CN110457472A - The emotion association analysis method for electric business product review based on SOM clustering algorithm - Google Patents
The emotion association analysis method for electric business product review based on SOM clustering algorithm Download PDFInfo
- Publication number
- CN110457472A CN110457472A CN201910639598.1A CN201910639598A CN110457472A CN 110457472 A CN110457472 A CN 110457472A CN 201910639598 A CN201910639598 A CN 201910639598A CN 110457472 A CN110457472 A CN 110457472A
- Authority
- CN
- China
- Prior art keywords
- product
- comment
- vocabulary
- data
- electric business
- 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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/374—Thesaurus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0282—Rating or review of business operators or products
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Computational Linguistics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of emotion association analysis method for electric business product review based on SOM clustering algorithm, comprising: the 1) keywords database of the product scope is constructed in analysis；2) the user comment data in product subdivision field are covered in acquisition, establish original comment text sample data；3) urtext is pre-processed；4) building that the original comment of m item is carried out to term vector according to bag of words again, is then 1 there are the word, there is no be then 00；5) term vector being converted to is commented on every using SOM (unsupervised Self-organizing Maps) algorithm carry out clustering；6) two kinds of vocabulary association is obtained by the analysis of analogous relationship word in cluster and the interpretation of similar sample, generates two kinds of association conclusion；7) the training dataset model generated by general comment sample obtains the information of user's portrait.By the method for this invention to the designing points of energy guide product iteration upgrading.The invention is analyzed for the product of multiple fields.
Description
Technical field
The invention belongs to sentiment analysis technical fields, are directed to electric business product based on SOM clustering algorithm more particularly to a kind of
The emotion association analysis method of comment.
Background technique
Such as development of day cat, Jingdone district electric business platform, this channel of online sale cover nearly all type in recent years
Product, and user is attracted to leave use feeling and subjective assessment to the product.The content of text such as these comments are for product
Developing design side has very high reference value, forms the corpus of text data of sufficient amount.
Text mining under the support of natural language processing (NLP) simultaneously is a kind of unstructured or semi-structured, weak to measuring
Potential, the data pattern of possibility, internal association, the rule of development or effective trend etc. are obtained in the text of structuring, and
Extraction goal in research is available, it is by intuitivism apprehension to be easy, the higher value content and knowledge, such as user's sense of distribution in the text
The content of interest, the prediction etc. of event development, to summarize as visual data model and rule to determine as auxiliary
The technical method of plan information[1]。
About the application of the Text Mining Technology in product design field, such as by utilizing magnanimity social media data,
Crossing research is carried out with each function focus of product, to obtain user to the focus of product, is ground with corresponding design
Hair[2]；Also have external related scholar simultaneously by text mining, by the Sentiment orientation of user and with the potential demand of Related product
It is attached excavation.In conjunction with Kano model, the products characteristics orientation analysis of user's negative emotion tendency is selected to determine potential demand,
And propose that the attribute of demand can be changed from glamour attribute to essential attribute by the optimization to potential demand[3]；And
The researcher of Northwestern Univ USA excavates comment of the user to each automobile brand, model by acquisition, it is established that automobile product
Multi-level incidence relation between board, model, prediction has separated the selection of most suitable user from the choice set of bulky complex
Collection[4]。
Emotion lexicon dictionary is relied solely in art methods, comment is subjected to text vector, and carry out association cluster
Analysis, the association obtained between the different levels of emotion influence, and eventually find the comment with valuable emotion vocabulary and progress
Specific comment and analysis finds most valuable comment in magnanimity comment, excavates the hiding demand of the user and to the production
Product carry out more targeted design and change[5]。
For existing technology when being analyzed for the stronger product of functional requirements, there are the following problems:
1) to the stronger product of functional requirements, the description of Kansei word may be more weak and about product function module
Corpus information could not be treated and be analyzed well, cause certain valuable information that may be ignored.
2) interference in terms of the association between emotion lexical hierarchy is easy the weight mixed there are a variety of emotions, therefore independent
On result accuracy, there may be influences when the valuable comment that analysis is tracked；
Bibliography:
[1]Talib R,Hanif M.K,Ayesha S,Fatimal F.Text Mining:Techniques,
Applications and Issues[J].International Journal of Advanced Computer Science
and Applications,2016,7(11):414–18.
