judges whether this microblogging is pure forwarding, if transmit microblogging, thinks that then this forwarding has reflected the mood that the bloger is same, the microblogging content of its forwarding is returned the first step carry out the emotion analysis.Set the tendentiousness of this forwarding according to transmitting the content emotion tendency.
Step 5, the emoticon in the statistics microblogging is saved in respectively among GoodEmotions and the BadEmotions according to general positive and negative emoticon dictionary.
Step 6, participle is carried out in comment to the whole piece microblogging, and part-of-speech tagging carries out emotion dictionary coupling to adjective, noun, adverbial word, verb, and the positive negative affect speech of appearance is saved in respectively among PositiveWords and the NegativeWords.
Step 7, if GoodEmotions, BadEmotions, PositiveWords and NegativeWords are sky, but this microblogging is the property a shared microblogging,
Step 8, search negative word NegWord, as comprise negative word, and judge then whether it modifies a certain emotion speech, then get negative in this way to this emotion speech polarity.
Step 9, speech IntensifyWord is stressed in search, as comprises the degree speech, judges then whether it modifies a certain emotion speech, is that the current polarity of emotion speech multiply by degree speech intensive parameter Strength (IntensifyWord) to adjustment emotion speech polarity then in this way.
Step 10 is calculated the positive negative sense result of comment, and computing formula is following:
Forward is the polarity of all speech among the expression number+PositiveWords among the PositiveSum=GoodEmotions as a result;
Negative sense is the polarity of all speech among the expression number+NegativeWords among the NegativeSum=BadEmotions as a result;
Step 11, microblogging emotion tendency SentimentOrientation
SentimentOrientation＝PositiveSum+Negative；
SentimentOrientation＝1.5*SentimentOrientation；
Step 13 is filed according to the date all microbloggings, the microblogging of issuing is on the same day carried out the microblogging mood index BloggerMoodIndex (day) that results added draws this day of bloger (day), that is:
BloggerMoodIndex(day)＝Sum(SentimentOrientation)；
The third situation that is directed against towards the emotion analytical approach of microblogging short text of the present invention is to carry out positive and negative emotional orientation analysis to the comment content of a certain microblogging; The user can be understood for specific blog article reviewer's the support or the comment and the ratio thereof of opposition viewpoint attitude, may further comprise the steps:
Step 1, microblogging comment pre-service, detailed step is following:
A) remove " transmit this microblogging: "
B) filter out the answer of bloger to the reviewer;
C) filter out the answer of reviewer to other people comment;
D) filter out link;
E) remove unnecessary punctuation mark, then only keep one like a plurality of fullstops or comma;
F) if a plurality of sentences are arranged, no punctuate then adds comma between sentence, between sentence ".", " ... " Change ", " into etc. the punctuate symbol unification outside the non-exclamation mark, exclamation carries out changing ", " equally into behind the mark.
Step 2, relevant dictionary loads, and comprises general positive negative affect dictionary, negates dictionary, degree dictionary, general positive and negative emoticon dictionary.
Step 3, the emoticon in the statistics comment is saved in respectively among GoodEmotions and the BadEmotions according to general positive and negative emoticon dictionary.
Step 4, participle is carried out in comment to the whole piece microblogging, and part-of-speech tagging carries out emotion dictionary coupling to adjective, noun, adverbial word, verb, and the positive negative affect speech of appearance is saved in respectively among PositiveWords and the NegativeWords.
Step 5 if GoodEmotions, BadEmotions, PositiveWords and NegativeWords are sky, thinks that then this comment is neutral comment, establishes its emotion tendency CommentOrientation=0.
Step 6, the search negative word, as comprise negative word, check that then whether it modifies a certain emotion speech, then gets negative to emotion speech polarity in this way.
Step 7, search degree speech, as comprise the degree speech, and check then whether it modifies a certain emotion speech, be that the current polarity of emotion speech multiply by degree speech intensive parameter then in this way to adjustment emotion speech polarity.
Step 8 is calculated the positive negative sense result of comment, and computing formula is following:
Forward is the polarity of all speech among the expression number+PositiveWords among the PositiveSum=GoodEmotions as a result;
Negative sense is the polarity of all speech among the expression number+NegativeWords among the NegativeSum=BadEmotions as a result;
Step 9, comment emotion tendency CommentOrientation
Step 10, according to the interjection of mark, if in the comment interjection is arranged, then end value multiply by certain final emotion score value of parameter conduct:
CommentOrientation＝1.5*CommentOrientation；
Further specify the emotion analytical approach towards the microblogging short text of the present invention below through specifically giving an example.
In example shown in Figure 8, be that a certain topic microblogging is carried out emotional orientation analysis, appointment theme as a certain personage.Analysis result is shown in accompanying drawing:
Collect 165 Sina's microbloggings altogether according to setting capture program; Carry out the data pre-service then; Also be left 128 microbloggings after filtering out invalid microbloggings such as advertisement, forwarding, carry out the relevant emotion analysis to this entity of theme, the analysis result that obtains is added up as follows:
The microblogging number of holding positive emotion to this entity has 37;
The microblogging number of holding negative emotion to this entity has 18;
The neutral microblogging number that does not have obvious emotion tendency has 73;
Through artificial checking, add up with recall rate to the emotional orientation analysis accuracy of special entity and to see the following form:
In another example shown in Figure 10; Be that the microblogging that a certain bloger issues is carried out emotional orientation analysis; The result is as shown in the figure: according to setting capture program acquisition time section is nearly one month; The microblogging of bloger's issue on Dec 25th, 25,2012 1 promptly on November collects 454 effective microbloggings altogether.Every microblogging is carried out the no theme emotional orientation analysis based on emotion speech and emoticon, and the mood of adding up the bloger according to analysis result is showed according to the page and can be found out mood score that the bloger is nearest 2 days and nearest 10 days mood tendency, sees accompanying drawing.Analysis result is added up as follows:
Actively the front microblogging number of phychology has: 334;
The negative microblogging number of passive phychology has: 47;
The neutral microblogging number that does not have obvious emotion tendency has: 73;
Through artificial checking, mood mining analysis accuracy and recall rate as a result sees the following form:
In another example shown in Figure 11, be that emotional orientation analysis is carried out in the comment of a certain microblogging, analysis result is shown in accompanying drawing: collect 297 comments altogether according to setting capture program.Through the comment data pre-service, obtain effectively commenting on 265 after filtering out the rubbish comment and replying other people comment to the microblogging content.Comment on positive and negative emotional orientation analysis, the analysis result that obtains is added up as follows:
Hold the comment number of supporting attitude and have 132;
Hold the comment number of opposing attitude and have 14;
The comment number that sits on the fence or do not have an obvious emotion tendency has 119;
Through artificial checking, microblogging comment and analysis accuracy is as a result added up like following table with recall rate:
Above-described specific embodiment; The object of the invention, technical scheme and beneficial effect have been carried out further explain, it should be understood that the above is merely specific embodiment of the present invention; Be not limited to the present invention; All within spirit of the present invention and principle, any modification of being made, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. emotion analytical approach towards the microblogging short text, wherein emotion analyze to as if the theme of entity, the method comprising the steps of:
Step 1 is gathered the microblogging data that comprise the designated key words and is deposited database in;
Step 2 reads the microblogging of special key words from database, filters out itself not comprise the configuration key word is expressed an opinion or the microblogging of message, and the microblogging data through filtration treatment are carried out denoising, removes the data lack of standardization in the microblogging;
Step 3 loads relevant dictionary, loads general positive negative affect dictionary, negates dictionary, degree dictionary, sentence formula dictionary, and according to field under the user configured key word, load corresponding field positive and negative evaluation dictionary commonly used;
