CN109034893A - A kind of tourist net comment sentiment analysis and QoS evaluating method - Google Patents
A kind of tourist net comment sentiment analysis and QoS evaluating method Download PDFInfo
- Publication number
- CN109034893A CN109034893A CN201810803792.4A CN201810803792A CN109034893A CN 109034893 A CN109034893 A CN 109034893A CN 201810803792 A CN201810803792 A CN 201810803792A CN 109034893 A CN109034893 A CN 109034893A
- Authority
- CN
- China
- Prior art keywords
- comment
- tourist
- model
- tourism
- net
- 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
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000004458 analytical method Methods 0.000 title claims abstract description 29
- 230000008451 emotion Effects 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 34
- 239000013598 vector Substances 0.000 claims abstract description 19
- 238000011156 evaluation Methods 0.000 claims abstract description 17
- 230000002349 favourable effect Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000001747 exhibiting effect Effects 0.000 claims description 3
- 238000012958 reprocessing Methods 0.000 claims description 3
- 238000012552 review Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 abstract description 10
- 230000006872 improvement Effects 0.000 abstract description 5
- 230000008569 process Effects 0.000 abstract description 5
- 230000015572 biosynthetic process Effects 0.000 abstract description 3
- 238000012706 support-vector machine Methods 0.000 abstract description 3
- 238000003786 synthesis reaction Methods 0.000 abstract description 3
- 230000008901 benefit Effects 0.000 description 6
- 238000013441 quality evaluation Methods 0.000 description 3
- 241000283074 Equus asinus Species 0.000 description 2
- 241000256856 Vespidae Species 0.000 description 2
- 239000002304 perfume Substances 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002996 emotional effect Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/14—Travel agencies
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Finance (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Accounting & Taxation (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Probability & Statistics with Applications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of tourist net comment sentiment analysis and QoS evaluating method, sentiment analysis is the following steps are included: pre-process tourist net evaluation；Construct emotion trend training set；Logic Regression Models, supporting vector machine model and model-naive Bayesian is respectively trained using emotion trend training set, exports the training result G of three kinds of propaedeutics modelsm(x)；Ballot device model is established, training result is handled, the Sentiment orientation of comment is calculated；Input ballot device model obtains the Sentiment orientation of every comment after handling tourist net comment data to be tested.Pass through the synthesis of a variety of machine learning algorithm models, three kinds of integrated logic recurrence, support vector machines and Bayes sorting algorithm models, overcome the efficiency and accuracy problem of single algorithm model, promote the accuracy of tourism comment sentiment analysis, it selects service delivery in tourism to help for tourist, provides decision support for the evaluation of tourism authorities and improvement region service quality in tourism.
Description
Technical field
The present invention relates to computer digital animations and analysis technical field, and in particular to a kind of tourist net comment emotion point
Analysis and QoS evaluating method.
Background technique
With the rise and development of Tourism E-commerce, more and more tourists buy tourist service by internet, and
Commented on after receiving tourist service, deliver the view for tourist service quality, express for tourist service emotion or
Tourist service, the quality evaluation of regional tourism service, improvement outing dress are assessed other tourists and are selected in impression, these comments
The quality of business has critically important reference significance.But in face of the tourism comment data of magnanimity, pass through artificial or simple statistics
How analysis method, value that is not only time-consuming and laborious but also being difficult profound performance tourism comment, effectively analyze the feelings of tourism comment
Sense tendency, and the quality of tourist service is evaluated on this basis, for " eating in tourist's selection tourist famous-city, tourism process
All kinds of services such as row trip purchase joy " provide decision-making foundation, provide regional tourism quality evaluation for tourism authorities, improve outing dress
Business offer tool and foundation.
Currently, the technology for carrying out emotional orientation analysis to comment text is broadly divided into two major classes: first is that based on emotion word
The method of allusion quotation or meaning of one's words knowledge, second is that being based on machine learning method.Currently, above two method is equal in text emotion analysis
There is utilization, in general, is become apparent using the advantage that the method for machine learning carries out text emotion analysis, elasticity and accuracy rate
It is higher.