[2]Tuarob,Suppawong Tucker,Conrad S.Automated Discovery of Lead Users
and Latent Product Features by Mining Large Scale Social Media Networks[J]
.Journal of mechanical design,2015,137(7):071402-1-11.
[3]Zhou Fengjiao,Roger Jianxin,Julie S,et al.Latent Customer Needs
Elicitation by Use Case Analogical Reasoning from Sentiment Analysis of
Online Product Reviews[J].Journal of mechanical design,2015,137(7):071401-1-
12.
[4]Wu X,Zhu X,WU G Q,Ding W.A Data-Driven Network Analysis Approach
to Predicting Customer Choice Sets for Choice Modeling in Engineering Design
[J].Journal of Mechanical Design,2015,137(7):071410-1-11.
[5]Wonjoon Kima,Taehoon Kob,et al.Mining affective experience for a
kansei design study on a recliner[J].Applied Ergonomics,74(2019):145-153.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of to be directed to electric business based on SOM clustering algorithm
The emotion association analysis method of product review, includes the following steps:
1) data are acquired or are crawled, and the user for obtaining the similar product of one or more functional structures uses comment；
2) the comment sample crawled is carried out data cleansing, removes and analyze incoherent text data by data prediction；
3) SOM clustering algorithm is executed；
4) the comment association analysis based on vocabulary；
5) conclusion exports subsequent applications.
The step 1) is specific as follows:
(1) it delimit and chooses that be analyzed a product or functional module structure be consistent or the product of similar a few moneys；
(2) webpage http protocol is utilized, obtains the comment in product electric business website using the crawler strategy based on Python,
By the data of crawl in the composition of sample deposit local data base of " flag code-comment ".
The step 2) is specific as follows:
(1) tentatively understand the related service of product, the incoherent comment content of others being likely to occur in analysis comment；
(2) stop words vocabulary is established, carries out text data comprising the symbol etc. that cannot be handled, and by compiling a little scripts
Cleaning；
(3) word cutting and word frequency analysis are carried out, constantly increases adjustment stop words vocabulary, improves the cleaning of text data；
(4) data after over cleaning obtain the vocabulary and function of the higher product usage experience of word frequency by word frequency analysis
Can, the vocabulary of component or module, choose wherein high word frequency and the vocabulary that has break-up value establish the dictionaries of bag of words.
The step 3) is specific as follows:
(1) comment text vectorization；
(2) input matrix of developing algorithm；
(3) SOM algorithm is executed.
The step 4) is specific as follows:
(1) division clustered on sample projection point diagram, obtains different clusters；
(2) each cluster is analyzed respectively, is found in cluster specifically comment and associated identical vocabulary and is compared intersection point
Analysis；
(3) association type is judged according to the conjunctive word of cluster comment, a point situation is analyzed；
(4) specific subsequent analysis is made according to type.
The step 5) is specific as follows: in the laggard row labelization processing of Clustering Model for obtaining training dataset, Ke Yiju
This carries out the projection classification of more text comments samples；Design of feedback content is jointly to user in conjunction with obtained in previous step
Portrait carries out description and definition more profound, and completes the feedback of product design iteration jointly.
In view of the analysis method being applied to text mining in perceptual engineering for being directed to product design field at present, the present invention
The association cluster algorithm with machine learning is wanted to, by emotion vocabulary in the product review of user and specific product function mould
Block is associated together, and is obtained product specific functional modules that mass data supports, more objective and is given user's bring emotion
Experience.
In addition, the present invention also wants to rely on existing Internet resources and natural language processing means, towards wider
General product scope is based on different product ontologies, targetedly builds the dictionary of different field, and utilize term vector
And machine.
The utility model has the advantages that
The present invention rapidly can divide a large amount of true user experience by way of NLP natural language processing
Analysis establishes targeted keyword dictionary for the product of different field, and establishes about the emotion of the type product and hard
Part experience between association or different levels between emotion vocabulary association and be incident upon in elements of product design.This aspect can
Using the research step as perceptual engineering method early period, extracts Kansei word and cooperate and establish between product form and emotion vocabulary
The part of correlation degree.The conclusion with general character is obtained in specific design point.It is fed back as design evaluatio and guides the type
The design of product.On the other hand it can become a macroscopical neural network model of performance appraisal of the type product, and be directed to newly
Special group user carries out projection classification, obtains understanding the more accurate of special group user rapidly.
