CN107944060A - A kind of product information search method towards automotive vertical website - Google Patents
A kind of product information search method towards automotive vertical website Download PDFInfo
- Publication number
- CN107944060A CN107944060A CN201810003077.2A CN201810003077A CN107944060A CN 107944060 A CN107944060 A CN 107944060A CN 201810003077 A CN201810003077 A CN 201810003077A CN 107944060 A CN107944060 A CN 107944060A
- Authority
- CN
- China
- Prior art keywords
- public praise
- product information
- data
- website
- retrieval result
- 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.)
- Granted
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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Abstract
The invention discloses a kind of product information search method towards automotive vertical website, comprise the following steps:Multiple automotive-type Vertical Website data are captured using crawler technology, and database is stored in the tree structure of website-vehicle-public praise;Regular and pretreatment operation is carried out to the data format in database so that the data format from different web sites, naming method are unified;Consider source web, public praise grade, public praise content, the user gradation of data, build retrieval result appraisement system, be retrieval result appraisement system indicator of distribution weight;Public praise feature vector is built according to retrieval result appraisement system, and combines sentiment analysis score, affiliated web site, public praise grade and the user gradation for delivering information of comprehensive public praise, obtain the retrieval result of automobile product information.The present invention solve the problems, such as to return the result in conventional retrieval method it is single, adapt to that emerging vocabulary is limited, can not carry out personalized ordering to specific user and particular demands.
Description
Technical field
The present invention relates to information retrieval field, more particularly to a kind of product information retrieval side towards automotive vertical website
Method.
Background technology
On August 4th, 2017, the 40th time of China Internet Network Information Center (CNNIC) issue《China Internet network develops
Situation statistical report》Display[1], by June, 2017, Chinese netizen's scale reaches 7.51 hundred million, accounts for five points of global netizen's sum
One of.Internet penetration is 54.3%, more than global 4.6 percentage points of average level.Electronics business using internet as representative
Business technology is accelerating to promote China's consumption upgrading, economic society transition, also profoundly promotes the change of people's consumption habit.
The user of product is delivered by the website of businessman, shopping portal website, associated vertical website and exchange community
The evaluation to product, on the one hand the consumption wish for other consumers there is considerable degree of influence[2], product
Buyer and user can influence to locate by delivering to product in the time of day of different behaviours in service, different service life
The desire to purchase of potential consumer during looking around, makes it consume the rationalization that more becomes[3];On the other hand, user comments product
The relation between consumer and enterprise is had a deep effect on by feedback, is the valuable source that enterprise understands user demand.Always can
The company of positive name recognition product is produced, the good relationship between consumer can be more cultivated and develop into brand loyalty[4].Cause
This, excavates the user feeling tendency for lying in product review behind, can more efficiently help enterprise to find that consumer concentrates
The defects of complaint, to cater to consumer demand[5]。
For automobile product due to higher price, vehicle and configuration categories are various, and need buyer's real experiences, therefore are difficult
Online purchase, dispatching are carried out as other ordinary consumption product.Also just because of this, user is sent out in the website such as automotive vertical website
The public praise (comment) of table believes purchased vehicle price, configuration, power, oil consumption, evaluation after sale etc. by containing car owner
Breath, judgement examination is carried out for consumer before purchase, and is understood product market reaction for enterprise businessman, found itself
Product defects, adjustment marketing strategy all have particularly important value[6].Gather as most active information automotive vertical website
Focus, is exactly optimal shopping guide's medium during user selects car, purchase car.Therefore search engine is using automotive-type as user
Vertical Website carries out important tool during information retrieval.Since automobile product type is various, complicated and user comment exists
Difference on expression form, causes retrieval result to be often inaccurate, while also lacks the personalized ordering to retrieval result.