Step 4 is carried out subordinate sentence, filters out not comprise the sentence that the user disposes key word;
Step 5 is carried out participle to the sentence that comprises key word, and part-of-speech tagging extracts adjective, noun, verb, adverbial word in the sentence, and uses corresponding field dictionary to search for, as appears at and then carry out mark in the dictionary; Speech for remaining matees in general emotion dictionary, and is same for appearing at the speech mark in the emotion vocabulary and adding the emotion set of words, if the emotion word set is combined into sky; Think that then this sentence does not have obvious emotion tendency; Be defaulted as neutrality, carry out next processing, otherwise carry out next step;
Step 6 utilizes the syntactic analysis instrument that the sentence that comprises theme is carried out interdependent syntactic analysis;
Step 7 is judged the polarity of each sentence of comprising descriptor;
Step 8; After having judged the polarity of the sentence that all comprise descriptor; Front sentence polarity sum is designated as PositiveSum and negative sentence polarity sum is NegativeSum in the result of calculation set, according to not counting the emotion tendency that PosSenNum value calculating whole piece microblogging counted in NegSenNum and forward sentence in the sentence result set to sentence:
2. method according to claim 1 is characterized in that, after step 6, also comprises step:
Step a searches for negative word in the sentence, degree speech, sentence formula speech and VOB structure and writes down the relevant position and syntactic information;
Step b, mark topic keyword position and syntactic information thereof add pending subject information tabulation;
Step c takes out a descriptor from the subject information set;
Steps d is taken out from the emotion word set and is waited to judge the emotion speech, and it is right to begin to travel through successively its grammatical relation from this emotion speech, if in traversal, find this descriptor, thinks that then this emotion speech modifies this descriptor, and this emotion speech of mark is for use, and matched indicia is true; Then do not carry out next step as having;
Step e judges whether the dependence of this descriptor is " SBV ", judges the part of speech of predicate in this way, as otherwise judge whether the sentence structure of this descriptor is " DE " structure, in this way then mark " " after speech as interim descriptor, carry out next step;
Step f carries out the predicate part of speech and judges, if predicate is a verb, carries out next step; As otherwise return step c;
Step g if this verb is the emotion speech, is then returned step c, otherwise is carried out next step;
Step h searches the VOB of coupling descriptor SBV structure in the VOB structure, if object be emotion speech then matched indicia for true, this emotion speech and VOB are labeled as and use; Like object is not the emotion speech, and then inquiry closes on the ADV structure and sees whether has before the object emotion speech to modify, if any same this emotion speech of mark and VOB for using; As not having, then return step c;
Step I if matched indicia is true, is then carried out negative match, defines negative word and modifies this emotion speech, and then the dynamic polarity of this emotion speech is got negative; Degree speech coupling defines the degree speech and modifies this emotion speech, and then the dynamic polarity of this emotion speech equals the intensive parameter that existing polarity multiply by the degree speech;
Step j deposits this descriptor and context polarity in the interim result set in, carries out next descriptor and handles, and turns back to step b;
Step k, this sentence disposes, and calculates this polarity, calculates this polarity according to not face polarity logarithm NegativeNum and positive polarity logarithm PositiveNum in the interim result set through following formula:
。
3. method according to claim 1 and 2 is characterized in that, the data pre-treatment step further comprises: filter out non-original microblogging; Filter out the microblogging of sharing picture or video content merely and not having comment; Filter out beginning of the sentence sentence tail and be " # " and middle content thereof; Filter out beginning of the sentence and be the content in " [] " and the bracket thereof; Filter out the bloger name of beginning of the sentence sentence tail for " " symbol and its back; "～" changed into fullstop; Replacing with Chinese character to "+" "-" "=" in the sentence " adds " " subtracting " and " equals "; Remove unnecessary punctuation mark; Remove all links in the microblogging; Remove the sentence that " seeing for details ", " play-by-play ", " little interview " place appear in the sentence tail.
4. method according to claim 3 is characterized in that, the word collection is used in the Chinese emotion analysis that general positive negative affect dictionary is based on Hownet to be provided, and it provides positive emotion word, negative emotion word, positive word, the negative evaluation word estimated.
5. emotion analytical approach towards the microblogging short text, wherein emotion analyze to as if bloger's mood index, then the method comprising the steps of:
Step 1 is carried out pre-service to bloger's microblogging;
Step 2, relevant dictionary loads, and comprises general positive negative affect dictionary, negates dictionary, degree dictionary, general positive and negative emoticon dictionary;
Step 3, according to this microblogging whether be purely share, the emotion tendency that microblogging confirmed in emoticon, emotion speech, negative word, degree speech in the pure forwarding, microblogging;
Step 4 is filed according to the date all microbloggings, and the emotion tendency according to all microbloggings of issuing on the same day draws bloger's microblogging mood index of this day.
6. method according to claim 5 is characterized in that step 3 further comprises:
Step 301 judges whether this microblogging is pure sharing, if, think that then this microblogging emotion is positive, establish its emotion tendency SentimentOrientation=1, carry out next bar microblogging analysis;
Step 302 judges whether this microblogging is pure forwarding, if transmit microblogging, the microblogging content of its forwarding is returned step 1 carry out the emotion analysis, sets the tendentiousness of this forwarding according to transmitting the content emotion tendency;
Step 303, the emoticon in the statistics microblogging is saved in respectively among forward expression collection GoodEmotions and the negative sense expression collection BadEmotions according to general positive and negative emoticon dictionary;
Step 304 is carried out participle to the whole piece microblogging, and part-of-speech tagging carries out emotion dictionary coupling to adjective, noun, adverbial word, verb, and the positive negative affect speech of appearance is saved in respectively among PositiveWords and the NegativeWords;
Step 305 if GoodEmotions, BadEmotions, PositiveWords and NegativeWords are sky, thinks that then this comment microblogging is neutral, establishes its emotion tendency SentimentOrientation=0;
Step 306, search negative word NegWord, as comprise negative word, judge then whether it modifies a certain emotion speech, in this way then to this emotion speech polarity negate;
Step 307, search degree speech IntensifyWord, as comprise the degree speech, and judge then whether it modifies a certain emotion speech, be that the current polarity number of emotion speech multiply by degree speech intensive parameter Degree (IntensifyWord) then in this way to adjustment emotion speech polarity;
Step 308 is calculated the positive negative sense result of this microblogging, and computing formula is following:
Forward is the polarity of all speech among the expression number+PositiveWords among the PositiveSum=GoodEmotions as a result;
Negative sense is the polarity of all speech among the expression number+NegativeWords among the NegativeSum=BadEmotions as a result;
Step 309, microblogging emotion tendency SentimentOrientation=PositiveSum+Negative.
7. method according to claim 6 is characterized in that step 4 further comprises:
Step 401 is according to the interjection of mark, if interjection is arranged in the microblogging, then
SentimentOrientation＝1.5*SentimentOrientation；
Step 402 is filed according to the date all microbloggings, the microblogging of issuing is on the same day carried out results added draw the bloger microblogging mood index BloggerMoodIndex (day) of this day, that is:
BloggerMoodIndex(day)＝Sum(SentimentOrientation)。
8. method according to claim 5 is characterized in that, bloger's microblogging is carried out pre-service further comprise:
Step 101 filters out beginning of the sentence sentence tail " # " and middle content thereof, and is right for " # " in the sentence, only removes symbol, keeps content;
Step 102 filters out the content in beginning of the sentence " [] " and the bracket thereof;
Step 103 filters out bloger's name of beginning of the sentence sentence tail " " symbol and its back, filters out " " symbol in the sentence;
Step 104, "～" changes fullstop into;
Step 105 is removed unnecessary punctuation mark;
Step 106 is removed all links in the microblogging;
Step 107 removes that the sentence tail " sees for details ", the sentence at " play-by-play ", " little interview " place.