Both of which more accurately can obtain rapidly accurate data from a large amount of comment samples, and be obtained by analysis
The feedback of true Emotion Design, eliminate traditional method of investigation and study to time drain on manpower and material resources, and can have very high
Accuracy.
Can be using the Relational Evaluation index and its collocation of different term vector algorithms and setting different modes, it can
Close to sentiment analysis accuracy of the invention.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Fig. 2 is composition of sample deposit local data base structure chart.
Fig. 3 is bag of words figure.
Fig. 4 is clustering algorithm method process.
Fig. 5 is clustering algorithm result diagram form.
Fig. 6 is the process flow diagram flow chart of association analysis.
The step of Fig. 7 is conclusion output subsequent applications.
Specific embodiment
Below by specific embodiments and the drawings, the present invention is further illustrated.The embodiment of the present invention is in order to more
So that those skilled in the art is more fully understood the present invention well, any limitation is not made to the present invention.
To solve the above-mentioned problems, the present invention provides a kind of feelings for electric business product review based on SOM clustering algorithm
Feel association analysis method, as shown in Figure 1, comprising the following steps:
Step 1: data are acquired or are crawled, and the user for obtaining the similar product of one or more functional structures uses comment.
1) it delimit and chooses that be analyzed a product or functional module structure be consistent or the product of similar a few moneys,
2) webpage http protocol is utilized, obtains the comment in product electric business website using the crawler strategy based on Python,
The data of crawl are stored in local data base with the composition of sample of " flag code-comment ", as shown in Figure 2.
Step 2: the comment sample crawled is carried out data cleansing, removes and analyze incoherent text by data prediction
Data.
When carrying out sentiment analysis to product scope, what product review included is but more than the actual usage experience of product,
It may can also include other related services of product, such as logistics information, the actual color difference of product, the subsidiary small gift of purchase product
Product and after-sale service etc., therefore other interference for servicing bring emotions feedback should be avoided when being analyzed for product
Factor；Some characters or emoticon that cannot be identified may be attached in comment simultaneously, have been needed by certain program language
(such as Python) writes script and is cleaned or substituted the text data for being organized into and more standardizing:
1) tentatively understand the related service of product, the incoherent comment content of others being likely to occur in analysis comment.
2) stop words vocabulary is established, carries out text data comprising the symbol etc. that cannot be handled, and by compiling a little scripts
Cleaning.
3) word cutting and word frequency analysis are carried out, constantly increases adjustment stop words vocabulary, improves the cleaning of text data.
4) data after over cleaning obtain the vocabulary and function of the higher product usage experience of word frequency by word frequency analysis
Can, the vocabulary of component or module, choose wherein high word frequency and the vocabulary that has break-up value establish the dictionaries of bag of words.
After tested, by aforementioned four step, it can be basically completed the task of data cleansing, and establish the word of bag of words
Allusion quotation, to carry out the processing on basis for subsequent text analyzing.It is as shown in Figure 3:
Step 3: executing SOM clustering algorithm.
According to bag of words obtained in previous step, each text comments are carried out with the vectorization of text, specifically
The method of definition are as follows: each of bag of words word is corresponded into a dimension, vocabulary representated by each dimension is commented in text
Occur one or many as 1 in, do not occur being then 0, then each is commented on to the vector converted for a hyperspace,
A sample data is also become, i.e., the vector of all comments is collectively formed into multi-dimensional matrix n*m, n a row and represents n dimension
Vocabulary, m represents m item and comments on the sample data that is converted to.
The focusing solutions analysis that the matrix inputs and carries out neural network SOM as training dataset (is occurred identical
The more comment similitude of vocabulary is bigger, and European space distance is smaller), which can throw the similitude of the sample of multidimensional data
The division and analysis penetrated in a two-dimensional plane, intuitively understand the similitude between sample, and clustered.It can be used and be based on
The kit of MATLAB more mature neural network clustering carries out this operation, and available SOM sample projects point diagram, individually
Characteristic attribute vocabulary weight distribution figure and SOM neighborhood distance map.
Method process is as shown in figure 4, obtained result diagram form (example diagram) is as shown in Figure 5.
Step 4: the comment association analysis based on vocabulary.
1. the division clustered on sample projection point diagram, obtains different clusters.
2. analyze each cluster respectively, find in cluster that specifically comment and associated identical vocabulary compare intersection point
Analysis, the cluster the big more intensively to illustrate that correlation degree is bigger.