Using the text emotion analysis method of sentiment dictionary, accuracy is mainly according to the quality of dictionary creation, Wu Fafen
The word not contained in analysis dictionary.Publication No. CN106156287A " the scenic spot evaluation data analysis based on tourism demand template
The patent of invention of public sentiment satisfaction method " discloses a kind of side that tourist attraction satisfaction is analyzed based on keyword template library
Method, it is believed that it is the sentiment analysis method using sentiment dictionary, depends critically upon user and emotion word assign and divide, it cannot
New word is handled, it is poor to Sentiment orientation precision of analysis expressed by a comment entirety.
Using the text emotion analysis method of machine learning method, mainly use the features such as emotion word, part of speech, syntax as
Basis of classification carries out emotion/tendentiousness to text by machine learning algorithm model and judges, but needs suitably and a certain number of
Manual identification.There are many algorithm model that currently used machine learning method text emotion analysis uses, such as KNN classification,
Maximum entropy classification, support vector cassification method etc., main process are manually to be marked to text, establish classification based training
Collection, with training set train classification models, each algorithm model has respective advantage and characteristic.Using machine learning method into
The technology of this sentiment analysis of composing a piece of writing mostly uses single disaggregated model, such as " the web topic tendency of Publication No. CN103116644A
Property excavate and the method for decision support " analysis that emotion tendentiousness of text is mainly carried out using support vector cassification method, by
It is limited to the preference of single classification method, accuracy and scalability are poor.
Therefore, how on the basis of assessing all kinds of machine learning algorithm models has scarce, select performance and accuracy mutual
Balance, and it is suitble to the sorting algorithm model of tourism industry feature, to the efficiency and validity for improving tourism comment and analysis, have very
High researching value.
Summary of the invention
In order to solve the above-mentioned technical problem the present invention provides a kind of tourist net comment sentiment analysis and service quality evaluation
Method.
The present invention is achieved through the following technical solutions:
A kind of method of tourist net comment sentiment analysis, comprising the following steps:
A, tourist net evaluation is pre-processed；
B, emotion trend training set is constructed；
C, Logic Regression Models, supporting vector machine model and naive Bayesian mould is respectively trained using emotion trend training set
Type exports the training result G of three kinds of propaedeutics modelsm(x)；
D, ballot device model is established, the step C training result exported is handled, the emotion that comment is calculated is inclined
To；
E, input ballot device model obtains the emotion of every comment after handling tourist net comment data to be tested
Tendency.
Synthesis of this programme by a variety of machine learning algorithm models, integrated logic recurrence, support vector machines and Bayes
Three kinds of sorting algorithm models overcome the efficiency and accuracy problem of single algorithm model, promote the standard of tourism comment sentiment analysis
True property is that tourist selects service delivery in tourism to help, provides for the evaluation of tourism authorities and improvement region service quality in tourism
Decision support.
Carrying out pretreated method to tourist net evaluation includes:
User is obtained to travel comment data, at least from eat, live, go, travel, entertain, shopping side faces data and stores；
Format specification data, and it is removed space, removal reprocessing.
Construct emotion trend training set specifically:
B1, word segmentation processing is carried out to data using participle tool；
B2, at least the building comment dictionary in terms of eating, living, go, travel, entertain, do shopping；
B3, building comment tfidfI, jVector,
tfidfi,j=tfi,j×idfi,
Wherein:
ni,jIndicate ith feature word in comment djIn frequency of occurrence, ∑knk,jIt is then comment djIn all words go out
The sum of occurrence number, | D | indicate the comment sum in corpus, | { j:ti∈djIndicate to include Feature Words tiNumber of reviews；
B4, mark emotion tend to.