9. emotion analytical approach towards the microblogging short text, wherein emotion analyze to as if microblogging comment tendentiousness, the method comprising the steps of:
Step 1 is to microblogging comment carrying out pre-service;
Step 2, relevant dictionary loads, and comprises general positive negative affect dictionary, negates dictionary, degree dictionary, general positive and negative emoticon dictionary;
Step 3, the emoticon in the statistics comment is saved in respectively among GoodEmotions and the BadEmotions according to general positive and negative emoticon dictionary;
Step 4, participle is carried out in comment to the whole piece microblogging, and part-of-speech tagging carries out emotion dictionary coupling to adjective, noun, adverbial word, verb, and the positive negative affect speech of appearance is saved in respectively among PositiveWords and the NegativeWords;
Step 5 if GoodEmotions, BadEmotions, PositiveWords and NegativeWords are sky, thinks that then this comment is neutral comment, establishes its emotion tendency CommentOrientation=0;
Step 6, the search negative word, as comprise negative word, check that then whether it modifies a certain emotion speech, then gets negative to emotion speech polarity in this way;
Step 7, search degree speech, as comprise the degree speech, and check then whether it modifies a certain emotion speech, be that the current polarity of emotion speech multiply by degree speech intensive parameter then in this way to adjustment emotion speech polarity;
Step 8 is calculated the positive negative sense result of comment, and computing formula is following:
Forward is the polarity of all speech among the expression number+PositiveWords among the PositiveSum=GoodEmotions as a result;
Negative sense is the polarity of all speech among the expression number+NegativeWords among the NegativeSum=BadEmotions as a result;
Step 9, the comment emotion tendency:
Step 10, according to the interjection of mark, if in the comment interjection is arranged, then end value multiply by certain final emotion score value of parameter conduct:
CommentOrientation＝1.5*CommentOrientation。
10. method according to claim 9 is characterized in that step 1 further comprises:
Step 101 is removed " transmit this microblogging: ";
Step 101 filters out the answer of bloger to the reviewer;
Step 101 filters out the answer of reviewer to other people comment;
Step 101 filters out link;
Step 101 is removed unnecessary punctuation mark;
Step 101, if a plurality of sentences are arranged, no punctuate then adds comma between sentence, and the punctuate symbol unification outside the non-exclamation mark between sentence changes ", " into, and exclamation carries out changing ", " equally into behind the mark.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210088366XA CN102663046A (en) | 2012-03-29 | 2012-03-29 | Sentiment analysis method oriented to micro-blog short text |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210088366XA CN102663046A (en) | 2012-03-29 | 2012-03-29 | Sentiment analysis method oriented to micro-blog short text |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102663046A true CN102663046A (en) | 2012-09-12 |
Family
ID=46772537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210088366XA Pending CN102663046A (en) | 2012-03-29 | 2012-03-29 | Sentiment analysis method oriented to micro-blog short text |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102663046A (en) |
Cited By (100)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930042A (en) * | 2012-11-13 | 2013-02-13 | 五邑大学 | Tendency text automatic classification system and achieving method of the same |
| CN103049435A (en) * | 2013-01-04 | 2013-04-17 | 浙江工商大学 | Text fine granularity sentiment analysis method and text fine granularity sentiment analysis device |
| CN103077207A (en) * | 2012-12-28 | 2013-05-01 | 深圳先进技术研究院 | Method and system for analyzing microblog happiness index |
| CN103116619A (en) * | 2013-01-29 | 2013-05-22 | 华为技术有限公司 | Collaboration business intelligence implementation method and device |
| CN103123620A (en) * | 2012-12-11 | 2013-05-29 | 中国互联网新闻中心 | Web text sentiment analysis method based on propositional logic |
| CN103150432A (en) * | 2013-03-07 | 2013-06-12 | 宁波成电泰克电子信息技术发展有限公司 | Method for internet public opinion analysis |
| CN103150367A (en) * | 2013-03-07 | 2013-06-12 | 宁波成电泰克电子信息技术发展有限公司 | Method for analyzing emotional tendency of Chinese microblogs |
| CN103207855A (en) * | 2013-04-12 | 2013-07-17 | 广东工业大学 | Fine-grained sentiment analysis system and method specific to product comment information |
| CN103440237A (en) * | 2013-03-15 | 2013-12-11 | 武汉元宝创意科技有限公司 | Microblog data processing visualization system based on 3D (3-dimensional) model |
| CN103455562A (en) * | 2013-08-13 | 2013-12-18 | 西安建筑科技大学 | Text orientation analysis method and product review orientation discriminator on basis of same |
| CN103544321A (en) * | 2013-11-06 | 2014-01-29 | 北京国双科技有限公司 | Data processing method and device for micro-blog emotion information |
| CN103559233A (en) * | 2012-10-29 | 2014-02-05 | 中国人民解放军国防科学技术大学 | Extraction method for network new words in microblogs and microblog emotion analysis method and system |
| CN103559176A (en) * | 2012-10-29 | 2014-02-05 | 中国人民解放军国防科学技术大学 | Microblog emotional evolution analysis method and system |
| CN103699626A (en) * | 2013-12-20 | 2014-04-02 | 华南理工大学 | Method and system for analysing individual emotion tendency of microblog user |
| CN103729456A (en) * | 2014-01-07 | 2014-04-16 | 合肥工业大学 | Microblog multi-modal sentiment analysis method based on microblog group environment |
| CN103744953A (en) * | 2014-01-02 | 2014-04-23 | 中国科学院计算机网络信息中心 | Network hotspot mining method based on Chinese text emotion recognition |
| CN103885933A (en) * | 2012-12-21 | 2014-06-25 | 富士通株式会社 | Method and equipment for evaluating text sentiment |
| CN103942340A (en) * | 2014-05-09 | 2014-07-23 | 电子科技大学 | Microblog user interest recognizing method based on text mining |
| CN103955451A (en) * | 2014-05-15 | 2014-07-30 | 北京优捷信达信息科技有限公司 | Method for judging emotional tendentiousness of short text |
| CN104111976A (en) * | 2014-06-24 | 2014-10-22 | 海南凯迪网络资讯有限公司 | Method and device for network speech emotion attitude localization |
| CN104182387A (en) * | 2014-07-21 | 2014-12-03 | 安徽华贞信息科技有限公司 | Text emotional tendency analysis system |
| CN104199845A (en) * | 2014-08-08 | 2014-12-10 | 杭州电子科技大学 | On-line comment sentiment classification method based on agent model |