The existing mode retrieved to automobile product information is mainly words matching, by accessing web page content, into
Establish index after row participle, and build index database, for the inquiry of query statement database, and according to correlation, when
Between or webpage weight be ranked up, retrieval result is finally presented[7]。
In the implementation of the present invention, discovery at least has the following disadvantages in the prior art and deficiency by inventor:
1) matched based on words, it is difficult to present to be retrieved the same or like product information of contents semantic with user;
2) new term and mutation vocabulary continue to bring out, and retrieval result is often difficult to reflect the letter really to be retrieved of user
Breath;
3) lacking individuality of retrieval result sorts, it is impossible to meets a certain specific area, specific crowd and particular demands
Searched targets, therefore also lack the personalized ordering to retrieval result.
Based on this, Internet automotive vertical website, with reference to text mining and data processing technique, the present invention constructs
It is a kind of to improve recall precision and the method for retrieval rate.
Bibliography
[1] Chinese netizen 7.51 hundred million accounts for global netizen 1/5th [J] the news world, and 2017, (09):5.
[2]Dincer H,Hacioglu U.Performance evaluation with fuzzy VIKOR and
AHP method based on customer satisfaction in Turkish banking sector[J]
.Kybernetes,2013,42(7):1072-1085.
[3]Xiao S,Wei C P,Dong M.Crowd intelligence:Analyzing online product
reviews for preference measurement[J].Information &Management,2016,53(2):169-
182.
[4]Farris P W,Bendle N,Pfeifer P,et al.Marketing metrics:The
definitive guide to measuring marketing performance[M].Pearson Education,
2010.
[5]Guo W,Liang R Y,Wang L,et al.Exploring sustained participation in
firm-hosted communities in China:the effects ofsocial capital and active
degree[J].Behaviour&Information Technology,2017,36(3):223-242.
[6] automobile Method for Sales Forecast researchs of Yuan Qingyu, Peng Geng, Liu Ying, the Lv Benfu based on Internet Keyword search data
[J] managerialists (scholarly edition), 2011, (01):12-24.
[7] Cao Shujin, Chen Yijin, Satisfaction of Library Users positive research [J] China figures of the Tao Yang based on user demand
Book shop journal, 2013,39 (05):60-75.
The content of the invention
The present invention provides a kind of product information search method towards automotive vertical website, the present invention solves traditional inspection
Returned the result in Suo Fangfa it is single, adapt to that emerging vocabulary is limited, can not carry out personalized ordering to specific user and particular demands
The problem of, it is described below:
A kind of product information search method towards automotive vertical website, the product information search method include following step
Suddenly:
Multiple automotive-type Vertical Website data are captured using crawler technology, and with the tree structure of website-vehicle-public praise
It is stored in database;
Regular and pretreatment operation is carried out to the data format in database so that the data format from different web sites,
Naming method is unified;
Consider source web, public praise grade, public praise content, the user gradation of data, build retrieval result appraisement system, be
Retrieval result appraisement system indicator of distribution weight;
Public praise feature vector is built according to retrieval result appraisement system, and combines sentiment analysis score, the institute of comprehensive public praise
Belong to website, public praise grade and the user gradation for delivering information, obtain the retrieval result of automobile product information.
The multiple automotive-type Vertical Website data are specially:
1) Vertical Website data, including:The anti-chain number of website, post number of users and comprising vehicle quantity;
2) model data, including:Vehicle classification, vehicle brand and vehicle price range in Vertical Website;
3) public praise data, including:User is for the scorings of different bus attributes, user class, purchase car time, purchase car
Point, public praise content, public praise grade.
The data format in database carries out regular and pretreatment operation:
Establish vehicle name mapping table:The different names of same vehicle in different car websites are mutually mapped, unified life
Name;
Establish bus attribute name mapping table:Bus attribute title in different automotive vertical websites is mutually mapped, is united
One name;
Delete the repetition public praise that user delivers, and the incomplete public praise of deleted data item;Structure automotive field keywords database,
Disable dictionary, emotion dictionary and synonym dictionary;Participle, part-of-speech tagging and go stop words to handle.
