CN103886501A - Post-loan risk early warning system based on semantic sentiment analysis - Google Patents
Post-loan risk early warning system based on semantic sentiment analysis Download PDFInfo
- Publication number
- CN103886501A CN103886501A CN201410138443.7A CN201410138443A CN103886501A CN 103886501 A CN103886501 A CN 103886501A CN 201410138443 A CN201410138443 A CN 201410138443A CN 103886501 A CN103886501 A CN 103886501A
- Authority
- CN
- China
- Prior art keywords
- sentiment analysis
- module
- relevant information
- warning system
- semantic
- 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
Images
Landscapes
- Machine Translation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a post-loan risk early warning system based on semantic sentiment analysis. The post-loan risk early warning system is characterized by comprising a network data mining module, a semantic sentiment analysis module, a total analysis module and a user interaction module. The network data mining module is used for collecting relevant information of customer enterprises from the network, wherein the relevant information comprises one or more of news, reviews, Microblogs, reports and complaints relevant to the client enterprises. The semantic sentiment analysis module is used for receiving the relevant information, analyzing the sentiment components of the relevant information and generating sentiment polarity K and sentiment intensity M. The total analysis module is used for obtaining the sentiment polarity K and the sentiment intensity M, generating the value of the sentiment polarity K and the value of the sentiment intensity M according to the source of the relevant information, and then obtaining a reliable coefficient P and an overall reliable coefficient W through calculation in sequence according to a predetermined formula. The user interaction module is used for giving a warning when the overall reliable coefficient W is smaller than a warning value. The post-loan risk early warning system based on semantic sentiment analysis can give an early warning for great changes of the client enterprises in time, help a bank to manage the client enterprises better, and effectively reduce post-loan risks.
Description
Technical field
The present invention relates to Warning System after a kind of loan based on semantic sentiment analysis, belong to computer realm.
Background technology
Along with socioeconomic high speed development, enterprises and individuals is likely to bank or financial institution's apply for loan.For example, enterprise, for expanding production scale of operation, needs the import of advanced technology and equipment, but these technology and equipment need to spend a large amount of funds conventionally, easily millions of, up to ten million units.Personal user is for floatation of a company or buy house, also needs to spend hundreds of thousands even up to a million.For these enterprises and individual, one-time payment so huge fund is very difficult, and solution just comprises and getting a bank loan.Enterprise or personal user, by bank's apply for loan, after bank verifies enterprise or individual's identity, sign loan agreement, then offer loans.
But, in prior art, user is between the operating period obtaining after loan, bank only can rely on artificial the going of its staff to collect with user-dependent various information, then information is carried out to Treatment Analysis, finally pass judgment on user's loan repayment capacity according to analysis result, can regain timely and effectively with loan and the interest of guaranteeing granting.But, in long-term practice, find, in huge information source, rely on completely and manually go collection, Treatment Analysis can exist with user-dependent information: the defect that workload is huge, information processing efficiency is lower and problem; To such an extent as to cannot notify in time related personnel and mechanism to trigger risk treatment scheme, cause bank not judge in time and to avoid risk.
Summary of the invention
The present invention proposes in view of the above problems, and its object is, Warning System after a kind of loan based on semantic sentiment analysis is provided, and to solve, workload is huge, information processing efficiency is compared with low and cannot trigger in time the problem of risk treatment scheme.
The invention provides Warning System after a kind of loan based on semantic sentiment analysis, it is characterized in that, this system comprises:
Network data excavation module, for collect the relevant information of Client Enterprise from network, described relevant information comprises following one or several: news, comment, microblogging, report, the complaint relevant to Client Enterprise;
Semantic sentiment analysis module, for receiving the market sense constituent analysis of going forward side by side of described relevant information, generates feeling polarities K and emotion intensity M;
Analyzing total module, for obtaining described feeling polarities K and described emotion intensity M, and generates feeling polarities K value and emotion intensity M value according to the source of described relevant information, calculates successively safety factor P and overall safety factor W afterwards according to predetermined formula;
User interactive module, for giving a warning during lower than warning value at described overall safety factor W.
The predetermined formula that calculates described safety factor P is: P=K*M.
The predetermined formula that calculates described overall safety factor W is: W=P
1+p
2+p
3+p
4+p
5+.。。。。。+p
n, wherein P
1, P
2, P
3, P
4, P
5,
.。。。。。p
nthe safety factor of respectively corresponding different described relevant informations.
Described network data excavation module Adoption Network reptile is collected the relevant information of Client Enterprise from network.