| CN104281694A (en) * | 2014-10-13 | 2015-01-14 | 安徽华贞信息科技有限公司 | Analysis system of emotional tendency of text |
| TWI477987B (en) * | 2012-10-30 | 2015-03-21 | Univ Ming Chuan | Methods for sentimental analysis of news text |
| CN104484336A (en) * | 2014-11-19 | 2015-04-01 | 湖州师范学院 | Chinese commentary analysis method and system |
| CN104516947A (en) * | 2014-12-03 | 2015-04-15 | 浙江工业大学 | Chinese microblog emotion analysis method fused with dominant and recessive characters |
| CN104731770A (en) * | 2015-03-23 | 2015-06-24 | 中国科学技术大学苏州研究院 | Chinese microblog emotion analysis method based on rules and statistical model |
| CN104765757A (en) * | 2014-12-05 | 2015-07-08 | 华中科技大学 | Micro-blog timing sequence ranking method based on heterogeneous network |
| CN104778209A (en) * | 2015-03-13 | 2015-07-15 | 国家计算机网络与信息安全管理中心 | Opinion mining method for ten-million-scale news comments |
| CN104915443A (en) * | 2015-06-29 | 2015-09-16 | 北京信息科技大学 | Extraction method of Chinese Microblog evaluation object |
| CN105005560A (en) * | 2015-08-26 | 2015-10-28 | 苏州大学张家港工业技术研究院 | Maximum entropy model-based evaluation type emotion sorting method and system |
| CN105095190A (en) * | 2015-08-25 | 2015-11-25 | 众联数据技术（南京）有限公司 | Chinese semantic structure and finely segmented word bank combination based emotional analysis method |
| CN105144227A (en) * | 2013-01-02 | 2015-12-09 | 微软技术许可有限责任公司 | Social media impact assessment |
| CN105224640A (en) * | 2015-09-25 | 2016-01-06 | 杭州朗和科技有限公司 | A kind of method and apparatus extracting viewpoint |
| CN105243054A (en) * | 2015-09-23 | 2016-01-13 | 中国传媒大学 | Television program satisfaction subjective evaluation method and construction system |
| CN105404674A (en) * | 2015-11-20 | 2016-03-16 | 焦点科技股份有限公司 | Knowledge-dependent webpage information extraction method |
| CN105447196A (en) * | 2015-12-31 | 2016-03-30 | 深圳中泓在线股份有限公司 | Key blogger tracking confirmation method and device |
| CN105488206A (en) * | 2015-12-09 | 2016-04-13 | 扬州大学 | Crowdsourcing based android application evolution recommendation method |
| CN105574092A (en) * | 2015-12-10 | 2016-05-11 | 百度在线网络技术（北京）有限公司 | Information mining method and device |
| CN105630928A (en) * | 2015-12-22 | 2016-06-01 | 北京奇虎科技有限公司 | Text marking method and apparatus |
| CN105740224A (en) * | 2014-12-11 | 2016-07-06 | 仲恺农业工程学院 | Text analysis based user psychology early warning method and apparatus |
| CN105843796A (en) * | 2016-03-28 | 2016-08-10 | 北京邮电大学 | Microblog emotional tendency analysis method and device |
| CN106021433A (en) * | 2016-05-16 | 2016-10-12 | 北京百分点信息科技有限公司 | Public praise analysis method and apparatus for product review data |
| CN106202032A (en) * | 2016-06-24 | 2016-12-07 | 广州数说故事信息科技有限公司 | A kind of sentiment analysis method towards microblogging short text and system thereof |
| CN106202047A (en) * | 2016-07-15 | 2016-12-07 | 国家计算机网络与信息安全管理中心 | A kind of character personality depicting method based on microblogging text |
| CN106202584A (en) * | 2016-09-20 | 2016-12-07 | 北京工业大学 | A kind of microblog emotional based on standard dictionary and semantic rule analyzes method |
| CN106294312A (en) * | 2015-06-29 | 2017-01-04 | 北京大学 | Information processing method and information processing system |
| CN106384245A (en) * | 2016-09-06 | 2017-02-08 | 合肥工业大学 | Product feature analysis method and system |
| CN106446147A (en) * | 2016-09-20 | 2017-02-22 | 天津大学 | Emotion analysis method based on structuring features |
| CN106796607A (en) * | 2014-12-29 | 2017-05-31 | 华为技术有限公司 | For the system and method that the search based on model and network data are retrieved |
| CN106796583A (en) * | 2014-07-07 | 2017-05-31 | 机械地带有限公司 | System and method for recognizing and advising emoticon |
| CN106776551A (en) * | 2016-12-06 | 2017-05-31 | 桂林电子科技大学 | A kind of analysis method of english composition emotion viewpoint |
| CN106980692A (en) * | 2016-05-30 | 2017-07-25 | 国家计算机网络与信息安全管理中心 | A kind of influence power computational methods based on microblogging particular event |
| CN107229610A (en) * | 2017-03-17 | 2017-10-03 | 咪咕数字传媒有限公司 | The analysis method and device of a kind of affection data |
| CN107315831A (en) * | 2017-07-10 | 2017-11-03 | 北京神州泰岳软件股份有限公司 | A kind of method and device of the unknown incidence relation of mining rule correlation model |
| CN107609132A (en) * | 2017-09-18 | 2018-01-19 | 杭州电子科技大学 | One kind is based on Ontology storehouse Chinese text sentiment analysis method |
| CN107748743A (en) * | 2017-09-20 | 2018-03-02 | 安徽商贸职业技术学院 | A kind of electric business online comment text emotion analysis method |
| CN107943787A (en) * | 2017-11-16 | 2018-04-20 | 北京百度网讯科技有限公司 | Collect method, apparatus, equipment and the computer-readable medium of user feedback |
| CN108038166A (en) * | 2017-12-06 | 2018-05-15 | 武汉大学 | A kind of Chinese microblog emotional analysis method based on the subjective and objective skewed popularity of lexical item |
| CN108319587A (en) * | 2018-02-05 | 2018-07-24 | 中译语通科技股份有限公司 | A kind of public sentiment value calculation method and system of more weights, computer |
| CN108363805A (en) * | 2018-03-01 | 2018-08-03 | 大连理工大学 | A kind of model sequencing method based on product feature public praise |
| CN108399158A (en) * | 2018-02-05 | 2018-08-14 | 华南理工大学 | Attribute sensibility classification method based on dependency tree and attention mechanism |
| CN108536762A (en) * | 2018-03-21 | 2018-09-14 | 上海蔚界信息科技有限公司 | A kind of high-volume text data automatically analyzes scheme |
| CN108563731A (en) * | 2018-04-08 | 2018-09-21 | 北京奇艺世纪科技有限公司 | A kind of sensibility classification method and device |
| CN108694165A (en) * | 2017-04-10 | 2018-10-23 | 南京理工大学 | Cross-cutting antithesis sentiment analysis method towards product review |
| CN108877336A (en) * | 2018-03-26 | 2018-11-23 | 深圳市波心幻海科技有限公司 | Teaching method, cloud service platform and tutoring system based on augmented reality |
| CN108932227A (en) * | 2018-06-05 | 2018-12-04 | 天津大学 | A kind of short text emotion value calculating method based on sentence structure and context |
| US10169419B2 (en) | 2010-12-27 | 2019-01-01 | Microsoft Technology Licensing, Llc | System and method for generating social summaries |
| CN109145306A (en) * | 2018-09-11 | 2019-01-04 | 刘瑞军 | The three-dimensional expression generation method of text-driven |
| CN109240558A (en) * | 2018-07-23 | 2019-01-18 | 中国农业大学 | A kind of the emotion initiation reason mask method and system of facing multiple users microblogging |
| 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 |
| CN109299226A (en) * | 2018-10-25 | 2019-02-01 | 北京奇艺世纪科技有限公司 | A kind of data processing method and system |