3. judging association type according to the conjunctive word of cluster comment, a point situation is analyzed,
If being associated with vocabulary of the vocabulary between emotion vocabulary and product function either emotion vocabulary and hardware aspect such as
Relationship between components module then establishes user and the emotion of Product Experience is associated with the perception of concrete function or module；
The relationship of emotion between different levels is then usually embodied if it is the cluster between emotion vocabulary and emotion vocabulary, is built
The primary association of specific Kansei word and macroscopical Kansei word is found；
4. making specific subsequent analysis according to type, the perception of the type product then can establish if it is the first situation
The relationship of vocabulary and design element, can studying as important reference frame for perceptual engineering.Can also directly it divide simultaneously
Comment sample in analysis cluster, selective analysis directly obtain the defect or advantage feedback in design；If it is second situation
The then available specific emotional experience for influencing emotion forward direction or negative sense, and targetedly certain emotional experiences are divided
Class and make a concrete analysis of comment sample be incident upon in specific product design.
The process of association analysis is as shown in Figure 6:
Step 5: conclusion exports subsequent applications.
In the laggard row labelization processing of Clustering Model for obtaining training dataset, more text comments samples can be carried out accordingly
This projection classification, as less investment measurer have same background information use comment (use of such as a certain amount of old man or children are commented
By), the focus of such available user.The design of feedback content in conjunction with obtained in previous step jointly draws a portrait to user
Description and definition more profound are carried out, and completes the feedback of product design iteration jointly.
Specific implementation step is as shown in Figure 7.
A kind of the technology of the present invention key point: emotion association analysis for electric business product review based on SOM clustering algorithm
Method, including the following steps: 1) acquisition cover the user comment data in product subdivision field, establishes original comment text
Sample data；2) urtext is pre-processed, including conversion between simplified and traditional Chinese, unusual character are deleted, segmented, removing stop words etc., and
Speech habits based on product ontology and certain product review, analysis are constructed the keywords database of the product scope, are picked out
The emotion vocabulary and product component of high word frequency or the vocabulary of function, establish n dimension bag of words, again by the original comment of m item according to
Bag of words carry out the building of term vector, are then 1 there are the word, there is no be then 00；3) using SOM, (unsupervised self-organizing is reflected
Penetrate) algorithm comments on the term vector that is converted to every and carries out clustering.4) by the analysis of analogous relationship word in cluster and similar
The interpretation of sample obtains two kinds of vocabulary association, generates two kinds of association conclusion, can more accurately obtain emotion
Vocabulary is associated with product module or function to the perception of user, building user's portrait；5) design of feedback of closed loop is formed.By total
Body comments on the training dataset model that sample generates, and can analyze the product of special group using comment or in-depth interview, comes
The information of the users such as use habit, scene, focus to special group portrait.
Described the prefered embodiments of the present invention in detail above in conjunction with attached drawing, for example, applied by programming language and soft
Part platform.But the present invention is not limited to the specific details in the above embodiment, within the scope of the technical concept of the present invention,
Can be with various simple variants of the technical solution of the present invention are made, these simple variants all belong to the scope of protection of the present invention.This
Any combination can be carried out between a variety of different embodiments of invention, it is as long as it does not violate the idea of the present invention, same
It should be considered as content disclosed in this invention.
Claims (6)
1. the emotion association analysis method for electric business product review based on SOM clustering algorithm, which is characterized in that including as follows
Step:
1) data are acquired or are crawled, and the user for obtaining the similar product of one or more functional structures uses comment；
2) the comment sample crawled is carried out data cleansing, removes and analyze incoherent text data by data prediction；
3) SOM clustering algorithm is executed；
4) the comment association analysis based on vocabulary；
5) conclusion exports subsequent applications.
2. the emotion association analysis method for electric business product review according to claim 1 based on SOM clustering algorithm,
It is characterized in that, the step 1) is specific as follows:
(1) it delimit and chooses that be analyzed a product or functional module structure be consistent or the product of similar a few moneys；
(2) webpage http protocol is utilized, the comment in product electric business website is obtained using the crawler strategy based on Python, will grab
The data taken are in the composition of sample deposit local data base of " flag code-comment ".