Described network data excavation module adopts focused crawler to collect the relevant information of Client Enterprise from network.
Described semantic sentiment analysis module adopts sentence level sentiment analysis to carry out emotion constituent analysis to described relevant information.
Described user interactive module comprises: administrative unit, and for Client Enterprise Data Enter, the setting of information search scope, early warning range setting with check Client Enterprise state.
Described administrative unit is the management system of B/S framework.
Described user interactive module comprises: prewarning unit, and for giving a warning during lower than warning value at described overall safety factor W.
Compared with prior art, beneficial effect of the present invention is: due to Warning System after the loan based on semantic sentiment analysis of the present invention, can be automatically successively by network data excavation module, semantic sentiment analysis module, analyzing total module completes the collection of Client Enterprise relevant information, sentiment analysis, and draw the overall safety factor of Client Enterprise, and given a warning automatically by user interactive module in the time that totally safety factor is lower than warning value, therefore reduce manual operation cost, increase work efficiency, so can make early warning to the significant movement of Client Enterprise timely, help better managing customer enterprise of bank, effectively reduce and borrow rear risk.
Brief description of the drawings
Fig. 1 is the structured flowchart of Warning System after the loan based on semantic sentiment analysis of the present invention.
Fig. 2 be semantic sentiment analysis module sentiment analysis shown in Fig. 1 process flow diagram.
Fig. 3 is the process flow diagram of semantic sentiment analysis module sentence level sentiment analysis shown in Fig. 1.
Fig. 4 is the process flow diagram of the work of bulk analysis module shown in Fig. 1.
Embodiment
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with embodiment and accompanying drawing, the present invention is described in further details.At this, exemplary embodiment of the present invention and explanation are used for explaining the present invention, but not as a limitation of the invention.
It shown in Fig. 1, is the structured flowchart of Warning System after the loan based on semantic sentiment analysis of the present invention, as shown in Figure 1, after the loan based on semantic sentiment analysis of the present invention, Warning System comprises: network data excavation module 101, semantic sentiment analysis module 102, analyzing total module 103 and user interactive module 104.Between network data excavation module 101 and semantic sentiment analysis module 102, interconnect; Between semantic sentiment analysis module 102 and analyzing total module 103, interconnect; Between analyzing total module 103 and user interactive module 104, interconnect.
Wherein:
Network data excavation module 101, this network data excavation module 101 is connected with internet, for collect the relevant information of Client Enterprise from network, this relevant information comprises following one or several: news, comment, microblogging, report, the complaint relevant to Client Enterprise;
The relevant informations such as main all and the Client Enterprises that rely on existing web crawlers program to collect and can find on network of network data excavation module 101 are relevant in the time collecting the relevant information of Client Enterprise news, comment, microblogging, report, complaint, then will send to semantic sentiment analysis module 102 after above-mentioned relevant information arrangement;
The web crawlers that network data excavation module 101 is used is known as again webpage spider, network robot or webpage are chased, be a kind of can be according to the computer program of setting the regular automatic capturing network information or procedure script, in addition, according to the search strategy using and the difference of web page analysis algorithm, web crawlers can be divided into universal network reptile, the multiple different type such as focused crawler, in practical application, the data that need due to Warning System after the loan based on semantic sentiment analysis of the present invention are the text message relevant with Client Enterprise, so the scope of data mining can significantly dwindle to improve efficiency and the information real-time of search.In general, news, comments etc. all appear at the website such as portal website, industry forum of main flow conventionally, report, calling information can obtain simply efficiently by the website of government department, Sina's microblogging, everybody, the social network sites of the main flow such as Tengxun also have the high information relevant to Client Enterprise that may occur, if client has transaction on the e-commerce websites such as Taobao, e-commerce website is also the focus of paying close attention to so.So there is very strong specific aim the hunting zone of network data excavation module 101, so focused crawler is the reptile program of first-selection of the present invention.
Semantic sentiment analysis module 102, excavates the market sense constituent analysis of going forward side by side of relevant information that module 101 collects for receiving network data, generates feeling polarities K and emotion intensity M;
Semantic sentiment analysis is emerging Computational Linguistics (computational linguistics) branch, no matter all there is important value in scientific research or in business application, it relates to the fundamental research of the aspects such as computational linguistics, data mining and machine learning, and be in the point of crossing of different subjects, thereby sentiment analysis can promote the development of different subjects, there is important value, it is mainly used in the analysis of emotion composition in natural language, and namely sentiment analysis refers to judge polarity and the intensity of emotion that text is held, viewpoint, attitude.Conventionally according to the difference of text granularity, emotionality analysis is mainly divided into the content of three aspects: word level sentiment analysis (Word-level Sentiment Analysis, WSA), sentence level sentiment analysis (Sentence-level Sentiment Analysis, and chapter level sentiment analysis (Document-level Sentiment Analysis, DSA) SSA).