| CN109344331A (en) * | 2018-10-26 | 2019-02-15 | 南京邮电大学 | A kind of user feeling analysis method based on online community network |
| CN109919641A (en) * | 2017-12-12 | 2019-06-21 | 优视科技有限公司 | A kind of advertisement placement method and platform |
| CN109933793A (en) * | 2019-03-15 | 2019-06-25 | 腾讯科技（深圳）有限公司 | Text polarity identification method, apparatus, equipment and readable storage medium storing program for executing |
| CN109933649A (en) * | 2019-03-14 | 2019-06-25 | 武汉烽火普天信息技术有限公司 | A kind of case means abstracting method based on classified lexicon and heuristic rule |
| CN109948139A (en) * | 2017-12-19 | 2019-06-28 | 优酷网络技术（北京）有限公司 | A kind of semantic tendency analysis method and system |
| CN110046239A (en) * | 2019-04-15 | 2019-07-23 | 合肥工业大学 | Dialogue method based on emotion editor |
| CN110096696A (en) * | 2018-06-11 | 2019-08-06 | 电子科技大学 | A kind of Chinese long text sentiment analysis method |
| CN110309506A (en) * | 2019-05-28 | 2019-10-08 | 北京三快在线科技有限公司 | Statement analytical method, device, electronic equipment and readable storage medium storing program for executing |
| CN110321562A (en) * | 2019-06-28 | 2019-10-11 | 广州探迹科技有限公司 | A kind of short text matching process and device based on BERT |
| CN110427621A (en) * | 2019-07-23 | 2019-11-08 | 北京语言大学 | A kind of Chinese classification term extraction method and system |
| CN110502744A (en) * | 2019-07-15 | 2019-11-26 | 同济大学 | A kind of text emotion recognition methods and device for history park evaluation |
| CN110705295A (en) * | 2019-09-11 | 2020-01-17 | 北京航空航天大学 | Entity name disambiguation method based on keyword extraction |
| CN110929026A (en) * | 2018-09-19 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Abnormal text recognition method and device, computing equipment and medium |
| CN110941759A (en) * | 2019-11-20 | 2020-03-31 | 国元证券股份有限公司 | Microblog emotion analysis method |
| CN110942337A (en) * | 2019-10-31 | 2020-03-31 | 天津中科智能识别产业技术研究院有限公司 | Accurate tourism marketing method based on internet big data |
| CN111310476A (en) * | 2020-02-21 | 2020-06-19 | 山东大学 | Public opinion monitoring method and system using aspect-based emotion analysis method |
| CN111985223A (en) * | 2020-08-25 | 2020-11-24 | 武汉长江通信产业集团股份有限公司 | Emotion calculation method based on combination of long and short memory networks and emotion dictionaries |
| CN112036980A (en) * | 2020-08-31 | 2020-12-04 | 北京明略昭辉科技有限公司 | Article recommendation method and device, electronic equipment and storage medium |
| CN112507071A (en) * | 2020-12-03 | 2021-03-16 | 南京邮电大学 | Network platform short text mixed emotion classification method based on novel emotion dictionary |
| CN112632272A (en) * | 2020-10-20 | 2021-04-09 | 浙江工业大学 | Microblog emotion classification method and system based on syntactic analysis |
| CN112765350A (en) * | 2021-01-15 | 2021-05-07 | 西华大学 | Microblog comment emotion classification method based on emoticons and text information |
| CN113111269A (en) * | 2021-05-10 | 2021-07-13 | 网易（杭州）网络有限公司 | Data processing method and device, computer readable storage medium and electronic equipment |
| WO2021139107A1 (en) * | 2020-01-10 | 2021-07-15 | 平安科技（深圳）有限公司 | Intelligent emotion recognition method and apparatus, electronic device, and storage medium |
| CN113298366A (en) * | 2021-05-12 | 2021-08-24 | 北京信息科技大学 | Tourism performance service value evaluation method |
| 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 |
| CN114676243A (en) * | 2022-05-25 | 2022-06-28 | 成都无糖信息技术有限公司 | User portrait analysis method and system for social text |
| US11860944B2 (en) | 2020-07-27 | 2024-01-02 | International Business Machines Corporation | State-aware interface |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236636A (en) * | 2010-04-26 | 2011-11-09 | 富士通株式会社 | Method and device for analyzing emotional tendency |
-
2012
- 2012-03-29 CN CN201210088366XA patent/CN102663046A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102236636A (en) * | 2010-04-26 | 2011-11-09 | 富士通株式会社 | Method and device for analyzing emotional tendency |
Non-Patent Citations (3)
| Title |
|---|
| 俞飞: "基于网络信息文本倾向性分析的领域应用研究", 《中国优秀硕士学位论文全文数据库》 * |
| 姚天昉等: "汉语语句主题语义倾向分析方法的研究", 《中文信息学报》 * |
| 谢丽星等: "基于层次结构的多策略中文微博情感分析和特征抽取", 《中文信息学报》 * |
Cited By (145)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10169419B2 (en) | 2010-12-27 | 2019-01-01 | Microsoft Technology Licensing, Llc | System and method for generating social summaries |
| CN103559233A (en) * | 2012-10-29 | 2014-02-05 | 中国人民解放军国防科学技术大学 | Extraction method for network new words in microblogs and microblog emotion analysis method and system |
| CN103559176B (en) * | 2012-10-29 | 2016-08-17 | 中国人民解放军国防科学技术大学 | Microblog emotional evolution analysis method and system |
| CN103559176A (en) * | 2012-10-29 | 2014-02-05 | 中国人民解放军国防科学技术大学 | Microblog emotional evolution analysis method and system |
| TWI477987B (en) * | 2012-10-30 | 2015-03-21 | Univ Ming Chuan | Methods for sentimental analysis of news text |
| CN102930042A (en) * | 2012-11-13 | 2013-02-13 | 五邑大学 | Tendency text automatic classification system and achieving method of the same |
| CN103123620A (en) * | 2012-12-11 | 2013-05-29 | 中国互联网新闻中心 | Web text sentiment analysis method based on propositional logic |
| CN103885933B (en) * | 2012-12-21 | 2017-03-01 | 富士通株式会社 | For evaluating emotion degree and the method and apparatus for evaluating entity of text |
| CN103885933A (en) * | 2012-12-21 | 2014-06-25 | 富士通株式会社 | Method and equipment for evaluating text sentiment |
| CN103077207B (en) * | 2012-12-28 | 2016-09-07 | 深圳先进技术研究院 | A kind of microblogging happy index analysis method and system |
| CN103077207A (en) * | 2012-12-28 | 2013-05-01 | 深圳先进技术研究院 | Method and system for analyzing microblog happiness index |
| CN105144227A (en) * | 2013-01-02 | 2015-12-09 | 微软技术许可有限责任公司 | Social media impact assessment |
| US10614077B2 (en) | 2013-01-02 | 2020-04-07 | Microsoft Corporation | Computer system for automated assessment at scale of topic-specific social media impact |
| CN103049435A (en) * | 2013-01-04 | 2013-04-17 | 浙江工商大学 | Text fine granularity sentiment analysis method and text fine granularity sentiment analysis device |
| CN103049435B (en) * | 2013-01-04 | 2015-10-14 | 浙江工商大学 | Text fine granularity sentiment analysis method and device |
| CN103116619B (en) * | 2013-01-29 | 2016-07-06 | 华为技术有限公司 | Collaborative business intelligence realizes method and device |
| CN103116619A (en) * | 2013-01-29 | 2013-05-22 | 华为技术有限公司 | Collaboration business intelligence implementation method and device |