Step D specifically:
Calculate the error rate of propaedeutics model
Calculate the weight of propaedeutics model
Construct the linear combination of propaedeutics model
Building ballot device model
Gm(x) aforementioned basic classification device, e are indicatedmIndicate basic classification device Gm(x) error rate, αmIndicate Gm(x) final
Importance in class device, effect of the smaller basic classification device of error classification rate in final classification device is bigger, m=1, and 2,3, M
=3.Wherein training data T={ (x1,y1),(x2,y2),...,(xN,yN), yi∈ { -1 ,+1 }, i=1,2 ..., N, N are represented
The quantity of training data.
A kind of service quality in tourism evaluation method, comprising the following steps:
The Sentiment orientation of the tourist famous-city comment to be detected with evaluation is obtained using the above method；
Obtained result is shown.
The methods of exhibiting includes at least one of method P1, method P2, method P3, method P4,
Method P1 are as follows: in terms of eating, living, go, travel, entertain, do shopping, calculate favorable comment and the poor number and percentage commented simultaneously
It shows；
Method P2 are as follows: by eating, living, going, travelling, entertaining, aspect of doing shopping, show the favorable comment of each aspect subordinate dimension and poor
Quantity, percentage and the corresponding representative comment content commented；
Method P3 are as follows: pass through fuzzy search and calculate and comment quantity, percentage with difference with the associated favorable comment of keyword and show.
Method P4 are as follows: calculating eats, lives, going, travelling, entertaining, the difference for various aspects of doing shopping comments number, when its accounting reaches threshold value
When, issue the user with warning.
Compared with prior art, the present invention having the following advantages and benefits:
1, synthesis of the present invention by a variety of machine learning algorithm models, integrated logic recurrence, support vector machines and pattra leaves
This three kinds of sorting algorithm models overcome the efficiency and accuracy problem of single algorithm model, promote tourism comment sentiment analysis
Accuracy is that tourist selects service delivery in tourism to help, mentions for the evaluation of tourism authorities and improvement region service quality in tourism
For decision support.
Detailed description of the invention
Attached drawing described herein is used to provide to further understand the embodiment of the present invention, constitutes one of the application
Point, do not constitute the restriction to the embodiment of the present invention.
Fig. 1 is the functional block diagram of 1 method of embodiment.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below with reference to embodiment and attached drawing, to this
Invention is described in further detail, and exemplary embodiment of the invention and its explanation for explaining only the invention, are not made
For limitation of the invention.
Embodiment 1
A kind of method of tourist net comment sentiment analysis as shown in Figure 1, comprising the following steps:
A, tourist net evaluation is pre-processed；
B, emotion trend training set is constructed；
C, Logic Regression Models, supporting vector machine model and naive Bayesian mould is respectively trained using emotion trend training set
Type exports the training result G of three kinds of propaedeutics modelsm(x)；
D, ballot device model is established, the step C training result exported is handled, the emotion that comment is calculated is inclined
To；
E, input ballot device model obtains the emotion of every comment after handling tourist net comment data to be tested
Tendency.
Embodiment 2
Principle based on the above embodiment, the present embodiment enumerate a specific embodiment and are described.
A, it obtains user using modes such as web crawlers to travel comment data, at least from eating, live, go, travel, entertain, purchase
Object space is stored in face of data,
Wherein, from the point of view of existing network data, it is above-mentioned eat, live, going, travelling, entertaining, do shopping aspect data can from
Lower website obtains:
It eats: taking journey, public comment etc.；
Firmly: hornet nest, way ox, skill dragon, donkey mother, same to journey, public comment etc.；
Row: public comment etc.；
Tourism: Baidu's tourism, Jingdone district, take journey, same to journey, go where, hornet nest, donkey mother, way ox；
Shopping: journey, public comment etc. are taken；
Amusement: public comment etc..
Using hive HQL format specification data, and it is removed space, removal reprocessing.It is calculated using Dynamic Programming
Method and python text-processing technology carry out reduplicated word, folded word, reiterative sentence processing to comment data.
B, emotion trend training set is constructed:
B1, word segmentation processing is carried out to data using python jieba participle tool, ython jieba is currently mainstream
Chinese word segmentation tool, and part of speech matter can be identified；
With for " the safe scented rice gone, taste or all well and good ", the result after being segmented by python jieba is " to go
The u of the u Thailand meters of pretty good a of q, wp taste n or d very d of perfume m of v
□”。
Wherein, the corresponding relationship of part-of-speech tagging is as shown in table 1.
1 part of speech corresponding relationship of table
| Label | Part of speech | Label | Part of speech | Label | Part of speech |
| Ag | Shape morpheme | k | It is followed by ingredient | tg | Tense morpheme |
| a | Adjective | l | Idiom | t | Time word |
| ad | Secondary shape word | m | Number | u | Auxiliary word |
| an | Adnoun | Ng | Name morpheme | vg | Dynamic morpheme |
| b | Distinction word | n | Noun | v | Verb |
| c | Conjunction | nr | Name | vd | Secondary verb |
| dg | Secondary morpheme | ns | Place name | vn | Name verb |
| d | Adverbial word | nt | Group of mechanism | w | Punctuation mark |
| e | Interjection | nz | Other proper names | x | Non- morpheme word |
| f | The noun of locality | o | Onomatopoeia | y | Modal particle |
| g | Morpheme | p | Preposition | z | Descriptive word |
| h | Enclitics | q | Quantifier | un | Unknown word |
| i | Chinese idiom | r | Pronoun | tg | Tense morpheme |
| j | Abbreviation abbreviation | s | Place word | t | Time word |
B2, at least the building comment dictionary in terms of eating, living, go, travel, entertain, do shopping；For example, comment relevant to eating
It include " taste ", " nice ", the words such as " fresh " that and comprising " room ", " hotel " is " preceding in firmly relevant comment dictionary in dictionary
The words such as platform " include " airport ", " railway station ", the words such as " subway ", comment dictionary relevant to trip in comment dictionary relevant to row
In include " scenic spot ", " admission ticket ", the words such as " scenery " include " environment ", " teacher ", " experience " etc. in comment dictionary relevant to joy
Word includes " service ", " market ", the words such as " activity " in comment dictionary relevant to purchase.
B3, building comment tfidfI, jVector,
tfidfi,j=tfi,j×idfi,
Wherein:
ni,jIndicate ith feature word in comment djIn frequency of occurrence, ∑knk,jIt is then comment djIn all words go out
The sum of occurrence number, | D | indicate the comment sum in corpus, | { j:ti∈djIndicate to include Feature Words tiNumber of reviews；
B4, mark emotion tend to, which can be used manual type realization.It is sampled and filters out from comment data concentration
30000 comment vectors carry out artificial emotion to comment vector using keyword retrieval technology and tend to mark, will such as contain and " take advantage of
Deceive ", the comment of the words such as " complaint " be labeled as difference and comment, will the comment mark containing the words such as " very satisfied ", " being worth recommending " preferably
It comments.Emotion manually is carried out to comment vector and tends to mark, difference, which is commented, is labeled as 0, and favorable comment is labeled as 1, and annotation results are stored in
Comment on the end of vector.
For example, abovementioned steps B expression is positive emotion, it is favorable comment by artificial judgment, is labeled as 1, then by artificial
After marking emotion trend, result is " to remove u Thailand meters of q, wp taste n or d of perfume m of v
The u 1 " of the pretty good a of very d.
Emotion trend training set T={ (x is obtained after step 2 is handled1,y1),(x2,y2),...,(xN,yN), It is j-th of feature of i-th of sample,ajlIt is j-th of feature
First of value that word may take, j=1,2 ..., n, l=1,2 ..., Sj。
C, Logic Regression Models, supporting vector machine model and naive Bayesian mould is respectively trained using emotion trend training set
Type exports the training result G of three kinds of propaedeutics modelsm(x)；
Wherein, the algorithm that Logic Regression Models use are as follows:
Training dataset T={ (x1,y1),(x2,y2),...,(xN,yN), yi∈ { 0,1 }, i=1,2 ..., N, w are power
It is worth vector, b is biasing, m=1.
Wherein, the calculating process of weight vector w and biasing b are as follows:
If β=(w；B),
The algorithm that supporting vector machine model uses are as follows:
Gm(x)=f (x)=sign (w*·x+b*)
Wherein training dataset T={ (x1,y1),(x2,y2),...,(xN,yN), yi∈ { -1 ,+1 }, i=1,2 ...,
N, m=2.
Wherein w*And b*Calculating process are as follows:
If w*And b*To meet yi(wT·xi+ b) >=1 most there are solution and w*·x+b*=0.
The algorithm that model-naive Bayesian uses are as follows:
Wherein training dataset T={ (x1,y1),(x2,y2),...,(xN,yN), It is
J-th of feature of i-th of sample,ajlIt is first of value that j-th of feature may take, j=1,
2 ..., n, l=1,2 ..., Sj, yi∈{c1,c2,...,cK, m=3.
Wherein, if CkMeet
D, ballot device model is established, the step C training result exported is handled, the emotion that comment is calculated is inclined
To；
Step D specifically:
Calculate the error rate of propaedeutics model
Calculate the weight of propaedeutics model
Construct the linear combination of propaedeutics model
Building ballot device model
Gm(x) aforementioned basic classification device, e are indicatedmIndicate basic classification device Gm(x) error rate, αmIndicate Gm(x) final
Importance in classifier, effect of the smaller basic classification device of error classification rate in final classification device is bigger, m=1, and 2,3,
M=3.
Wherein training data T={ (x1,y1),(x2,y2),...,(xN,yN), yi∈ { -1 ,+1 }, i=1,2 ..., N, N
Represent the quantity of training data.
E, tourist net comment data to be tested is stored and is cleaned, data are divided by the method for step B
Word processing, obtains the tfidf of the dataI, jVector；By tfidfI, jBallot device model in vector input step four, is calculated
The Sentiment orientation of every comment.
Embodiment 3
Based on the above embodiment, the present embodiment discloses a kind of service quality in tourism evaluation method, comprising the following steps:
The Sentiment orientation of the tourist famous-city comment to be detected with evaluation is obtained using the method for above-described embodiment；
Obtained result is shown.
The methods of exhibiting includes at least one of method P1, method P2, method P3, method P4,
Method P1 are as follows: in terms of eating, living, go, travel, entertain, do shopping, calculate favorable comment and the poor number and percentage commented simultaneously
It shows；
Method P2 are as follows: by eating, living, going, travelling, entertaining, aspect of doing shopping, show the favorable comment of each aspect subordinate dimension and poor
Quantity, percentage and the corresponding representative comment content commented；
Specifically, the subordinate that eats, live, going, travelling, entertaining, doing shopping analyze dimension can according to the form below conclude:
Method P3 are as follows: pass through fuzzy search and calculate and comment quantity, percentage with difference with the associated favorable comment of keyword and show.
Method P4 are as follows: calculating eats, lives, going, travelling, entertaining, the difference for various aspects of doing shopping comments number, when its accounting reaches threshold value
When, issue the user with warning.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not intended to limit the present invention the foregoing is merely a specific embodiment of the invention
Protection scope, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (6)
1. a kind of method of tourist net comment sentiment analysis, which comprises the following steps:
A, tourist net evaluation is pre-processed；
B, emotion trend training set is constructed；
C, Logic Regression Models, supporting vector machine model and model-naive Bayesian is respectively trained using emotion trend training set,
Export the training result G of three kinds of propaedeutics modelsm(x)；
D, ballot device model is established, the step C training result exported is handled, the Sentiment orientation of comment is calculated；
E, input ballot device model obtains the Sentiment orientation of every comment after handling tourist net comment data to be tested.
2. a kind of method of tourist net comment sentiment analysis according to claim 1, which is characterized in that tourist net
Evaluation carries out pretreated method
User is obtained to travel comment data, at least from eat, live, go, travel, entertain, shopping side faces data and stores；
Format specification data, and it is removed space, removal reprocessing.
3. a kind of method of tourist net comment sentiment analysis according to claim 1, which is characterized in that building emotion becomes
Gesture training set specifically:
B1, word segmentation processing is carried out to data using participle tool；
B2, at least the building comment dictionary in terms of eating, living, go, travel, entertain, do shopping；
B3, building comment tfidfI, jVector,
tfidfi,j=tfi,j×idfi,
Wherein:
ni,jIndicate ith feature word in comment djIn frequency of occurrence, ∑knk,jIt is then comment djIn all words go out occurrence
The sum of number, | D | indicate the comment sum in corpus, | { j:ti∈djIndicate to include Feature Words tiNumber of reviews；
B4, mark emotion tend to.
4. a kind of method of tourist net comment sentiment analysis according to claim 1, which is characterized in that step D is specific
Are as follows:
Calculate the error rate of propaedeutics model
Calculate the weight of propaedeutics model
Construct the linear combination of propaedeutics model
Building ballot device model
Gm(x) aforementioned basic classification device, e are indicatedmIndicate basic classification device Gm(x) error rate, αmIndicate Gm(x) in final classification
Importance in device, effect of the smaller basic classification device of error classification rate in final classification device is bigger, m=1, and 2,3, M=
3。
Wherein training data T={ (x1,y1),(x2,y2),...,(xN,yN), yi∈ { -1 ,+1 }, i=1,2 ..., N, N are represented
The quantity of training data.
5. a kind of service quality in tourism evaluation method, which comprises the following steps:
The Sentiment orientation of the tourist famous-city comment to be detected with evaluation is obtained using Claims 1-4 either method；
Obtained result is shown.
6. a kind of service quality in tourism evaluation method according to claim 5, which is characterized in that the methods of exhibiting includes
At least one of method P1, method P2, method P3, method P4,
Method P1 are as follows: in terms of eating, living, go, travel, entertain, do shopping, calculate the number and percentage that favorable comment and difference are commented and exhibition
Show；
Method P2 are as follows: in terms of eating, living, go, travel, entertain, do shopping, show what the favorable comment of each aspect subordinate dimension and difference were commented
Quantity, percentage and corresponding representative comment content；
Method P3 are as follows: pass through fuzzy search and calculate and comment quantity, percentage with difference with the associated favorable comment of keyword and show.
Method P4 are as follows: calculating eats, lives, going, travelling, entertaining, the difference for various aspects of doing shopping comments number, when its accounting reaches threshold value, to
User gives a warning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810803792.4A CN109034893A (en) | 2018-07-20 | 2018-07-20 | A kind of tourist net comment sentiment analysis and QoS evaluating method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810803792.4A CN109034893A (en) | 2018-07-20 | 2018-07-20 | A kind of tourist net comment sentiment analysis and QoS evaluating method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109034893A true CN109034893A (en) | 2018-12-18 |
Family
ID=64644661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810803792.4A Pending CN109034893A (en) | 2018-07-20 | 2018-07-20 | A kind of tourist net comment sentiment analysis and QoS evaluating method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109034893A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858973A (en) * | 2019-02-18 | 2019-06-07 | 成都中科大旗软件有限公司 | A kind of analysis method of regional tourism industry development |
| CN109978389A (en) * | 2019-03-28 | 2019-07-05 | 吴道钰 | The method that a kind of pair of tourist attractions carry out consumer's overall merit |
| CN110059922A (en) * | 2019-03-11 | 2019-07-26 | 北京比速信息科技有限公司 | Satisfaction evaluation method on the line of data is commented on based on internet tourist |
| CN110083726A (en) * | 2019-03-11 | 2019-08-02 | 北京比速信息科技有限公司 | A kind of destination image cognitive method based on UGC image data |
| CN111159342A (en) * | 2019-12-26 | 2020-05-15 | 北京大学 | Park text comment emotion scoring method based on machine learning |
| CN111340385A (en) * | 2020-03-10 | 2020-06-26 | 深圳华侨城创新研究院有限公司 | Scientific measuring method for measuring joy index of tourist attraction |
| CN111414754A (en) * | 2020-03-19 | 2020-07-14 | 中国建设银行股份有限公司 | Emotion analysis method and device of event, server and storage medium |
| CN111737978A (en) * | 2020-07-02 | 2020-10-02 | 武汉卓尔数字传媒科技有限公司 | Shopping evaluation emotion analysis method and device and electronic equipment |
| CN112100367A (en) * | 2019-05-28 | 2020-12-18 | 贵阳海信网络科技有限公司 | Public opinion early warning method and device for scenic spot |
| CN112948482A (en) * | 2021-04-28 | 2021-06-11 | 云景文旅科技有限公司 | Data preprocessing method and system for machine learning of travel online clothing platform |
| CN113298367A (en) * | 2021-05-12 | 2021-08-24 | 北京信息科技大学 | Theme park perception value evaluation method |
| CN113298366A (en) * | 2021-05-12 | 2021-08-24 | 北京信息科技大学 | Tourism performance service value evaluation method |
| CN113393276A (en) * | 2021-06-25 | 2021-09-14 | 食亨(上海)科技服务有限公司 | Comment data classification method and device and computer readable medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103034626A (en) * | 2012-12-26 | 2013-04-10 | 上海交通大学 | Emotion analyzing system and method |
| CN107015965A (en) * | 2017-03-24 | 2017-08-04 | 苏州希格玛科技有限公司 | A kind of Chinese text sentiment analysis device and method |
| CN107885849A (en) * | 2017-11-13 | 2018-04-06 | 成都蓝景信息技术有限公司 | A kind of moos index analysis system based on text classification |
-
2018
- 2018-07-20 CN CN201810803792.4A patent/CN109034893A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103034626A (en) * | 2012-12-26 | 2013-04-10 | 上海交通大学 | Emotion analyzing system and method |
| CN107015965A (en) * | 2017-03-24 | 2017-08-04 | 苏州希格玛科技有限公司 | A kind of Chinese text sentiment analysis device and method |
| CN107885849A (en) * | 2017-11-13 | 2018-04-06 | 成都蓝景信息技术有限公司 | A kind of moos index analysis system based on text classification |
Non-Patent Citations (1)
| Title |
|---|
| 王新宇等: "基于机器学习的旅游大数据分析研究-以旅游网络评价情感分析为例", 《中国旅游评论：2016第二辑》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858973A (en) * | 2019-02-18 | 2019-06-07 | 成都中科大旗软件有限公司 | A kind of analysis method of regional tourism industry development |
| CN110059922A (en) * | 2019-03-11 | 2019-07-26 | 北京比速信息科技有限公司 | Satisfaction evaluation method on the line of data is commented on based on internet tourist |
| CN110083726A (en) * | 2019-03-11 | 2019-08-02 | 北京比速信息科技有限公司 | A kind of destination image cognitive method based on UGC image data |
| CN110083726B (en) * | 2019-03-11 | 2021-10-22 | 北京比速信息科技有限公司 | Destination image perception method based on UGC picture data |
| CN109978389A (en) * | 2019-03-28 | 2019-07-05 | 吴道钰 | The method that a kind of pair of tourist attractions carry out consumer's overall merit |
| CN112100367A (en) * | 2019-05-28 | 2020-12-18 | 贵阳海信网络科技有限公司 | Public opinion early warning method and device for scenic spot |
| CN111159342A (en) * | 2019-12-26 | 2020-05-15 | 北京大学 | Park text comment emotion scoring method based on machine learning |
| CN111340385A (en) * | 2020-03-10 | 2020-06-26 | 深圳华侨城创新研究院有限公司 | Scientific measuring method for measuring joy index of tourist attraction |
| CN111414754A (en) * | 2020-03-19 | 2020-07-14 | 中国建设银行股份有限公司 | Emotion analysis method and device of event, server and storage medium |
| CN111737978A (en) * | 2020-07-02 | 2020-10-02 | 武汉卓尔数字传媒科技有限公司 | Shopping evaluation emotion analysis method and device and electronic equipment |
| CN112948482A (en) * | 2021-04-28 | 2021-06-11 | 云景文旅科技有限公司 | Data preprocessing method and system for machine learning of travel online clothing platform |
| CN113298367A (en) * | 2021-05-12 | 2021-08-24 | 北京信息科技大学 | Theme park perception value evaluation method |
| 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 |
| CN113298367B (en) * | 2021-05-12 | 2023-12-12 | 北京信息科技大学 | Theme park perception value evaluation method |
| CN113393276A (en) * | 2021-06-25 | 2021-09-14 | 食亨(上海)科技服务有限公司 | Comment data classification method and device and computer readable medium |
| CN113393276B (en) * | 2021-06-25 | 2023-06-16 | 食亨(上海)科技服务有限公司 | Comment data classification method, comment data classification device and computer-readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109034893A (en) | A kind of tourist net comment sentiment analysis and QoS evaluating method | |
| Balbi et al. | Combining different evaluation systems on social media for measuring user satisfaction | |
| Annisa et al. | Opinion mining on Mandalika hotel reviews using latent dirichlet allocation | |
| Xu et al. | Problems in current text simplification research: New data can help | |
| Thelwall | Sentiment analysis for tourism | |
| Chang et al. | Why customers don’t revisit in tourism and hospitality industry? | |
| Khan et al. | Sentiment classification using sentence-level lexical based | |
| CN103646088A (en) | Product comment fine-grained emotional element extraction method based on CRFs and SVM | |
| Refaee et al. | Subjectivity and sentiment analysis of arabic twitter feeds with limited resources | |
| Khasawneh et al. | Sentiment analysis of Arabic social media content: a comparative study | |
| CN103995853A (en) | Multi-language emotional data processing and classifying method and system based on key sentences | |
| KR20180131146A (en) | Apparatus and Method for Identifying Core Issues of Each Evaluation Criteria from User Reviews | |
| CN111353044A (en) | Comment-based emotion analysis method and system | |
| Van Thin et al. | Two new large corpora for Vietnamese aspect-based sentiment analysis at sentence level | |
| Di Fabbrizio et al. | Starlet: multi-document summarization of service and product reviews with balanced rating distributions | |
| Strelluf | needs+ PAST PARTICIPLE in regional Englishes on Twitter | |
| Li et al. | Improved new word detection method used in tourism field | |
| Peterlin et al. | Automated content analysis: The review of the big data systemic discourse in tourism and hospitality | |
| George et al. | Comparison of LDA and NMF topic modeling techniques for restaurant reviews | |
| Ferreira et al. | Generating flexible proper name references in text: Data, models and evaluation | |
| Kaur et al. | Sentiment detection from Punjabi text using support vector machine | |
| Renganathan et al. | Dubai restaurants: A sentiment analysis of tourist reviews | |
| KR101265467B1 (en) | Method for extracting experience and classifying verb in blog | |
| Alsudais et al. | Large-scale sentiment analysis on airbnb reviews from 15 cities | |
| Piao et al. | Product reputation mining: bring informative review summaries to producers and consumers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181218 |