Further, the retrieval result appraisement system is specially:
Retrieval result appraisement system is total to two-stage, and first class index is integrated retrieval result P, and two-level index belongs to for the related of automobile
Property Ii, association attributes IiDetermined by expert estimation.
Wherein, the association attributes IiIncluding:Power, manipulation, appearance, interior trim, space, oil consumption, comfort, manipulation and
After sale.
Further, the expert estimation is specially:
Build the multilevel iudge matrix based on expert estimation, multilevel iudge matrix represent when first class index and two-level index it
Between when there is contact, the comparison of relative importance between two-level index.
Further, the public praise grade is specially:
Wherein, the user gradation is specially:
The beneficial effect of technical solution provided by the invention is:
1st, the present invention can be in time in hours section, and quick crawl automobile product is online on multiple websites
Public praise data, constantly update corpus and text database;
2nd, the present invention can integrate multiple automotive vertical website datas, while by user and its deliver the grade of public praise and include
Scope is considered, from automobile body attribute angle, by improving related dictionary and construction feature vector, is obtained more accurate
Automobile product retrieval result;
3rd, the present invention can be obtained by information such as the purchase car place in screening mouth landmark data, purchase car purpose, purchase car times
Product considers retrieval result and its variation tendency under scope, specific aim in different time scope, different geographical, different demands etc.
Meet the fine-grained cognitive need of consumer and manufacturing enterprise to product.
Brief description of the drawings
Fig. 1 is a kind of flow chart of product information search method towards automotive vertical website;
Fig. 2 is the schematic diagram of data store organisation;
Fig. 3 is the schematic diagram of retrieval result assessment indicator system;
Fig. 4 is the schematic diagram that retrieval operates in detail.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, embodiment of the present invention is made below further
It is described in detail on ground.
Embodiment 1
An embodiment of the present invention provides a kind of product information search method towards automotive vertical website, referring to Fig. 1, the party
Method comprises the following steps:
101:Multiple automotive-type Vertical Website data are captured using crawler technology, and with the tree of " website-vehicle-public praise "
Shape structure is stored in database;
102:Regular and pretreatment operation is carried out to the data format in database so that the data lattice from different web sites
Formula, naming method are unified;
103:Consider source web, public praise grade, public praise content, the user gradation of data, structure retrieval result evaluation body
System, is retrieval result appraisement system indicator of distribution weight;
104:Public praise feature vector is built according to retrieval result appraisement system, and combines sentiment analysis score, comprehensive public praise
Affiliated web site, public praise grade and the user gradation for delivering information, obtain the retrieval result of automobile product information.
In conclusion the embodiment of the present invention is realized towards automotive-type Vertical Website by above-mentioned steps 101- steps 104,
The magnanimity word-of-mouth information that the different user of multiple websites can be utilized to deliver, improves overall inspection of the user using automotive vertical website
Rope efficiency and retrieval rate.
Embodiment 2
The scheme in embodiment 1 is further introduced with reference to specific calculation formula, example, it is as detailed below
Description:
201:The related datas such as the acceptance of the users of multiple automotive vertical websites are obtained by data grabber;
Wherein, which is specially:
1) according to webpage http protocol, using the acceptance of the users data of crawler technology crawl mainstream automotive class Vertical Website,
By the regular expression made (expression formula is known to those skilled in the art, and the embodiment of the present invention does not repeat this)
All vehicles in website are gathered for traveling through, and the word-of-mouth information for crawling all users and delivering is traveled through further directed to every money vehicle.
Wherein, the data of above-mentioned crawl are specially:
1. Vertical Website data, including:Post in the anti-chain number of website, Vertical Website in number of users and Vertical Website
Comprising vehicle quantity;
2. model data, including:Vehicle classification, vehicle brand and vehicle price range for including in Vertical Website etc.;
3. public praise data, including:User is for the scorings of different bus attributes, user class, purchase car time, purchase car
Point, public praise content, public praise grade etc..