| CN103150432B (en) * | 2013-03-07 | 2016-05-11 | 宁波成电泰克电子信息技术发展有限公司 | A kind of Internet public opinion analysis method |
| CN103150367B (en) * | 2013-03-07 | 2016-01-20 | 宁波成电泰克电子信息技术发展有限公司 | A kind of Sentiment orientation analytical approach of Chinese microblogging |
| CN103150432A (en) * | 2013-03-07 | 2013-06-12 | 宁波成电泰克电子信息技术发展有限公司 | Method for internet public opinion analysis |
| CN103150367A (en) * | 2013-03-07 | 2013-06-12 | 宁波成电泰克电子信息技术发展有限公司 | Method for analyzing emotional tendency of Chinese microblogs |
| CN103440237A (en) * | 2013-03-15 | 2013-12-11 | 武汉元宝创意科技有限公司 | Microblog data processing visualization system based on 3D (3-dimensional) model |
| CN103207855A (en) * | 2013-04-12 | 2013-07-17 | 广东工业大学 | Fine-grained sentiment analysis system and method specific to product comment information |
| CN103207855B (en) * | 2013-04-12 | 2019-04-26 | 广东工业大学 | For the fine granularity sentiment analysis system and method for product review information |
| CN103455562A (en) * | 2013-08-13 | 2013-12-18 | 西安建筑科技大学 | Text orientation analysis method and product review orientation discriminator on basis of same |
| CN103544321A (en) * | 2013-11-06 | 2014-01-29 | 北京国双科技有限公司 | Data processing method and device for micro-blog emotion information |
| CN103699626A (en) * | 2013-12-20 | 2014-04-02 | 华南理工大学 | Method and system for analysing individual emotion tendency of microblog user |
| CN103699626B (en) * | 2013-12-20 | 2017-02-01 | 华南理工大学 | Method and system for analysing individual emotion tendency of microblog user |
| CN103744953A (en) * | 2014-01-02 | 2014-04-23 | 中国科学院计算机网络信息中心 | Network hotspot mining method based on Chinese text emotion recognition |
| CN103729456B (en) * | 2014-01-07 | 2016-09-28 | 合肥工业大学 | Microblog multi-modal sentiment analysis method based on microblog group environment |
| CN103729456A (en) * | 2014-01-07 | 2014-04-16 | 合肥工业大学 | Microblog multi-modal sentiment analysis method based on microblog group environment |
| CN103942340A (en) * | 2014-05-09 | 2014-07-23 | 电子科技大学 | Microblog user interest recognizing method based on text mining |
| CN103955451A (en) * | 2014-05-15 | 2014-07-30 | 北京优捷信达信息科技有限公司 | Method for judging emotional tendentiousness of short text |
| CN103955451B (en) * | 2014-05-15 | 2017-04-19 | 北京优捷信达信息科技有限公司 | Method for judging emotional tendentiousness of short text |
| CN104111976B (en) * | 2014-06-24 | 2017-04-05 | 海南凯迪网络资讯股份有限公司 | Network speech emotion attitude localization method and device |
| CN104111976A (en) * | 2014-06-24 | 2014-10-22 | 海南凯迪网络资讯有限公司 | Method and device for network speech emotion attitude localization |
| CN106796583A (en) * | 2014-07-07 | 2017-05-31 | 机械地带有限公司 | System and method for recognizing and advising emoticon |
| CN104182387A (en) * | 2014-07-21 | 2014-12-03 | 安徽华贞信息科技有限公司 | Text emotional tendency analysis system |
| CN104199845B (en) * | 2014-08-08 | 2018-05-29 | 杭州电子科技大学 | Line Evaluation based on agent model discusses sensibility classification method |
| CN104199845A (en) * | 2014-08-08 | 2014-12-10 | 杭州电子科技大学 | On-line comment sentiment classification method based on agent model |
| CN104281694A (en) * | 2014-10-13 | 2015-01-14 | 安徽华贞信息科技有限公司 | Analysis system of emotional tendency of text |
| CN104484336B (en) * | 2014-11-19 | 2017-12-19 | 湖州师范学院 | A kind of Chinese comment and analysis method and its system |
| CN104484336A (en) * | 2014-11-19 | 2015-04-01 | 湖州师范学院 | Chinese commentary analysis method and system |
| CN104516947B (en) * | 2014-12-03 | 2017-08-22 | 浙江工业大学 | A kind of Chinese microblog emotional analysis method for merging dominant and recessive character |
| CN104516947A (en) * | 2014-12-03 | 2015-04-15 | 浙江工业大学 | Chinese microblog emotion analysis method fused with dominant and recessive characters |
| CN104765757A (en) * | 2014-12-05 | 2015-07-08 | 华中科技大学 | Micro-blog timing sequence ranking method based on heterogeneous network |
| CN105740224A (en) * | 2014-12-11 | 2016-07-06 | 仲恺农业工程学院 | Text analysis based user psychology early warning method and apparatus |
| CN106796607A (en) * | 2014-12-29 | 2017-05-31 | 华为技术有限公司 | For the system and method that the search based on model and network data are retrieved |
| CN104778209B (en) * | 2015-03-13 | 2018-04-27 | 国家计算机网络与信息安全管理中心 | A kind of opining mining method for millions scale news analysis |
| CN104778209A (en) * | 2015-03-13 | 2015-07-15 | 国家计算机网络与信息安全管理中心 | Opinion mining method for ten-million-scale news comments |
| CN104731770A (en) * | 2015-03-23 | 2015-06-24 | 中国科学技术大学苏州研究院 | Chinese microblog emotion analysis method based on rules and statistical model |
| CN106294312B (en) * | 2015-06-29 | 2019-02-26 | 北京大学 | Information processing method and information processing system |
| CN104915443B (en) * | 2015-06-29 | 2018-11-23 | 北京信息科技大学 | A kind of abstracting method of Chinese microblogging evaluation object |
| CN106294312A (en) * | 2015-06-29 | 2017-01-04 | 北京大学 | Information processing method and information processing system |
| CN104915443A (en) * | 2015-06-29 | 2015-09-16 | 北京信息科技大学 | Extraction method of Chinese Microblog evaluation object |
| CN105095190B (en) * | 2015-08-25 | 2018-01-12 | 众联数据技术（南京）有限公司 | A kind of sentiment analysis method combined based on Chinese semantic structure and subdivision dictionary |
| CN105095190A (en) * | 2015-08-25 | 2015-11-25 | 众联数据技术（南京）有限公司 | Chinese semantic structure and finely segmented word bank combination based emotional analysis method |
| CN105005560A (en) * | 2015-08-26 | 2015-10-28 | 苏州大学张家港工业技术研究院 | Maximum entropy model-based evaluation type emotion sorting method and system |
| CN105243054A (en) * | 2015-09-23 | 2016-01-13 | 中国传媒大学 | Television program satisfaction subjective evaluation method and construction system |
| CN105243054B (en) * | 2015-09-23 | 2017-12-29 | 中国传媒大学 | A kind of TV programme satisfaction subjective evaluation method and construction system |
| CN105224640A (en) * | 2015-09-25 | 2016-01-06 | 杭州朗和科技有限公司 | A kind of method and apparatus extracting viewpoint |
| CN105404674B (en) * | 2015-11-20 | 2017-02-22 | 焦点科技股份有限公司 | Knowledge-dependent webpage information extraction method |
| CN105404674A (en) * | 2015-11-20 | 2016-03-16 | 焦点科技股份有限公司 | Knowledge-dependent webpage information extraction method |
| CN105488206A (en) * | 2015-12-09 | 2016-04-13 | 扬州大学 | Crowdsourcing based android application evolution recommendation method |
| CN105488206B (en) * | 2015-12-09 | 2019-03-26 | 扬州大学 | A kind of Android application evolution recommended method based on crowdsourcing |
| CN105574092A (en) * | 2015-12-10 | 2016-05-11 | 百度在线网络技术（北京）有限公司 | Information mining method and device |
