CN106354754A - Bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis - Google Patents
Bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis Download PDFInfo
- Publication number
- CN106354754A CN106354754A CN201610677146.9A CN201610677146A CN106354754A CN 106354754 A CN106354754 A CN 106354754A CN 201610677146 A CN201610677146 A CN 201610677146A CN 106354754 A CN106354754 A CN 106354754A
- Authority
- CN
- China
- Prior art keywords
- feature
- word
- viewpoint
- words
- component analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis. The method includes the steps that multiple pieces of historical comment information are collected; a plurality of viewpoint words and a plurality of characteristic words of the multiple pieces of historical comment information are obtained with the dispersed independent component analysis method; according to the multiple viewpoint words and the multiple characteristic words, the viewpoint characteristic association rule is formed; current viewpoint words are obtained, and according to the viewpoint characteristic association rule, characteristic words corresponding to the current viewpoint words are obtained. The bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis have the advantages that extracting is carried out through bootstrap-type characteristics, association between explicit characteristic words and characteristic words, association between characteristic words and view words and association between viewpoint words and viewpoint words are obtained, matched characteristic words of viewpoint words in implicit characteristic comments are obtained according to the association rule, viewpoint mining samples can be completed, and the viewpoint mining effect is improved.
Description
Technical field
The present invention relates to computer utility machine learning field and in particular to a kind of based on discrete independent component analysis from
Exhibition formula implicit features method for digging and system.
Background technology
Opining mining, also referred to as emotion are analyzed, and are to be produced with regard to the feature of a certain entity, assembly, attribute etc. for people
A kind of technology that the viewpoint of life, attitude and emotion are excavated and analyzed, is an important neck in natural language processing subject
Domain, has obtained the attention of academia and industry.
Bonded products are commented on, and opining mining can highly desirable react the concrete viewpoint for certain concrete things of user,
There is real-time, topic sensitivity and polytropy.Most models often utilize supervised learning or half to product review
Carrying out feature extraction, this is often because what the field limitation of product review caused supervised learning: a same feature
Word is mostly unequal in the weight of different professional fields, and for example " power " one word is in the product review of the vehicles such as automobile
Occupy an important position, but be then useless " noise " comment in the comment of the digital products such as mobile phone.In conventional work,
Extraction for feature generally requires manually to demarcate so that the efficiency of feature extraction is very low.
Feature Words do not occur directly in sentence in comment and are referred to as implicit expression comment, such as: " someone very like it is simply that not
Easily put pocket into." at this with regard in the comment of mobile phone, Feature Words " size " with regard to mobile phone characteristic or " handset size " and
Do not occur, but people are it can be seen that this comment is that mobile phone size is described.The spy being described by implicit features
Levy and be known as implicit features.Quality a big chunk of opining mining depends on the quality of feature mining.In conventional work,
The researchers of opining mining often pay close attention to display feature critiques, and that is, feature occurs directly in the selection of the comment in sentence, and
Have ignored the comment of implicit features, thus leading to the accuracy to opining mining low.
Content of the invention
It is contemplated that at least solving one of above-mentioned technical problem.
For this reason, it is an object of the present invention to proposing a kind of auto-expanding type implicit features based on discrete independent component analysis
Method for digging.
Further object is that proposition is a kind of being dug based on the auto-expanding type implicit features of discrete independent component analysis
Pick system.
To achieve these goals, embodiment of the invention discloses that a kind of auto-expanding type based on discrete independent component analysis
Implicit features method for digging, comprises the following steps: s1: gathers multiple historical review information;S2: using discrete independent component analysis
Method obtains multiple viewpoint words and multiple Feature Words in the plurality of historical review information;S3: according to the plurality of viewpoint word
Form viewpoint feature association rule with the plurality of Feature Words;S4: obtain current view word, and according to described viewpoint feature association
Rule obtains the corresponding Feature Words of described current view word.
Auto-expanding type implicit features method for digging based on discrete independent component analysis according to embodiments of the present invention, by certainly
Exhibition formula feature extraction, has obtained associating between associating between explicit features word and Feature Words, Feature Words and viewpoint word, viewpoint
Associating between word and viewpoint word, the viewpoint root during implicit features are commented on obtains the Feature Words matching according to correlation rule,
The sample of opining mining then can be improved, improve opining mining effect.