3. the emotion association analysis method for electric business product review according to claim 1 based on SOM clustering algorithm,
It is characterized in that, the step 2) is specific as follows:
(1) tentatively understand the related service of product, the incoherent comment content of others being likely to occur in analysis comment；
(2) stop words vocabulary is established, carries out the clear of text data comprising the symbol etc. that cannot be handled, and by compiling a little scripts
It washes；
(3) word cutting and word frequency analysis are carried out, constantly increases adjustment stop words vocabulary, improves the cleaning of text data；
(4) data after over cleaning obtain the vocabulary and function of the higher product usage experience of word frequency by word frequency analysis,
The vocabulary of component or module, choose wherein high word frequency and the vocabulary that has break-up value establish the dictionaries of bag of words.
4. the emotion association analysis method for electric business product review according to claim 1 based on SOM clustering algorithm,
It is characterized in that, the step 3) is specific as follows:
(1) comment text vectorization；
(2) input matrix of developing algorithm；
(3) SOM algorithm is executed.
5. the emotion association analysis method for electric business product review according to claim 1 based on SOM clustering algorithm,
It is characterized in that, the step 4) is specific as follows:
(1) division clustered on sample projection point diagram, obtains different clusters；
(2) each cluster is analyzed respectively, is found in cluster specifically comment and associated identical vocabulary and is compared alternate analysis；
(3) association type is judged according to the conjunctive word of cluster comment, a point situation is analyzed；
(4) specific subsequent analysis is made according to type.
6. the emotion association analysis method for electric business product review according to claim 1 based on SOM clustering algorithm,
It is characterized in that, the step 5) is specific as follows:, can be in the laggard row labelization processing of the Clustering Model that obtains training dataset
The projection classification of more text comments samples is carried out accordingly；The design of feedback content in conjunction with obtained in previous step jointly to
Family portrait carries out description and definition more profound, and completes the feedback of product design iteration jointly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910639598.1A CN110457472A (en) | 2019-07-16 | 2019-07-16 | The emotion association analysis method for electric business product review based on SOM clustering algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910639598.1A CN110457472A (en) | 2019-07-16 | 2019-07-16 | The emotion association analysis method for electric business product review based on SOM clustering algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110457472A true CN110457472A (en) | 2019-11-15 |
Family
ID=68481709
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910639598.1A Pending CN110457472A (en) | 2019-07-16 | 2019-07-16 | The emotion association analysis method for electric business product review based on SOM clustering algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110457472A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113792544A (en) * | 2021-07-06 | 2021-12-14 | 中国地质大学（武汉） | Text emotion classification method and device considering geographic spatial distribution |
| CN114970446A (en) * | 2022-07-14 | 2022-08-30 | 深圳前海环融联易信息科技服务有限公司 | Text conversion display method and device, equipment, medium and product thereof |
| CN117131427A (en) * | 2023-10-26 | 2023-11-28 | 戎行技术有限公司 | Social platform account data correlation analysis method based on multiple NLP |
| CN117455555A (en) * | 2023-12-25 | 2024-01-26 | 厦门理工学院 | Big data-based electric business portrait analysis method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156184A (en) * | 2015-04-21 | 2016-11-23 | 苏州优估营网络科技有限公司 | The expert's comment inductive algorithm clustered based on emotional semantic classification and SOM |
| CN107704558A (en) * | 2017-09-28 | 2018-02-16 | 北京车慧互动广告有限公司 | A kind of consumers' opinions abstracting method and system |
| CN108062304A (en) * | 2017-12-19 | 2018-05-22 | 北京工业大学 | A kind of sentiment analysis method of the comment on commodity data based on machine learning |
| CN108319734A (en) * | 2018-04-11 | 2018-07-24 | 中国计量大学 | A kind of product feature structure tree method for auto constructing based on linear combiner |
| CN108446813A (en) * | 2017-12-19 | 2018-08-24 | 清华大学 | A kind of method of electric business service quality overall merit |
| CN108491377A (en) * | 2018-03-06 | 2018-09-04 | 中国计量大学 | A kind of electric business product comprehensive score method based on multi-dimension information fusion |
| CN109101478A (en) * | 2018-06-04 | 2018-12-28 | 东南大学 | A kind of Aspect grade sentiment analysis method towards electric business comment text |