| CN105574092B (en) * | 2015-12-10 | 2019-08-23 | 百度在线网络技术（北京）有限公司 | Information mining method and device |
| CN105630928A (en) * | 2015-12-22 | 2016-06-01 | 北京奇虎科技有限公司 | Text marking method and apparatus |
| CN105447196A (en) * | 2015-12-31 | 2016-03-30 | 深圳中泓在线股份有限公司 | Key blogger tracking confirmation method and device |
| CN105447196B (en) * | 2015-12-31 | 2019-03-05 | 深圳中泓在线股份有限公司 | A kind of emphasis bloger tracks confirmation method and device |
| CN105843796A (en) * | 2016-03-28 | 2016-08-10 | 北京邮电大学 | Microblog emotional tendency analysis method and device |
| CN106021433B (en) * | 2016-05-16 | 2019-05-10 | 北京百分点信息科技有限公司 | A kind of the public praise analysis method and device of comment on commodity data |
| CN106021433A (en) * | 2016-05-16 | 2016-10-12 | 北京百分点信息科技有限公司 | Public praise analysis method and apparatus for product review data |
| CN106980692A (en) * | 2016-05-30 | 2017-07-25 | 国家计算机网络与信息安全管理中心 | A kind of influence power computational methods based on microblogging particular event |
| CN106980692B (en) * | 2016-05-30 | 2020-12-08 | 国家计算机网络与信息安全管理中心 | Influence calculation method based on microblog specific events |
| CN106202032A (en) * | 2016-06-24 | 2016-12-07 | 广州数说故事信息科技有限公司 | A kind of sentiment analysis method towards microblogging short text and system thereof |
| CN106202032B (en) * | 2016-06-24 | 2018-08-28 | 广州数说故事信息科技有限公司 | A kind of sentiment analysis method and its system towards microblogging short text |
| CN106202047A (en) * | 2016-07-15 | 2016-12-07 | 国家计算机网络与信息安全管理中心 | A kind of character personality depicting method based on microblogging text |
| CN106384245A (en) * | 2016-09-06 | 2017-02-08 | 合肥工业大学 | Product feature analysis method and system |
| CN106202584A (en) * | 2016-09-20 | 2016-12-07 | 北京工业大学 | A kind of microblog emotional based on standard dictionary and semantic rule analyzes method |
| CN106446147A (en) * | 2016-09-20 | 2017-02-22 | 天津大学 | Emotion analysis method based on structuring features |
| CN106776551B (en) * | 2016-12-06 | 2020-05-08 | 桂林电子科技大学 | Method for analyzing emotion viewpoints of English composition |
| CN106776551A (en) * | 2016-12-06 | 2017-05-31 | 桂林电子科技大学 | A kind of analysis method of english composition emotion viewpoint |
| CN107229610B (en) * | 2017-03-17 | 2019-06-21 | 咪咕数字传媒有限公司 | A kind of analysis method and device of affection data |
| CN107229610A (en) * | 2017-03-17 | 2017-10-03 | 咪咕数字传媒有限公司 | The analysis method and device of a kind of affection data |
| CN108694165A (en) * | 2017-04-10 | 2018-10-23 | 南京理工大学 | Cross-cutting antithesis sentiment analysis method towards product review |
| CN108694165B (en) * | 2017-04-10 | 2021-11-09 | 南京理工大学 | Cross-domain dual emotion analysis method for product comments |
| CN107315831B (en) * | 2017-07-10 | 2019-06-07 | 北京神州泰岳软件股份有限公司 | A kind of method and device of the unknown incidence relation of mining rule correlation model |
| CN107315831A (en) * | 2017-07-10 | 2017-11-03 | 北京神州泰岳软件股份有限公司 | A kind of method and device of the unknown incidence relation of mining rule correlation model |
| CN107609132A (en) * | 2017-09-18 | 2018-01-19 | 杭州电子科技大学 | One kind is based on Ontology storehouse Chinese text sentiment analysis method |
| CN107609132B (en) * | 2017-09-18 | 2020-03-20 | 杭州电子科技大学 | Semantic ontology base based Chinese text sentiment analysis method |
| CN107748743A (en) * | 2017-09-20 | 2018-03-02 | 安徽商贸职业技术学院 | A kind of electric business online comment text emotion analysis method |
| CN107943787A (en) * | 2017-11-16 | 2018-04-20 | 北京百度网讯科技有限公司 | Collect method, apparatus, equipment and the computer-readable medium of user feedback |
| CN108038166A (en) * | 2017-12-06 | 2018-05-15 | 武汉大学 | A kind of Chinese microblog emotional analysis method based on the subjective and objective skewed popularity of lexical item |
| CN109919641A (en) * | 2017-12-12 | 2019-06-21 | 优视科技有限公司 | A kind of advertisement placement method and platform |
| CN109948139A (en) * | 2017-12-19 | 2019-06-28 | 优酷网络技术（北京）有限公司 | A kind of semantic tendency analysis method and system |
| CN108319587A (en) * | 2018-02-05 | 2018-07-24 | 中译语通科技股份有限公司 | A kind of public sentiment value calculation method and system of more weights, computer |
| CN108319587B (en) * | 2018-02-05 | 2021-11-19 | 中译语通科技股份有限公司 | Multi-weight public opinion value calculation method and system and computer |
| CN108399158A (en) * | 2018-02-05 | 2018-08-14 | 华南理工大学 | Attribute sensibility classification method based on dependency tree and attention mechanism |
| CN108363805B (en) * | 2018-03-01 | 2020-09-29 | 大连理工大学 | Product sorting method based on product feature public praise |
| CN108363805A (en) * | 2018-03-01 | 2018-08-03 | 大连理工大学 | A kind of model sequencing method based on product feature public praise |
| CN108536762A (en) * | 2018-03-21 | 2018-09-14 | 上海蔚界信息科技有限公司 | A kind of high-volume text data automatically analyzes scheme |
| CN108877336A (en) * | 2018-03-26 | 2018-11-23 | 深圳市波心幻海科技有限公司 | Teaching method, cloud service platform and tutoring system based on augmented reality |
| CN108563731A (en) * | 2018-04-08 | 2018-09-21 | 北京奇艺世纪科技有限公司 | A kind of sensibility classification method and device |
| CN108932227A (en) * | 2018-06-05 | 2018-12-04 | 天津大学 | A kind of short text emotion value calculating method based on sentence structure and context |
| CN110096696A (en) * | 2018-06-11 | 2019-08-06 | 电子科技大学 | A kind of Chinese long text sentiment analysis method |
| CN109240558A (en) * | 2018-07-23 | 2019-01-18 | 中国农业大学 | A kind of the emotion initiation reason mask method and system of facing multiple users microblogging |
| CN109145306A (en) * | 2018-09-11 | 2019-01-04 | 刘瑞军 | The three-dimensional expression generation method of text-driven |
| 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 |
| CN109271634B (en) * | 2018-09-17 | 2022-07-01 | 重庆理工大学 | Microblog text emotion polarity analysis method based on user emotion tendency perception |
| CN110929026A (en) * | 2018-09-19 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Abnormal text recognition method and device, computing equipment and medium |
| CN110929026B (en) * | 2018-09-19 | 2023-04-25 | 阿里巴巴集团控股有限公司 | Abnormal text recognition method, device, computing equipment and medium |
| CN109299226A (en) * | 2018-10-25 | 2019-02-01 | 北京奇艺世纪科技有限公司 | A kind of data processing method and system |
| CN109344331A (en) * | 2018-10-26 | 2019-02-15 | 南京邮电大学 | A kind of user feeling analysis method based on online community network |