After first crawl, it is every 24 it is small when increment crawl is carried out to several above-mentioned data item, it is ensured that the steady growth of data
With temporal continuity.
2) data of crawl are stored in local data base with the tree structure of " website-vehicle-public praise ", such as Fig. 2 institutes
Show.
202:Data format arrangement, data cleansing and Text Pretreatment;
1) vehicle name mapping table is established:The different names of same vehicle in different car websites are mutually mapped, it is unified
Name;
2) bus attribute name mapping table is established:Bus attribute title in different automotive vertical websites is mutually mapped,
Uniform Name, such as " comfort " and " comfort level ";
3) the repetition public praise that user delivers, and the incomplete public praise of deleted data item are deleted;
Wherein, the incomplete public praise of data item is determined according to actual needs, and the embodiment of the present invention does not repeat this.
4) build automotive field keywords database, disable dictionary, emotion dictionary and synonym dictionary.
When carrying out text-processing to user vehicle public praise, original general dictionary can not cover the key of automotive field
Word, therefore automotive field should be directed to and build corresponding dictionary, specifically include:
1. automotive field keywords database:Based on automobile body structure vocabulary and product requirement specification book, more moneys are merged
Automotive field dictionary in Chinese character coding input method, be aided with artificial supplementation automotive field writes a Chinese character in simplified form vocabulary, emerging vocabulary and and automobile
The relevant attached vocabulary (such as " passability ", " abnormal sound ") of attribute, builds automotive field keywords database;
2. disable dictionary:It will be deleted with bus attribute and the relevant stop words of evaluation from deactivation vocabulary, will be with automobile category
Property and the unrelated vocabulary of evaluation added to disabling vocabulary;
3. emotion dictionary:The adjective with Sentiment orientation for describing, describing bus attribute and adverbial word are added into emotion
Word (such as), can more accurately hold the Sentiment orientation of the bus attribute of user's description;
4. synonym dictionary:For user when public praise evaluation is carried out, different users often uses different vocabulary
Or form of presentation describes the same attribute (such as " headlight " and " front car light ") of automobile product, therefore synonym dictionary should be built,
Conclude and classify easy to feature.
5) participle, part-of-speech tagging and go stop words to handle:Original public praise text message includes more and automobile product
Entity information is unrelated or the less information of correlation, these information can be that following entities feature extraction and sentiment analysis bring and make an uproar
Sound, therefore text should be simplified.
Automotive field keywords database is imported in the Chinese word segmentation instrument ICTCLAS that the Chinese Academy of Sciences releases with disabling dictionary, will
All acceptance of the users information obtained are segmented, part-of-speech tagging and go stop words to handle, and are retained relevant with automotive field
Noun, noun phrase, adjective, adverbial word etc..
203:Automobile product retrieval result two-level index appraisement system is established, and distributes two-level index weight;
Driven according to assessment item of the user when public praise is delivered in automotive vertical website, and composite vehicle owner daily
The bus attribute paid close attention to during sailing, establishes retrieval result assessment indicator system, as shown in Figure 3.
Wherein, retrieval result assessment indicator system is total to two-stage, and first class index is integrated retrieval result P, and two-level index is vapour
The association attributes I of cari, i ∈ [1,9].Respectively power, manipulation, appearance, interior trim, space, oil consumption, comfort, manipulate and sell
Afterwards, association attributes IiDetermined by expert estimation.
Application level analytic approach (AHP), builds the multilevel iudge matrix M based on expert estimation (i.e. to association attributes IiInto
Row marking).Multilevel iudge matrix M is represented when first class index P (referred to herein as automobile product retrieval result) is between two-level index I
During in the presence of contact, two-level index IiBetween relative importance comparison.The relational expression of multilevel iudge matrix M is represented by:
Wherein, n=9.