In addition, the auto-expanding type implicit features excavation side based on discrete independent component analysis according to the above embodiment of the present invention
Method, can also have as follows add technical characteristic:
Further, step s2 further includes: s201: carries out data cleansing to described historical review data;S202: right
Historical review data after cleaning carries out feature extraction and Feature Dimension Reduction to obtain the plurality of viewpoint word and the plurality of feature
Word.
Further, step s201 further includes: removes web page interlinkage, topic label, positional information and duplicon
Sentence;The corresponding Word message that network popular word and network abbreviation are converted into;Replace emoticon with word.
Further, described network popular word and network abbreviation are obtained by way of searching for dictionary or artificial mark
The corresponding Word message being converted into.
Further, in step s202, will occur under each theme screening out less than the word of described threshold value by given threshold,
Complete described Feature Dimension Reduction.
To achieve these goals, embodiment of the invention discloses that a kind of auto-expanding type based on discrete independent component analysis
Implicit features digging system, comprising: information acquisition module, for gathering multiple historical review information;Correlation rule generation module,
For obtaining multiple viewpoint words and multiple feature in the plurality of historical review information using discrete Independent Component Analysis
Word, and viewpoint feature association rule is formed according to the plurality of viewpoint word and the plurality of Feature Words;Implicit features excavate module,
For current view word, and the corresponding Feature Words of described current view word are obtained according to described viewpoint feature association rule.
Auto-expanding type implicit features digging system based on discrete independent component analysis according to embodiments of the present invention, by certainly
Exhibition formula feature extraction, has obtained associating between associating between explicit features word and Feature Words, Feature Words and viewpoint word, viewpoint
Associating between word and viewpoint word, the viewpoint root during implicit features are commented on obtains the Feature Words matching according to correlation rule,
The sample of opining mining then can be improved, improve opining mining effect.
In addition, the auto-expanding type implicit features based on discrete independent component analysis according to the above embodiment of the present invention excavate system
System, can also have as follows add technical characteristic:
Further, described correlation rule generation module includes: data cleansing module, for described historical review data
Carry out data cleansing;Feature extraction and Feature Dimension Reduction module, for cleaning after historical review data carry out feature extraction and
Feature Dimension Reduction is to obtain the plurality of viewpoint word and the plurality of Feature Words;Viewpoint feature association rule generation module, for root
Generate described viewpoint feature association rule according to the plurality of viewpoint word and the plurality of Feature Words.
Further, described data cleansing module is further used for: remove web page interlinkage, topic label, positional information with
And repeat clause;The corresponding Word message that network popular word and network abbreviation are converted into;Replace emoticon with word.
Further, described data cleansing module is further used for obtaining by way of searching for dictionary or artificial mark
The corresponding Word message being converted into described network popular word and network abbreviation.
Further, described feature extraction and Feature Dimension Reduction module are passed through given threshold and will be occurred under each theme being less than institute
The word stating threshold value screens out, and completes described Feature Dimension Reduction.
The additional aspect of the present invention and advantage will be set forth in part in the description, and partly will become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Brief description
The above-mentioned and/or additional aspect of the present invention and advantage will become from reference to the description to embodiment for the accompanying drawings below
Substantially and easy to understand, wherein:
Fig. 1 is the auto-expanding type implicit features method for digging based on discrete independent component analysis of one embodiment of the invention
Flow chart;
Fig. 2 is the auto-expanding type implicit features digging system based on discrete independent component analysis of one embodiment of the invention
Structured flowchart.
Specific embodiment
Embodiments of the invention are described below in detail, the example of described embodiment is shown in the drawings, wherein from start to finish
The element that same or similar label represents same or similar element or has same or like function.Below with reference to attached
The embodiment of figure description is exemplary, is only used for explaining the present invention, and is not considered as limiting the invention.
In describing the invention it is to be understood that term " " center ", " longitudinal ", " horizontal ", " on ", D score,
The orientation of instruction such as "front", "rear", "left", "right", " vertical ", " level ", " top ", " bottom ", " interior ", " outward " or position relationship are
Based on orientation shown in the drawings or position relationship, it is for only for ease of the description present invention and simplifies description, rather than instruction or dark
Show the device of indication or element must have specific orientation, with specific azimuth configuration and operation, therefore it is not intended that right
The restriction of the present invention.Additionally, term " first ", " second " are only used for describing purpose, and it is not intended that instruction or hint are relative
Importance.