Sentiment analysis relates to two important elements: feeling polarities and emotion intensity.Feeling polarities refers to the emotion classification that text is corresponding, and feeling polarities is divided into commendation, derogatory sense and objective conventionally; And emotion intensity is the quantitative description to text representation emotion power.After a certain relevant information is carried out to sentiment analysis, we can obtain the value of a feeling polarities and an emotion intensity, for example, shown in Figure 2, semantic sentiment analysis module 102 starts text to be analyzed to carry out sexy analysis receiving after text to be analyzed, and draw commendation, derogatory sense or objective feeling polarities, draw again afterwards commendation rank or derogatory sense rank;
The Client Enterprise relevant information major part of collecting from network due to network data excavation module 101 is all several fragments or simple sentence.So Warning System mainly adopts sentence level sentiment analysis to analyze and classification the Emotional Factors of information after the loan based on semantic sentiment analysis of the present invention.Shown in Figure 3, use the sentiment analysis of sentence level first to need to build emotion sentence sorter, corpus is carried out to pre-service (participle, part-of-speech tagging, the identification of name body and subordinate sentence etc.), and then extract affective characteristics, training emotion classifiers, then predicts sentence feeling polarities.
Analyzing total module 103, for obtaining feeling polarities K and emotion intensity M, and generates feeling polarities K value and emotion intensity M value according to the source of relevant information, calculates successively safety factor P and overall safety factor W afterwards according to predetermined formula; Shown in Figure 4, it is negative value that concrete user in analyzing total module 103 can pre-define K in the time that feeling polarities is derogatory sense, when commendation K be on the occasion of.In the time that K is negative value, the occurrence of K is definite by the source of relevant information, for example: come from the comparison authoritative websites such as government department during when relevant information, the value of k is-3; In the time that relevant information comes from e-commerce website, the value of k is-2; In the time that relevant information comes from social platform, k is-1.When K be on the occasion of time, the occurrence of K is decided by the advertisement putting situation of Client Enterprise and the character of Client Enterprise, for example: when Client Enterprise is electric business website, when the enterprise of the types such as Internet service, the value of K is 0.5; When Client Enterprise is food and drink, when traditional industries that retail etc. can be carried out the publicity of internet to a certain degree, the value of K is 1; In the time of the little conventional industries associated with internet such as Client Enterprise is traditional manufacture, the value of K is 2.Wherein, emotion intensity M is obtained by the emotion intensity rank of analysis word, the comprehensive emotion intensity of statement by semantic sentiment analysis module 102, be different emotion intensity ranks, the corresponding digital value of comprehensive emotion intensity of statement, this digital value can define in advance, and in the time that data enter analyzing total module 103, emotion intensity M has been determined like this.
After determining feeling polarities K value and emotion intensity M value by the way, just can calculate safety factor P according to predetermined formula, predetermined formula can be: P=K*M, user also can set other formula according to actual conditions, the reliability of the Client Enterprise that the current relevant information of measurement that just can be quantitative by safety factor P embodies.The safety factor p afterwards all associated information calculation being drawn adds up, and has just obtained the overall safety factor W of Client Enterprise, i.e. W=P
1+p
2+p
3+p
4+p
5+.。。。。。+p
n, the P here
1, P
2, P
3,
.。。。。。p
nrespectively safety factors corresponding to different relevant informations, when the overall safety factor W of Client Enterprise is during lower than warning value, user interactive module 104 just can give a warning and the overall safety factor W of key monitoring lower than the Client Enterprise of warning value, and the information such as company information, the negative information collected are issued to related personnel or mechanism in the lump.
User interactive module 104, for giving a warning in the time that totally safety factor W is lower than warning value, comprises administration module and a warning module of a B/S framework in it.Administration module is mainly used in Client Enterprise Data Enter, the setting of information search scope, early warning range setting, checks the work such as Client Enterprise state.Warning module can be arranged in the computer of bank clerk as a background service, when there being Client Enterprise to exist when abnormal, warning module can give a warning, and provide some information relevant to this Client Enterprise for reference, warning can not disappear before client unit being made at bank clerk and investigating and give a response, and guarantees that problem Client Enterprise is effectively investigated.