-
2019
- 2019-07-16 CN CN201910639598.1A patent/CN110457472A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156184A (en) * | 2015-04-21 | 2016-11-23 | 苏州优估营网络科技有限公司 | The expert's comment inductive algorithm clustered based on emotional semantic classification and SOM |
| CN107704558A (en) * | 2017-09-28 | 2018-02-16 | 北京车慧互动广告有限公司 | A kind of consumers' opinions abstracting method and system |
| CN108062304A (en) * | 2017-12-19 | 2018-05-22 | 北京工业大学 | A kind of sentiment analysis method of the comment on commodity data based on machine learning |
| CN108446813A (en) * | 2017-12-19 | 2018-08-24 | 清华大学 | A kind of method of electric business service quality overall merit |
| CN108491377A (en) * | 2018-03-06 | 2018-09-04 | 中国计量大学 | A kind of electric business product comprehensive score method based on multi-dimension information fusion |
| CN108319734A (en) * | 2018-04-11 | 2018-07-24 | 中国计量大学 | A kind of product feature structure tree method for auto constructing based on linear combiner |
| CN109101478A (en) * | 2018-06-04 | 2018-12-28 | 东南大学 | A kind of Aspect grade sentiment analysis method towards electric business comment text |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113792544A (en) * | 2021-07-06 | 2021-12-14 | 中国地质大学（武汉） | Text emotion classification method and device considering geographic spatial distribution |
| CN113792544B (en) * | 2021-07-06 | 2023-08-29 | 中国地质大学（武汉） | Text emotion classification method and device considering geospatial distribution |
| CN114970446A (en) * | 2022-07-14 | 2022-08-30 | 深圳前海环融联易信息科技服务有限公司 | Text conversion display method and device, equipment, medium and product thereof |
| CN117131427A (en) * | 2023-10-26 | 2023-11-28 | 戎行技术有限公司 | Social platform account data correlation analysis method based on multiple NLP |
| CN117131427B (en) * | 2023-10-26 | 2024-01-12 | 戎行技术有限公司 | Social platform account data correlation analysis method based on multiple NLP |
| CN117455555A (en) * | 2023-12-25 | 2024-01-26 | 厦门理工学院 | Big data-based electric business portrait analysis method and system |
| CN117455555B (en) * | 2023-12-25 | 2024-03-08 | 厦门理工学院 | Big data-based electric business portrait analysis method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kumar et al. | Aspect-based sentiment analysis using deep networks and stochastic optimization | |
| Wesslen | Computer-assisted text analysis for social science: Topic models and beyond | |
| CN110457472A (en) | The emotion association analysis method for electric business product review based on SOM clustering algorithm | |
| CN102123172B (en) | Implementation method of Web service discovery based on neural network clustering optimization | |
| Lakshminarayana et al. | Automatic generation and optimization of test case using hybrid cuckoo search and bee colony algorithm | |
| Sevastjanova et al. | Going beyond visualization: Verbalization as complementary medium to explain machine learning models | |
| KR20200007713A (en) | Method and Apparatus for determining a topic based on sentiment analysis | |
| Sridevi et al. | AI based suitability measurement and prediction between job description and job seeker profiles | |
| Mehndiratta et al. | Identification of sarcasm using word embeddings and hyperparameters tuning | |
| Jiang et al. | A constituency parsing tree based method for relation extraction from abstracts of scholarly publications | |
| Dulău et al. | Cryptocurrency–sentiment analysis in social media | |
| Singh et al. | Towards improving e-commerce customer review analysis for sentiment detection | |
| Fares et al. | Difficulties and improvements to graph-based lexical sentiment analysis using lisa | |
| Upreti | Convolutional neural network (cnn). a comprehensive overview | |
| Cong et al. | A small sample data-driven method: User needs elicitation from online reviews in new product iteration | |
| Babayoff et al. | The role of semantics in the success of crowdfunding projects | |
| Rawat et al. | Topic modelling of legal documents using NLP and bidirectional encoder representations from transformers | |
| CN110069686A (en) | User behavior analysis method, apparatus, computer installation and storage medium | |
| Kim et al. | Categorizing affective response of customer with novel explainable clustering algorithm: The case study of Amazon reviews | |
| Saeed et al. | A framework to predict early news popularity using deep temporal propagation patterns | |
| CN116450827A (en) | Event template induction method and system based on large-scale language model | |
| Haddad et al. | Toward a prediction approach based on deep learning in Big Data analytics | |
| Ruiz-Navas et al. | Developing a framework to track knowledge convergence in'big data' | |
| Zhao et al. | Missing RDF triples detection and correction in knowledge graphs | |
| Kamel et al. | Robust sentiment fusion on distribution of news |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191115 |