In describing the invention, it should be noted that unless otherwise clearly defined and limited, term " installation ", " phase
Even ", " connection " should be interpreted broadly, for example, it may be being fixedly connected or being detachably connected, or is integrally connected;Can
To be to be mechanically connected or electrical connection;Can be to be joined directly together it is also possible to be indirectly connected to by intermediary, Ke Yishi
The connection of two element internals.For the ordinary skill in the art, above-mentioned term can be understood at this with concrete condition
Concrete meaning in invention.
With reference to explained below and accompanying drawing it will be clear that these and other aspects of embodiments of the invention.In these descriptions
In accompanying drawing, specifically disclose some particular implementation in embodiments of the invention, to represent the enforcement implementing the present invention
Some modes of the principle of example are but it is to be understood that the scope of embodiments of the invention is not limited.On the contrary, the present invention
Embodiment includes falling into all changes in the range of the spirit of attached claims and intension, modification and equivalent.
Special below in conjunction with the Description of Drawings auto-expanding type implicit expression based on discrete independent component analysis according to embodiments of the present invention
Levy method for digging and system.
Fig. 1 is the auto-expanding type implicit features method for digging based on discrete independent component analysis of one embodiment of the invention
Flow chart.
As shown in figure 1, a kind of auto-expanding type implicit features method for digging based on discrete independent component analysis, walk including following
Rapid:
S1: gather multiple historical review information.
In an example of the present invention, multiple historical review information mobile phone being commented on to prefecture can be gathered.History
Review information includes the related commentary information of the performance to mobile phone and size.For example, so-and-so mobile phone, size too big it is not easy to
Put pocket into.
S2: obtain the multiple viewpoint words and many in the plurality of historical review information using discrete Independent Component Analysis
Individual Feature Words.
Specifically, discrete independent component analysis (discrete independent component analysis, dica)
It is a kind of text model, its good mode inference ability becomes recently popular machine learning and text analyzing method.
Its advantage having drawn topic model: word in text is carried out counting the method being mapped to vector space, being capable of high-order spy
Levy space reflection to low-dimensional feature space;The statistical property of topic model ensure that it in discrete data simultaneously, on text
Good analysis ability.Semantic association between word and word can be showed in the form of probability by topic model, extremely meets
The demand that unsupervised opining mining embodies for text semantic.The embodiment of the present invention adopts dica to obtain the plurality of historical review
Multiple viewpoint words in information and multiple Feature Words.
In one embodiment of the invention, step s2 further includes:
S201: data cleansing is carried out to described historical review data.
Specifically, product review is a text having mixed polyglot form, such as network popular word, cyberspeak
Abbreviation, web page interlinkage and emoticon etc..Some data types substantially unrelated with product itself it is therefore desirable to carry out clearly to data
Wash and screen.
In one embodiment of the invention, be accomplished by clear to data is carried out to described historical review data
Wash: remove web page interlinkage, topic label, positional information and repeat clause;Network popular word and network abbreviation are converted into
Corresponding Word message;Replace emoticon with word.
In one embodiment of the invention, obtain described network flow by way of searching for dictionary or artificial mark
Row language and the network corresponding Word message that is converted into of abbreviation.
S202: feature extraction and Feature Dimension Reduction are carried out to obtain the plurality of viewpoint word to the historical review data after cleaning
With the plurality of Feature Words.Feature Words are generally noun, and viewpoint word is generally adjective or adverbial word.For example in " so-and-so mobile phone, chi
Very little too big it is not easy to put pocket into " in, Feature Words are size, and viewpoint word is that (too) is big.
In one embodiment of the invention, the word sieve less than described threshold value will be occurred by given threshold under each theme
Fall, complete described Feature Dimension Reduction.For example under the theme of mobile phone, by arranging threshold value (such as Feature Words threshold value and viewpoint word threshold
Value), the word that flat rate ground occurs can be screened, and leave out the high word of this probability (such as size, photographic head pixel, weight, standby
Time and size, weight, length etc.).
S3: viewpoint feature association rule is formed according to the plurality of viewpoint word and the plurality of Feature Words.
In an example of the present invention, for example can be by Feature Words " size " and viewpoint word " big or little " and " difficult
With/be easily put into pocket " between formed correlation rule.
S4: obtain current view word, and it is corresponding to obtain described current view word according to described viewpoint feature association rule
Feature Words.
In an example of the present invention, when for new review information " someone very like it is simply that being not easy to put mouth into
Bag ", under the theme of mobile phone, due to the correlation rule of " size ", " (too) big " and " being difficult to put into pocket " of being stored with before, then
Recessive character " size " can be obtained, that is, new review information is the commentary carrying out for " size ".
By the auto-expanding type implicit features method for digging based on discrete independent component analysis for the embodiment of the present invention, by certainly
Exhibition formula feature extraction, has obtained associating between associating between explicit features word and Feature Words, Feature Words and viewpoint word, viewpoint
Associating between word and viewpoint word, the viewpoint root during implicit features are commented on obtains the Feature Words matching according to correlation rule,
The sample of opining mining then can be improved.Averagely account for the 20%-30% of global feature by counting implicit features, opining mining is tied
Fruit has very important impact.
Additionally, embodiments of the invention are also disclosed and a kind of are dug based on the auto-expanding type implicit features of discrete independent component analysis
Pick system, excavates module 230 including information acquisition module 210, correlation rule generation module 220 and implicit features.Wherein, information
Acquisition module 210 is used for gathering multiple historical review information.Correlation rule generation module 220 is used for dividing using discrete independent element
Analysis method obtains multiple viewpoint words and multiple Feature Words in the plurality of historical review information, and according to the plurality of viewpoint word
Form viewpoint feature association rule with the plurality of Feature Words.Implicit features are excavated module 230 and are used for current view word, and according to
Described viewpoint feature association rule obtains the corresponding Feature Words of described current view word.
In one embodiment of the invention, described correlation rule generation module 220 includes data cleansing module, feature is taken out
Take and Feature Dimension Reduction module and viewpoint feature association rule generation module.Wherein, data cleansing module is used for described history is commented
Carry out data cleansing by data.Feature extraction and Feature Dimension Reduction module are taken out for carrying out feature to the historical review data after cleaning
Take with Feature Dimension Reduction to obtain the plurality of viewpoint word and the plurality of Feature Words.Viewpoint feature association rule generation module is used for
Described viewpoint feature association rule is generated according to the plurality of viewpoint word and the plurality of Feature Words.
In one embodiment of the invention, described data cleansing module is further used for: removes web page interlinkage, topic mark
Label, positional information and repetition clause;The corresponding Word message that network popular word and network abbreviation are converted into;Use word generation
For emoticon.
In one embodiment of the invention, described data cleansing module is further used for by searching for dictionary or people
The mode of work mark obtains the corresponding Word message that described network popular word and network abbreviation are converted into.
In one embodiment of the invention, each is led by described feature extraction and Feature Dimension Reduction module by given threshold
Topic is lower to be occurred screening out less than the word of described threshold value, completes described Feature Dimension Reduction.
It should be noted that the auto-expanding type implicit features based on discrete independent component analysis of the embodiment of the present invention excavate system
The specific embodiment of system and the auto-expanding type implicit features method for digging based on discrete independent component analysis of the embodiment of the present invention
Specific embodiment identical, repeat no more.
In addition, the auto-expanding type implicit features method for digging based on discrete independent component analysis of the embodiment of the present invention and system
Other constitute and effect is all known for a person skilled in the art, in order to reduce redundancy, do not repeat.
In the description of this specification, reference term " embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or the spy describing with reference to this embodiment or example
Point is contained at least one embodiment or the example of the present invention.In this manual, to the schematic representation of above-mentioned term not
Necessarily refer to identical embodiment or example.And, the specific features of description, structure, material or feature can be any
One or more embodiments or example in combine in an appropriate manner.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that: not
Multiple changes, modification, replacement and modification can be carried out to these embodiments in the case of the principle of the disengaging present invention and objective, this
The scope of invention by claim and its is equal to limit.
Claims (10)
1. a kind of auto-expanding type implicit features method for digging based on discrete independent component analysis is it is characterised in that include following walking
Rapid:
S1: gather multiple historical review information;
S2: obtain multiple viewpoint words and the multiple spy in the plurality of historical review information using discrete Independent Component Analysis
Levy word;
S3: viewpoint feature association rule is formed according to the plurality of viewpoint word and the plurality of Feature Words;
S4: obtain current view word, and the corresponding feature of described current view word is obtained according to described viewpoint feature association rule
Word.
2. the auto-expanding type implicit features method for digging based on discrete independent component analysis according to claim 1, its feature
It is, step s2 further includes:
S201: data cleansing is carried out to described historical review data;
S202: feature extraction and Feature Dimension Reduction are carried out to obtain the plurality of viewpoint word and institute to the historical review data after cleaning
State multiple Feature Words.
3. the auto-expanding type implicit features method for digging based on discrete independent component analysis according to claim 2, its feature
It is, step s201 further includes:
Remove web page interlinkage, topic label, positional information and repeat clause;
The corresponding Word message that network popular word and network abbreviation are converted into;
Replace emoticon with word.
4. the auto-expanding type implicit features method for digging based on discrete independent component analysis according to claim 3, its feature
It is, obtain the correspondence that described network popular word and network abbreviation are converted into by way of searching for dictionary or artificial mark
Word message.
5. the auto-expanding type implicit features method for digging based on discrete independent component analysis according to claim 2, its feature
It is, in step s202, will occur under each theme screening out less than the word of described threshold value by given threshold, complete described feature
Dimensionality reduction.
6. a kind of auto-expanding type implicit features digging system based on discrete independent component analysis is it is characterised in that include:
Information acquisition module, for gathering multiple historical review information;
Correlation rule generation module, for being obtained in the plurality of historical review information using discrete Independent Component Analysis
Multiple viewpoint words and multiple Feature Words, and viewpoint feature association rule are formed according to the plurality of viewpoint word and the plurality of Feature Words
Then;
Implicit features excavate module, for current view word, and obtain described current sight according to described viewpoint feature association rule
The point corresponding Feature Words of word.
7. the auto-expanding type implicit features digging system based on discrete independent component analysis according to claim 6, its feature
It is, described correlation rule generation module includes:
Data cleansing module, for carrying out data cleansing to described historical review data;
Feature extraction and Feature Dimension Reduction module, for cleaning after historical review data carry out feature extraction and Feature Dimension Reduction with
Obtain the plurality of viewpoint word and the plurality of Feature Words;
Viewpoint feature association rule generation module, for generating described sight according to the plurality of viewpoint word and the plurality of Feature Words
Point feature association rule.
8. the auto-expanding type implicit features digging system based on discrete independent component analysis according to claim 7, its feature
It is, described data cleansing module is further used for:
Remove web page interlinkage, topic label, positional information and repeat clause;
The corresponding Word message that network popular word and network abbreviation are converted into;
Replace emoticon with word.
9. the auto-expanding type implicit features digging system based on discrete independent component analysis according to claim 6, its feature
It is, described data cleansing module is further used for obtaining described network flow by way of searching for dictionary or artificial mark
Row language and the network corresponding Word message that is converted into of abbreviation.
10. the auto-expanding type implicit features digging system based on discrete independent component analysis according to claim 6, its feature
It is, described feature extraction and Feature Dimension Reduction module are passed through given threshold and will the word sieve less than described threshold value under each theme
Fall, complete described Feature Dimension Reduction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610677146.9A CN106354754A (en) | 2016-08-16 | 2016-08-16 | Bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610677146.9A CN106354754A (en) | 2016-08-16 | 2016-08-16 | Bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106354754A true CN106354754A (en) | 2017-01-25 |
Family
ID=57844920
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610677146.9A Pending CN106354754A (en) | 2016-08-16 | 2016-08-16 | Bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106354754A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109299460A (en) * | 2018-09-18 | 2019-02-01 | 北京三快在线科技有限公司 | Analyze method, apparatus, electronic equipment and the storage medium of the evaluation data in shop |
CN110020439A (en) * | 2019-04-16 | 2019-07-16 | 中森云链(成都)科技有限责任公司 | A kind of multi-field text implicit features abstracting method based on hiding related network |
CN112559672A (en) * | 2021-02-22 | 2021-03-26 | 深圳市优讯通信息技术有限公司 | Information detection method, electronic device and computer storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102945268A (en) * | 2012-10-25 | 2013-02-27 | 北京腾逸科技发展有限公司 | Method and system for excavating comments on characteristics of product |
CN103399916A (en) * | 2013-07-31 | 2013-11-20 | 清华大学 | Internet comment and opinion mining method and system on basis of product features |
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 |
CN105573983A (en) * | 2015-12-17 | 2016-05-11 | 清华大学 | Topic model based hierarchical classification method and system for microblog user emotions |
-
2016
- 2016-08-16 CN CN201610677146.9A patent/CN106354754A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102945268A (en) * | 2012-10-25 | 2013-02-27 | 北京腾逸科技发展有限公司 | Method and system for excavating comments on characteristics of product |
CN103399916A (en) * | 2013-07-31 | 2013-11-20 | 清华大学 | Internet comment and opinion mining method and system on basis of product features |
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 |
CN105573983A (en) * | 2015-12-17 | 2016-05-11 | 清华大学 | Topic model based hierarchical classification method and system for microblog user emotions |
Non-Patent Citations (3)
Title |
---|
LINGWEI ZENG.ETC: "A classification-based approach for implicit feature identification", 《CONFERENCE PROCEEDINGS NLP-NABD 2013,CCL 2013》 * |
SOUJANYA PORIA.ETC: "A Rule-Based Approach to Aspect Extraction from Product Reviews", 《PROCEEDINGS OF THE SECOND WORKSHOP ON NATURAL LANGUAGE PROCESSING FOR SOCIAL MEDIA (SOCIALNLP)》 * |
ZHEN HAI.ETC: "Implicit Feature Identification via Co-occurrence Association Rule Mining", 《ZHEN H, CHANG K, KIM J. IMPLICIT FEATURE IDENTIFICATION VIA CO-OCCURRENCE ASSOCIATION RULE MINING[C]// INTERNATIONAL CONFERENCE ON COMPUTATIONAL LINGUISTICS & INTELLIGENT TEXT PROCESSING》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109299460A (en) * | 2018-09-18 | 2019-02-01 | 北京三快在线科技有限公司 | Analyze method, apparatus, electronic equipment and the storage medium of the evaluation data in shop |
CN109299460B (en) * | 2018-09-18 | 2022-07-12 | 北京三快在线科技有限公司 | Method and device for analyzing evaluation data of shop, electronic device and storage medium |
CN110020439A (en) * | 2019-04-16 | 2019-07-16 | 中森云链(成都)科技有限责任公司 | A kind of multi-field text implicit features abstracting method based on hiding related network |
CN111859898A (en) * | 2019-04-16 | 2020-10-30 | 中森云链(成都)科技有限责任公司 | Hidden associated network-based multi-field text implicit feature extraction method and computer storage medium |
CN111859898B (en) * | 2019-04-16 | 2024-01-16 | 中森云链(成都)科技有限责任公司 | Hidden association network-based multi-domain text implicit feature extraction method and computer storage medium |
CN112559672A (en) * | 2021-02-22 | 2021-03-26 | 深圳市优讯通信息技术有限公司 | Information detection method, electronic device and computer storage medium |
CN112559672B (en) * | 2021-02-22 | 2021-07-13 | 深圳市优讯通信息技术有限公司 | Information detection method, electronic device and computer storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110781317B (en) | Method and device for constructing event map and electronic equipment | |
CN107153713A (en) | Overlapping community detection method and system based on similitude between node in social networks | |
CN107590134A (en) | Text sentiment classification method, storage medium and computer | |
CN104699766A (en) | Implicit attribute mining method integrating word correlation and context deduction | |
CN108009228A (en) | A kind of method to set up of content tab, device and storage medium | |
CN105512687A (en) | Emotion classification model training and textual emotion polarity analysis method and system | |
CN107133214A (en) | A kind of product demand preference profiles based on comment information are excavated and its method for evaluating quality | |
CN107578292B (en) | User portrait construction system | |
CN104268160A (en) | Evaluation object extraction method based on domain dictionary and semantic roles | |
CN107992481A (en) | A kind of matching regular expressions method, apparatus and system based on multiway tree | |
CN105243129A (en) | Commodity property characteristic word clustering method | |
CN102033880A (en) | Marking method and device based on structured data acquisition | |
CN107038229A (en) | A kind of use-case extracting method based on natural semantic analysis | |
CN107463658A (en) | File classification method and device | |
CN108062304A (en) | A kind of sentiment analysis method of the comment on commodity data based on machine learning | |
CN102298638A (en) | Method and system for extracting news webpage contents by clustering webpage labels | |
CN101980199A (en) | Method and system for discovering network hot topic based on situation assessment | |
CN105654144B (en) | A kind of social network ontologies construction method based on machine learning | |
CN106354754A (en) | Bootstrap-type implicit characteristic mining method and system based on dispersed independent component analysis | |
CN102053974B (en) | Chinese character input method and device | |
CN105893582A (en) | Social network user emotion distinguishing method | |
CN110413787A (en) | Text Clustering Method, device, terminal and storage medium | |
CN106202034B (en) | A kind of adjective word sense disambiguation method and device based on interdependent constraint and knowledge | |
CN101404033A (en) | Automatic generation method and system for noumenon hierarchical structure | |
CN107092605A (en) | A kind of entity link method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170125 |
|
RJ01 | Rejection of invention patent application after publication |