Claims (9)
1. a Warning System after the loan based on semantic sentiment analysis, is characterized in that, this system comprises:
Network data excavation module, for collect the relevant information of Client Enterprise from network, described relevant information comprises following one or several: news, comment, microblogging, report, the complaint relevant to Client Enterprise;
Semantic sentiment analysis module, for receiving the market sense constituent analysis of going forward side by side of described relevant information, generates feeling polarities K and emotion intensity M;
Analyzing total module, for obtaining described feeling polarities K and described emotion intensity M, and generates feeling polarities K value and emotion intensity M value according to the source of described relevant information, calculates successively safety factor P and overall safety factor W afterwards according to predetermined formula;
User interactive module, for giving a warning during lower than warning value at described overall safety factor W.
2. Warning System after the loan based on semantic sentiment analysis according to claim 1, is characterized in that, the predetermined formula that calculates described safety factor P is: P=K*M.
3. Warning System after the loan based on semantic sentiment analysis according to claim 1 and 2, is characterized in that, the predetermined formula that calculates described overall safety factor W is:
, wherein
the safety factor of respectively corresponding different described relevant informations.
4. Warning System after the loan based on semantic sentiment analysis according to claim 3, is characterized in that: described network data excavation module Adoption Network reptile is collected the relevant information of Client Enterprise from network.
5. Warning System after the loan based on semantic sentiment analysis according to claim 3, is characterized in that: described network data excavation module adopts focused crawler to collect the relevant information of Client Enterprise from network.
6. Warning System after the loan based on semantic sentiment analysis according to claim 3, is characterized in that: described semantic sentiment analysis module adopts sentence level sentiment analysis to carry out emotion constituent analysis to described relevant information.
7. Warning System after the loan based on semantic sentiment analysis according to claim 3, is characterized in that: described user interactive module comprises:
Administrative unit, for Client Enterprise Data Enter, the setting of information search scope, early warning range setting with check Client Enterprise state.
8. Warning System after the loan based on semantic sentiment analysis according to claim 7, is characterized in that: described administrative unit is the management system of B/S framework.
9. Warning System after the loan based on semantic sentiment analysis according to claim 3, is characterized in that: described user interactive module comprises:
Prewarning unit, for giving a warning during lower than warning value at described overall safety factor W.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410138443.7A CN103886501B (en) | 2014-04-08 | 2014-04-08 | Post-loan risk early warning system based on semantic sentiment analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410138443.7A CN103886501B (en) | 2014-04-08 | 2014-04-08 | Post-loan risk early warning system based on semantic sentiment analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103886501A true CN103886501A (en) | 2014-06-25 |
| CN103886501B CN103886501B (en) | 2017-04-26 |
Family
ID=50955379
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410138443.7A Active CN103886501B (en) | 2014-04-08 | 2014-04-08 | Post-loan risk early warning system based on semantic sentiment analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103886501B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104616198A (en) * | 2015-02-12 | 2015-05-13 | 哈尔滨工业大学 | P2P (peer-to-peer) network lending risk prediction system based on text analysis |
| CN106530127A (en) * | 2016-11-09 | 2017-03-22 | 国网江苏省电力公司南京供电公司 | Complaint early warning and monitoring analysis system based on text mining |
| CN108062423A (en) * | 2018-01-24 | 2018-05-22 | 北京百度网讯科技有限公司 | Information-pushing method and device |
| CN108197178A (en) * | 2017-12-22 | 2018-06-22 | 国云科技股份有限公司 | A kind of business risk appraisal procedure |
| CN108229806A (en) * | 2017-12-27 | 2018-06-29 | 中国银行股份有限公司 | A kind of method and system for analyzing business risk |
| CN110443236A (en) * | 2019-08-06 | 2019-11-12 | 中国工商银行股份有限公司 | Text will put information extracting method and device after loan |