| CN109933649A (en) * | 2019-03-14 | 2019-06-25 | 武汉烽火普天信息技术有限公司 | A kind of case means abstracting method based on classified lexicon and heuristic rule |
| CN109933793B (en) * | 2019-03-15 | 2023-01-06 | 腾讯科技（深圳）有限公司 | Text polarity identification method, device and equipment and readable storage medium |
| CN109933793A (en) * | 2019-03-15 | 2019-06-25 | 腾讯科技（深圳）有限公司 | Text polarity identification method, apparatus, equipment and readable storage medium storing program for executing |
| CN110046239A (en) * | 2019-04-15 | 2019-07-23 | 合肥工业大学 | Dialogue method based on emotion editor |
| CN110309506A (en) * | 2019-05-28 | 2019-10-08 | 北京三快在线科技有限公司 | Statement analytical method, device, electronic equipment and readable storage medium storing program for executing |
| CN110321562A (en) * | 2019-06-28 | 2019-10-11 | 广州探迹科技有限公司 | A kind of short text matching process and device based on BERT |
| CN110502744A (en) * | 2019-07-15 | 2019-11-26 | 同济大学 | A kind of text emotion recognition methods and device for history park evaluation |
| CN110427621B (en) * | 2019-07-23 | 2020-11-20 | 北京语言大学 | Chinese classified word extraction method and system |
| CN110427621A (en) * | 2019-07-23 | 2019-11-08 | 北京语言大学 | A kind of Chinese classification term extraction method and system |
| CN110705295A (en) * | 2019-09-11 | 2020-01-17 | 北京航空航天大学 | Entity name disambiguation method based on keyword extraction |
| CN110705295B (en) * | 2019-09-11 | 2021-08-24 | 北京航空航天大学 | Entity name disambiguation method based on keyword extraction |
| CN110942337A (en) * | 2019-10-31 | 2020-03-31 | 天津中科智能识别产业技术研究院有限公司 | Accurate tourism marketing method based on internet big data |
| CN110941759A (en) * | 2019-11-20 | 2020-03-31 | 国元证券股份有限公司 | Microblog emotion analysis method |
| WO2021139107A1 (en) * | 2020-01-10 | 2021-07-15 | 平安科技（深圳）有限公司 | Intelligent emotion recognition method and apparatus, electronic device, and storage medium |
| CN111310476B (en) * | 2020-02-21 | 2021-11-02 | 山东大学 | Public opinion monitoring method and system using aspect-based emotion analysis method |
| CN111310476A (en) * | 2020-02-21 | 2020-06-19 | 山东大学 | Public opinion monitoring method and system using aspect-based emotion analysis method |
| US11860944B2 (en) | 2020-07-27 | 2024-01-02 | International Business Machines Corporation | State-aware interface |
| CN111985223A (en) * | 2020-08-25 | 2020-11-24 | 武汉长江通信产业集团股份有限公司 | Emotion calculation method based on combination of long and short memory networks and emotion dictionaries |
| CN112036980A (en) * | 2020-08-31 | 2020-12-04 | 北京明略昭辉科技有限公司 | Article recommendation method and device, electronic equipment and storage medium |
| CN112632272A (en) * | 2020-10-20 | 2021-04-09 | 浙江工业大学 | Microblog emotion classification method and system based on syntactic analysis |
| CN112632272B (en) * | 2020-10-20 | 2022-07-19 | 浙江工业大学 | Microblog emotion classification method and system based on syntactic analysis |
| CN112507071A (en) * | 2020-12-03 | 2021-03-16 | 南京邮电大学 | Network platform short text mixed emotion classification method based on novel emotion dictionary |
| CN112507071B (en) * | 2020-12-03 | 2022-10-14 | 南京邮电大学 | Network platform short text mixed emotion classification method based on novel emotion dictionary |
| CN112765350A (en) * | 2021-01-15 | 2021-05-07 | 西华大学 | Microblog comment emotion classification method based on emoticons and text information |
| 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 |
| CN113378576A (en) * | 2021-05-08 | 2021-09-10 | 重庆航天信息有限公司 | Food safety data mining method |
| CN113111269A (en) * | 2021-05-10 | 2021-07-13 | 网易（杭州）网络有限公司 | Data processing method and device, computer readable storage medium and electronic equipment |
| CN113298366A (en) * | 2021-05-12 | 2021-08-24 | 北京信息科技大学 | Tourism performance service value evaluation method |
| CN113298366B (en) * | 2021-05-12 | 2023-12-12 | 北京信息科技大学 | Travel performance service value assessment method |
| CN114676243A (en) * | 2022-05-25 | 2022-06-28 | 成都无糖信息技术有限公司 | User portrait analysis method and system for social text |
| CN114676243B (en) * | 2022-05-25 | 2022-08-19 | 成都无糖信息技术有限公司 | User portrait analysis method and system for social text |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102663046A (en) | Sentiment analysis method oriented to micro-blog short text | |
| Mäntylä et al. | The evolution of sentiment analysis—A review of research topics, venues, and top cited papers | |
| Ma et al. | Sentiment analysis–a review and agenda for future research in hospitality contexts | |
| CN107577759B (en) | Automatic recommendation method for user comments | |
| Kaur et al. | A survey on sentiment analysis and opinion mining techniques | |
| Cambria et al. | New avenues in opinion mining and sentiment analysis | |
| Kanan et al. | A review of natural language processing and machine learning tools used to analyze arabic social media | |
| Malouf et al. | Taking sides: User classification for informal online political discourse | |
| Sun et al. | Pre-processing online financial text for sentiment classification: A natural language processing approach | |
| Hui et al. | Effects of word class and text position in sentiment-based news classification | |
| Yergesh et al. | Sentiment analysis of Kazakh text and their polarity | |
| Bach et al. | Big data text mining in the financial sector | |
| Atoum | Detecting cyberbullying from tweets through machine learning techniques with sentiment analysis | |
| Stylios et al. | Using Bio-inspired intelligence for Web opinion Mining | |
| Jeevanandam Jotheeswaran | Sentiment analysis: A survey of current research and techniques | |
| Wang et al. | Unsupervised opinion phrase extraction and rating in Chinese blog posts | |
| Sainger | Sentiment analysis-an assessment of online public opinion: a conceptual review | |
| Li et al. | Opinion mining of camera reviews based on Semantic Role Labeling | |
| Hamroun et al. | Large scale microblogging intentions analysis with pattern based approach | |
| Wang et al. | Natural language processing systems and Big Data analytics | |
| Tsai et al. | User feedback in controllable and explainable social recommender systems: a linguistic analysis | |
| Abuteir et al. | Automatic Sarcasm Detection in Arabic Text: A Supervised Classification Approach | |
| Kumar et al. | Twitter based information extraction | |
| Zainab et al. | Comparative analysis of machine learning algorithms for author age and gender identification | |
| Nandy et al. | Filtering-Based Text Sentiment Analysis for Twitter Dataset |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120912 |