Multilevel iudge matrix M represents index IiWith index IjWhen being compared relative to first class index P, both are relatively important
The degree of membership of degree, following multilevel iudge matrix M is obtained according to foregoing description:
Above-mentioned multilevel iudge matrix M is solved according to matrix theory, obtains the corresponding spy of maximum eigenvalue of multilevel iudge matrix M
Levy vector μ*, it is normalized to μ=[μ1,μ2,…,μn]T, be each two-level index weight.
204:Calculate automobile product retrieval result evaluation index.
1) according to automobile product retrieval result two-level index appraisement system structure public praise feature vector;
The automobile product retrieval result two-level index appraisement system built according to step 203, by public praise subordinate sentence, and for every
One public praise builds formFive yuan of Vector Groups of public praise.
Wherein, α is product entity (such as certain vehicle), and β is characteristic attribute (such as space, manipulation, an appearance of entity alpha
The characteristic attributes such as design), γ is the publisher of public praise, and η is the time that public praise is delivered,For subjective assessment emotion score, represent
User γ delivers the emotion score of subjective assessment in time η for the characteristic attribute β of entity alpha.α in These parameters dimension,
Beta, gamma, η can be directly obtained by the data captured,It need to be obtained by affection computation.
2) with Chinese sentiment analysis instrument SnowNLP, affection computation is carried out to public praise feature vector, obtains two-level index
J-th strip public praise emotion score d under iij;
Wherein, dij∈ [0,1], for ease of calculating, by dijIt is transformed into [1,5] section, dij'=5dij∈[0,5]。
3) public praise score weight is adjusted;
Since the initial data of automobile product retrieval result two-level index appraisement system comes from the automotive vertical net of crawl
Stand data, its confidence level is influenced by factors such as its source web, user class, public praise grades, therefore should be according to these factor tune
Whole sentiment analysis score, includes in detail:
A) public praise attribute C;
The recommendation degree that public praise grade is divided by each website for the public praise of user.Uniformly be set as elite public praise,
Recommend public praise and common public praise.Public praise higher grade, its public praise content reliability is higher.
B) user property U;
User in automotive vertical website is divided into certification user and non-authentication user.Certification user is to carry out purchase car in website
The user of information registration, correspondingly, non-authentication user is the use that the information registration of purchase car is only registered but not yet carried out in website
Family.The confidence level with a high credibility that content is delivered in non-authentication user is commented in the public praise that certification user is delivered.Empirically it is worth,
Have:
C) website attribute W;
The influence power of website is an important factor for influencing customer consumption selection.Net with higher authoritative and the degree of recognition
Stand information, can largely be propagated by its influence power, so as to influence the consumption policy of consumer.Website herein
Influence power selects the anti-chain number (Anti-chain number) of website.
Anti-chain number refers to the number of links that certain website is imported into from other website, and the quality for importing link directly determines certain
The weight of website in a search engine.The anti-chain number of website is more, then the concerned degree of website and authority are higher.I.e. some hangs down
The weight of straight website is proportions of its anti-chain number in all crawl website anti-chain number summations, i.e.,:
Wherein, WjInfluence power for the affiliated Vertical Websites of public praise j gathered, bjFor the anti-chain of the affiliated Vertical Websites of public praise j
Number,To gather total anti-chain number of multiple websites, the data of N number of website are gathered altogether.
It can be obtained by above attribute, certain vehicle j-th strip public praise is directed to the score p of two-level index iij=dij′CjUjWj.If
K bar public praise data are grabbed altogether for two-level index i, then two-level index i is scored atpiAs public praise feature
In vectorValue.Similarly, according to the obtained each two-level index weight mu of step 203i, the retrieval result of such vehicle can be obtained
It is scored at
In conclusion the embodiment of the present invention can obtain the retrieval of a certain vehicle, a certain bus attribute by above-mentioned steps
Score, and retrieval result is ranked up accordingly, so that generate is needed based on automotive vertical class website, user oriented personalized retrieval
The retrieval result asked, improves the accuracy of retrieval, meets a variety of needs in practical application.
To the model of each device in addition to specified otherwise is done, the model of other devices is not limited the embodiment of the present invention,
As long as the device of above-mentioned function can be completed.
It will be appreciated by those skilled in the art that attached drawing is the schematic diagram of a preferred embodiment, the embodiments of the present invention
Sequence number is for illustration only, does not represent the quality of embodiment.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent replacement, improvement and so on, should all be included in the protection scope of the present invention.
Claims (8)
- A kind of 1. product information search method towards automotive vertical website, it is characterised in that the product information search method Comprise the following steps:Multiple automotive-type Vertical Website data are captured using crawler technology, and are stored in the tree structure of website-vehicle-public praise Database;Regular and pretreatment operation is carried out to the data format in database so that the data format from different web sites, name Mode is unified;Consider source web, public praise grade, public praise content, the user gradation of data, build retrieval result appraisement system, for retrieval Evaluation of result system indicator of distribution weight;Public praise feature vector is built according to retrieval result appraisement system, and combines sentiment analysis score, the affiliated net of comprehensive public praise Stand, public praise grade and the user gradation for delivering information, obtain the retrieval result of automobile product information.
- A kind of 2. product information search method towards automotive vertical website according to claim 1, it is characterised in that institute Stating multiple automotive-type Vertical Website data is specially:1) Vertical Website data, including:The anti-chain number of website, post number of users and comprising vehicle quantity;2) model data, including:Vehicle classification, vehicle brand and vehicle price range in Vertical Website;3) public praise data, including:User is for the scoring of different bus attributes, user class, purchase car time, purchase car place, mouth Upright stone tablet content, public praise grade.
- A kind of 3. product information search method towards automotive vertical website according to claim 1, it is characterised in that institute State and be specially to the regular and pretreatment operation of data format progress in database:Establish vehicle name mapping table:The different names of same vehicle in different car websites are mutually mapped, Uniform Name;Establish bus attribute name mapping table:Bus attribute title in different automotive vertical websites is mutually mapped, unified life Name;Delete the repetition public praise that user delivers, and the incomplete public praise of deleted data item;Build automotive field keywords database, disable Dictionary, emotion dictionary and synonym dictionary;Participle, part-of-speech tagging and go stop words to handle.
- A kind of 4. product information search method towards automotive vertical website according to claim 1, it is characterised in that institute Stating retrieval result appraisement system is specially:Retrieval result appraisement system is total to two-stage, and first class index is integrated retrieval result P, and two-level index is the association attributes I of automobilei, Association attributes IiDetermined by expert estimation.
- A kind of 5. product information search method towards automotive vertical website according to claim 4, it is characterised in that institute State association attributes IiIncluding:Power, manipulation, appearance, interior trim, space, oil consumption, comfort, manipulation and after sale.
- A kind of 6. product information search method towards automotive vertical website according to claim 4, it is characterised in that institute Stating expert estimation is specially:The multilevel iudge matrix based on expert estimation is built, multilevel iudge matrix is represented to work as and deposited between first class index and two-level index In contact, the comparison of relative importance between two-level index.
- A kind of 7. product information search method towards automotive vertical website according to claim 1, it is characterised in that institute Stating public praise grade is specially:
- A kind of 8. product information search method towards automotive vertical website according to claim 1, it is characterised in that institute Stating user gradation is specially:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810003077.2A CN107944060B (en) | 2018-01-02 | 2018-01-02 | Product information retrieval method oriented to automobile vertical website |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810003077.2A CN107944060B (en) | 2018-01-02 | 2018-01-02 | Product information retrieval method oriented to automobile vertical website |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107944060A true CN107944060A (en) | 2018-04-20 |
CN107944060B CN107944060B (en) | 2020-07-31 |
Family
ID=61937320
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810003077.2A Expired - Fee Related CN107944060B (en) | 2018-01-02 | 2018-01-02 | Product information retrieval method oriented to automobile vertical website |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107944060B (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109408809A (en) * | 2018-09-25 | 2019-03-01 | 天津大学 | A kind of sentiment analysis method for automobile product comment based on term vector |
CN109862116A (en) * | 2019-03-13 | 2019-06-07 | 深圳市元征科技股份有限公司 | A kind of information of vehicles method for pushing and device |
CN110543547A (en) * | 2019-08-13 | 2019-12-06 | 广东数鼎科技有限公司 | automobile public praise semantic emotion analysis system |
CN110704391A (en) * | 2019-09-23 | 2020-01-17 | 车智互联(北京)科技有限公司 | Word stock construction method and computing device |
CN111143653A (en) * | 2019-12-27 | 2020-05-12 | 上海汽笛生网络科技有限公司 | Credibility verification method for mass science popularization resources |
CN111401988A (en) * | 2020-03-02 | 2020-07-10 | 重庆邮电大学 | Product configuration demand response system based on semantics and order generation method |
CN111581479A (en) * | 2020-05-11 | 2020-08-25 | 北京精准沟通传媒科技股份有限公司 | One-stop data processing method and device, storage medium and electronic equipment |
CN111723162A (en) * | 2020-06-19 | 2020-09-29 | 广州小鹏车联网科技有限公司 | Dictionary processing method, processing device, server and voice interaction system |
CN112016964A (en) * | 2020-08-27 | 2020-12-01 | 李忠耘 | Second-hand vehicle dynamic pricing method and device, electronic equipment and storage medium |
CN112395484A (en) * | 2020-11-18 | 2021-02-23 | 天津大学 | User satisfaction evaluation method for automatically driving automobile |
CN113408809A (en) * | 2021-06-29 | 2021-09-17 | 奇瑞汽车股份有限公司 | Automobile design scheme evaluation method and device and computer storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010176599A (en) * | 2009-02-02 | 2010-08-12 | Mitsubishi Electric Building Techno Service Co Ltd | Web function monitoring system |
CN103049435A (en) * | 2013-01-04 | 2013-04-17 | 浙江工商大学 | Text fine granularity sentiment analysis method and text fine granularity sentiment analysis device |
CN106844403A (en) * | 2015-12-07 | 2017-06-13 | 北京国双科技有限公司 | Information processing method and device |
CN106844787A (en) * | 2017-03-16 | 2017-06-13 | 四川大学 | It is a kind of for automobile industry finds targeted customer and matches the recommendation method of target product |
CN107403255A (en) * | 2017-06-30 | 2017-11-28 | 亳州中药材商品交易中心有限公司 | Chinese medicine price composite index assessment method and system |
-
2018
- 2018-01-02 CN CN201810003077.2A patent/CN107944060B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010176599A (en) * | 2009-02-02 | 2010-08-12 | Mitsubishi Electric Building Techno Service Co Ltd | Web function monitoring system |
CN103049435A (en) * | 2013-01-04 | 2013-04-17 | 浙江工商大学 | Text fine granularity sentiment analysis method and text fine granularity sentiment analysis device |
CN106844403A (en) * | 2015-12-07 | 2017-06-13 | 北京国双科技有限公司 | Information processing method and device |
CN106844787A (en) * | 2017-03-16 | 2017-06-13 | 四川大学 | It is a kind of for automobile industry finds targeted customer and matches the recommendation method of target product |
CN107403255A (en) * | 2017-06-30 | 2017-11-28 | 亳州中药材商品交易中心有限公司 | Chinese medicine price composite index assessment method and system |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109408809A (en) * | 2018-09-25 | 2019-03-01 | 天津大学 | A kind of sentiment analysis method for automobile product comment based on term vector |
CN109862116A (en) * | 2019-03-13 | 2019-06-07 | 深圳市元征科技股份有限公司 | A kind of information of vehicles method for pushing and device |
CN110543547A (en) * | 2019-08-13 | 2019-12-06 | 广东数鼎科技有限公司 | automobile public praise semantic emotion analysis system |
CN110704391A (en) * | 2019-09-23 | 2020-01-17 | 车智互联(北京)科技有限公司 | Word stock construction method and computing device |
CN111143653B (en) * | 2019-12-27 | 2023-06-23 | 上海杰图智能科技有限公司 | Credibility verification method for mass science popularization resources |
CN111143653A (en) * | 2019-12-27 | 2020-05-12 | 上海汽笛生网络科技有限公司 | Credibility verification method for mass science popularization resources |
CN111401988A (en) * | 2020-03-02 | 2020-07-10 | 重庆邮电大学 | Product configuration demand response system based on semantics and order generation method |
CN111581479A (en) * | 2020-05-11 | 2020-08-25 | 北京精准沟通传媒科技股份有限公司 | One-stop data processing method and device, storage medium and electronic equipment |
CN111723162A (en) * | 2020-06-19 | 2020-09-29 | 广州小鹏车联网科技有限公司 | Dictionary processing method, processing device, server and voice interaction system |
CN111723162B (en) * | 2020-06-19 | 2023-08-25 | 北京小鹏汽车有限公司 | Dictionary processing method, processing device, server and voice interaction system |
CN112016964A (en) * | 2020-08-27 | 2020-12-01 | 李忠耘 | Second-hand vehicle dynamic pricing method and device, electronic equipment and storage medium |
CN112395484A (en) * | 2020-11-18 | 2021-02-23 | 天津大学 | User satisfaction evaluation method for automatically driving automobile |
CN113408809A (en) * | 2021-06-29 | 2021-09-17 | 奇瑞汽车股份有限公司 | Automobile design scheme evaluation method and device and computer storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN107944060B (en) | 2020-07-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107944060A (en) | A kind of product information search method towards automotive vertical website | |
Park et al. | Tourism demand forecasting with online news data mining | |
Lu et al. | BizSeeker: a hybrid semantic recommendation system for personalized government‐to‐business e‐services | |
Król | Forgotten agritourism: abandoned websites in the promotion of rural tourism in Poland | |
Chang et al. | An improved model for sentiment analysis on luxury hotel review | |
Yadav et al. | Effect of trip mode on opinion about hotel aspects: A social media analysis approach | |
US20130311288A1 (en) | Online advertising valuation apparatus and method | |
US20090076887A1 (en) | System And Method Of Collecting Market-Related Data Via A Web-Based Networking Environment | |
Kim et al. | Why am I satisfied? See my reviews–Price and location matter in the restaurant industry | |
JP2009521750A (en) | Analyzing content to determine context and providing relevant content based on context | |
CN104268292A (en) | Label word library update method of portrait system | |
Liu et al. | What affects the online ratings of restaurant consumers: a research perspective on text-mining big data analysis | |
KR20160079299A (en) | System and Method on Intelligent Recommendation of Used Cars | |
KR102121901B1 (en) | System for online public fund investment management assessment service | |
Al Adwan et al. | E-marketing strategic for Jordanian small business to increase sale in local e-market | |
Jia | Measuring tourists’ meal experience by mining online user generated content about restaurants | |
US20070219987A1 (en) | Self Teaching Thesaurus | |
CN104050243B (en) | It is a kind of to search for the network search method combined with social activity and its system | |
Niu et al. | Product hierarchy-based customer profiles for electronic commerce recommendation | |
KR100707361B1 (en) | Public opinion research system using internet network and its method | |
Zhang et al. | Examining differences among book reviews from various online platforms | |
Shi et al. | Service quality assessment of travel agency O2O model based on improved evidence theory | |
La et al. | Understanding consumers’ sentiment expressions in online reviews: a hybrid approach | |
CN112395484A (en) | User satisfaction evaluation method for automatically driving automobile | |
Abbasi et al. | Website quality and evaluation: a perspective of Iranian airline industry |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200731 Termination date: 20210102 |
|
CF01 | Termination of patent right due to non-payment of annual fee |