| TWI719246B (en) * | 2017-08-29 | 2021-02-21 | 彰化商業銀行股份有限公司 | Enterprise customer intelligent risk control system |
| CN115659995A (en) * | 2022-12-30 | 2023-01-31 | 荣耀终端有限公司 | Text emotion analysis method and device |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040267660A1 (en) * | 2003-02-21 | 2004-12-30 | Automated Financial Systems, Inc. | Risk management system |
| CN1916956A (en) * | 2006-09-05 | 2007-02-21 | 中国工商银行股份有限公司 | System and method for quantizing risks of financial assets |
| CN102800018A (en) * | 2012-07-09 | 2012-11-28 | 贵州摇钱树软件开发有限公司 | Credit management system and credit processing method thereof |
-
2014
- 2014-04-08 CN CN201410138443.7A patent/CN103886501B/en active Active
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104616198A (en) * | 2015-02-12 | 2015-05-13 | 哈尔滨工业大学 | P2P (peer-to-peer) network lending risk prediction system based on text analysis |
| CN104616198B (en) * | 2015-02-12 | 2018-01-26 | 哈尔滨工业大学 | A kind of P2P network loan Risk Forecast Systems based on text analyzing |
| CN106530127A (en) * | 2016-11-09 | 2017-03-22 | 国网江苏省电力公司南京供电公司 | Complaint early warning and monitoring analysis system based on text mining |
| TWI719246B (en) * | 2017-08-29 | 2021-02-21 | 彰化商業銀行股份有限公司 | Enterprise customer intelligent risk control system |
| CN108197178A (en) * | 2017-12-22 | 2018-06-22 | 国云科技股份有限公司 | A kind of business risk appraisal procedure |
| CN108229806A (en) * | 2017-12-27 | 2018-06-29 | 中国银行股份有限公司 | A kind of method and system for analyzing business risk |
| CN108062423A (en) * | 2018-01-24 | 2018-05-22 | 北京百度网讯科技有限公司 | Information-pushing method and device |
| CN110443236A (en) * | 2019-08-06 | 2019-11-12 | 中国工商银行股份有限公司 | Text will put information extracting method and device after loan |
| CN110443236B (en) * | 2019-08-06 | 2022-04-29 | 中国工商银行股份有限公司 | Method and device for extracting essential information of post-loan documents |
| CN115659995A (en) * | 2022-12-30 | 2023-01-31 | 荣耀终端有限公司 | Text emotion analysis method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103886501B (en) | 2017-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Neogi et al. | Sentiment analysis and classification of Indian farmers’ protest using twitter data | |
| CN103886501B (en) | Post-loan risk early warning system based on semantic sentiment analysis | |
| Zimbra et al. | The state-of-the-art in Twitter sentiment analysis: A review and benchmark evaluation | |
| Li et al. | Media-aware quantitative trading based on public Web information | |
| CN104137128B (en) | The method and system of green score are generated for using data and mood to analyze | |
| CN104995650B (en) | The method and system of composite index are generated for using the data for being derived from social media and mood analysis | |
| Chen et al. | The determinants of online customer ratings: a combined domain ontology and topic text analytics approach | |
| Patel et al. | A survey of opinion mining and sentiment analysis | |
| Botchway et al. | Deductions from a Sub-Saharan African Bank’s Tweets: A sentiment analysis approach | |
| Kim et al. | Comparing machine learning classifiers for movie WOM opinion mining | |
| CN113449108A (en) | Financial news stream burst detection method based on hierarchical clustering | |
| Sigari et al. | Analyzing the past, improving the future: a multiscale opinion tracking model for optimizing business performance | |
| Suman et al. | Analysis of stock price flow based on social media sentiments | |
| Karmagatri et al. | Uncovering User Perceptions toward Digital Banks in Indonesia: A Naive Bayes Sentiment Analysis of Twitter Data | |
| Ikram et al. | Open source software adoption evaluation through feature level sentiment analysis using Twitter data | |
| Prom-on et al. | DOM: A big data analytics framework for mining Thai public opinions | |
| Kaur et al. | Twitter sentiment analysis of the Indian union budget 2020 | |
| Verma et al. | Knowledge discovery and Twitter sentiment analysis: mining public opinion and studying its correlation with popularity of Indian movies | |
| Rouhani et al. | Sentiment analysis researches story narrated by topic modeling approach | |
| Xia et al. | ICSH 2018: LSTM based sentiment analysis for patient experience narratives in E-survey tools | |
| Chandankhede et al. | ISAR: implicit sentiment analysis of user reviews | |
| Deshmukh et al. | Spin-offs in Indian stock market owing to twitter sentiments, commodity prices and analyst recommendations | |
| Joshi et al. | A new approach to target dependent sentiment analysis with onto-fuzzy logic | |
| Călin et al. | Analysis of Romanian online news articles regarding COVID-19 vaccines using NLP techniques | |
| Kolajo et al. | Sentiment analysis on twitter health news |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |