CN108229749A - Bad booking behavior management method based on deep learning - Google Patents
Bad booking behavior management method based on deep learning Download PDFInfo
- Publication number
- CN108229749A CN108229749A CN201810040220.5A CN201810040220A CN108229749A CN 108229749 A CN108229749 A CN 108229749A CN 201810040220 A CN201810040220 A CN 201810040220A CN 108229749 A CN108229749 A CN 108229749A
- Authority
- CN
- China
- Prior art keywords
- booking
- bad
- user
- behavior
- booking behavior
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06Q30/01—Customer relationship services
- G06Q30/012—Providing warranty services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of bad booking behavior management methods based on deep learning, including, establish model, step S2, training pattern, real-time data acquisition, according to the customer transaction behavioral data of acquisition and trained model analysis, predict whether the current booking behavior of user is bad booking behavior, if the current booking behavior of user is handled for bad booking behavior into bad booking behavior, otherwise it is normal booking behavior, the bad booking behavior management method of the present invention, actively discover network brush ticket behavior, and it is handled automatically, abnormal user is piped off, forbid black list user's booking, safeguard the just operation of ticket-booking system.
Description
Technical field
The present invention relates to booking technical fields, and in particular to a kind of bad booking behavior management side based on deep learning
Method.
Background technology
With popularizing for internet electronic business, various network booking channels and platform greatly facilitate people’s lives
Mode.Presell pattern is generally taken in network booking, and the presell phase is all longer, and can return ticket at any time.As train ticket, air ticket
The period is subscribed up to several days, and can freely return ticket except time restriction and not generate service charge.
The personnel such as many third-party agent mechanisms and ox are consequently led to, are occupied very much greatly using the loophole of ticket sale system
Measure the supply of tickets.Generally systematic automatic operation holds the popular supply of tickets to when closing on charge when just putting ticket always with regard to all buying
It returns ticket again, so as to achieve the purpose that zero cost controls a large amount of supply of tickets.Inconvenience is brought to the people of normal need purchase ticket, is also given
Operating agency brings unfavorable factor, such as can not correctly estimate the target group's quantity for really having demand, it is possible to sell out
Ticket can largely be retracted suddenly in some node.
Invention content
The purpose of the present invention overcomes prior art problem, proposes a kind of bad booking behavior management side based on deep learning
Method, the present invention adopt the following technical scheme that：
A kind of bad booking behavior management method based on deep learning, includes the following steps：
Step S1, establishes model：
Model is established using the recurrence of SVM classifier, Bayes classifier and random forest；
Step S2, training pattern：
User existing in ticket sale system and its booking record and its ticket satellite information are collected, arranged, and to mould
Type is trained；The booking record and its satellite information for the black list user that ticket sale system has been identified or has been shielded also are received
Collection, model training；The ticket satellite information is including in booking time, booking place, terminal type, quantity, payment information, booking
Hold；
Step S3, real-time data acquisition：
Acquisition customer transaction behavioral data in real time, the customer transaction behavioral data include but not limited to user identity letter
Breath, booking time, booking place, terminal type, booking content；
The terminal type is computer, mobile phone or operating system；
The booking content includes the section of ticket, time；
Step S4, user's behavior prediction：
According to the customer transaction behavioral data of acquisition and trained model analysis, the current booking behavior of prediction user whether be
Bad booking behavior, if the current booking behavior of user enters step the bad booking behavior processing of S5 for bad booking behavior, otherwise
For normal booking behavior.
Further, the subscriber identity information includes gender, age, area, phone.
Wherein, the bad booking behavior processing of step S5, includes the following steps：
Step S51 judges the user type of current bad booking behavior user,
If current bad booking behavior user is old user, S52 is entered step, if current bad booking behavior user is
New user, then currently bad booking behavior user gives warning label；
Step S52, it is current bad according to the booking behavior processing before the current booking of current bad booking behavior user
Booking behavior：
If current bad booking behavior user's mark is is potential bad user for the first time by current bad booking behavior；
If current bad booking behavior has been labeled for the bad booking behavior of n-th or current bad booking behavior user
For potential bad user, then currently bad booking behavior user gives warning label；
If current bad booking behavior is warned for the N+M times bad booking behavior or current bad booking behavior user
Label is accused, then current bad booking behavior user's mark is black list user.
In embodiments of the present invention, N=2, M=1.It should be noted that the value of N, M are not limited to the embodiment of the present invention,
Model errors judge in order to prevent, apply releasing blacklist label with reference to black list user, further analyze, can suitably adjust
M, N correlations.
In order to safeguard justice, in the method for the present invention, the registered user of blacklist is added into, booking can not be carried out again.
In order to safeguard justice, in the method for the present invention, ticket-booking system will by force return ticket the booking of this user, and using
Family behavioral data is added in black list database, is used for model training.
In order to safeguard justice, in the method for the present invention, the user that pipes off applies releasing, and need to manually be audited.
Further, if the user that pipes off does not apply releasing blacklist, it is black list user to confirm this user.
By the bad booking behavior management method based on deep learning of the present invention, the present invention is deep by establishing model
Degree study a large number of users data, so as to which whether the behavior for judge user is bad user, and the corresponding countermeasure of conduct, it can be with
Reach following advantageous effect：
(1) user existing in ticket sale system and its booking record are subjected to model training, acquire user's booking number in real time
According to, and whether model prediction user booking behavior is abnormal after training, actively discovers network brush ticket behavior, and located automatically
Reason is potential bad user, warning, blacklist, improves working efficiency；
(2) forbid black list user's booking, while allow to apply for that manual examination and verification release contact blacklist label, ensure network
The fairness of ticket-booking system and the public praise of ticketing company；
(3) ticketing company grasps truthful data, would be much more convenient and carries out the relevant resource allocation of enterprise operation (such as increase and decrease vehicle
Secondary, adjustment time etc.).
Description of the drawings
Attached drawing described herein is used for providing further understanding invention, forms the part of the present invention, the present invention
Illustrative embodiments and their description for explaining the present invention, do not constitute improper limitations of the present invention.In the accompanying drawings：
Fig. 1 is the flow diagram of the bad booking behavior management method the present invention is based on deep learning；
Fig. 2 is the flow diagram of bad booking behavior processing in the method for the present invention.
Specific embodiment
In order to make technical problems, technical solutions and advantages to be solved clearer, clear, tie below
Drawings and examples are closed, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only used
To explain the present invention, it is not intended to limit the present invention.
The bad booking behavior management method based on deep learning of the present invention, as shown in Figure 1, includes the following steps：
Step S1, establishes model：
Model is established using the recurrence of SVM classifier, Bayes classifier and random forest；
Step S2, training pattern：
User existing in ticket sale system and its booking record and its ticket satellite information are collected, arranged, and to mould
Type is trained；The booking record and its satellite information for the black list user that ticket sale system has been identified or has been shielded also are received
Collection, model training；The ticket satellite information is including in booking time, booking place, terminal type, quantity, payment information, booking
Hold；
Step S3, real-time data acquisition：
Acquisition customer transaction behavioral data in real time, the customer transaction behavioral data include but not limited to user identity letter
Breath, booking time, booking place, terminal type, booking content；
The subscriber identity information includes gender, age, area, phone；
The terminal type is computer, mobile phone or operating system；
The booking content includes the section of ticket, time；
Step S4, user's behavior prediction：
According to the customer transaction behavioral data of acquisition and trained model analysis, the current booking behavior of prediction user whether be
Bad booking behavior, if the current booking behavior of user enters step the bad booking behavior processing of S5 for bad booking behavior, otherwise
For normal booking behavior.
It should be noted that in the present invention, the ticket satellite information can be configured according to actual needs, can be increased
It can also reduce.Equally, acquisition customer transaction behavioral data is also not necessarily limited to cited by the present embodiment in real time, can be more than this hair
It is bright cited, it can also be less than cited by the present invention.
The bad booking behavior processing of step S5, includes the following steps：
Step S51 judges the user type of current bad booking behavior user,
If current bad booking behavior user is old user, S52 is entered step, if current bad booking behavior user is
New user, then currently bad booking behavior user gives warning label；
Step S52, it is current bad according to the booking behavior processing before the current booking of current bad booking behavior user
Booking behavior：
If current bad booking behavior user's mark is is potential bad user for the first time by current bad booking behavior；
If current bad booking behavior has been labeled for the 2nd bad booking behavior or current bad booking behavior user
For potential bad user, then currently bad booking behavior user gives warning label；
If current bad booking behavior is warned for the 2+1 times bad booking behavior or current bad booking behavior user
Label is accused, then current bad booking behavior user's mark is black list user；
Further, the registered user of blacklist is added into, booking can not be carried out again.
Further, ticket-booking system will by force return ticket the booking of this user, and user behavior data is added to
In black list database, used for model training.
Further, the user that pipes off applies releasing, and need to manually be audited.
Further, if user does not apply releasing blacklist, it is black list user to confirm this user.
Description above describe the preferred embodiment of the present invention, it is to be understood that the present invention is not limited to above-mentioned implementation
Example, and the exclusion to other embodiment should not be regarded as.By the enlightenment of the present invention, those skilled in the art combine known or existing
The change that technology, knowledge are carried out also should be regarded as within the scope of the present invention.
Claims (8)
- A kind of 1. bad booking behavior management method based on deep learning, which is characterized in that include the following steps：Step S1, establishes model：Model is established using the recurrence of SVM classifier, Bayes classifier and random forest；Step S2, training pattern：User existing in ticket sale system and its booking record and its ticket satellite information are collected, arrange, and to model into Row training；The booking record and its satellite information for the black list user that ticket sale system has been identified or has been shielded also be collected, Model training；The ticket satellite information includes booking time, booking place, terminal type, quantity, payment information, booking content；Step S3, real-time data acquisition：Acquisition customer transaction behavioral data in real time, the customer transaction behavioral data include but not limited to subscriber identity information, purchase Ticket time, booking place, terminal type, booking content；The terminal type is computer, mobile phone or operating system；The booking content includes the section of ticket, time；Step S4, user's behavior prediction：According to the customer transaction behavioral data of acquisition and trained model analysis, whether the prediction current booking behavior of user is bad Booking behavior, if the current booking behavior of user enters step the bad booking behavior processing of S5 for bad booking behavior, otherwise for just Normal booking behavior.
- 2. the bad booking behavior management method according to claim 1 based on deep learning, which is characterized in thatThe subscriber identity information includes gender, age, area, phone.
- 3. the bad booking behavior management method according to claim 1 based on deep learning, which is characterized in thatThe bad booking behavior processing of step S5, includes the following steps：Step S51 judges the user type of current bad booking behavior user,If current bad booking behavior user is old user, S52 is entered step, if current bad booking behavior user uses to be new Family, then currently bad booking behavior user gives warning label；Step S52, according to the current bad booking of booking behavior processing before the current booking of current bad booking behavior user Behavior；If current bad booking behavior user's mark is is potential bad user for the first time by current bad booking behavior；If current bad booking behavior has been marked as diving for the bad booking behavior of n-th or current bad booking behavior user In bad user, then currently bad booking behavior user gives warning label；If current bad booking behavior has been warned mark for the N+M times bad booking behavior or current bad booking behavior user Note, then current bad booking behavior user's mark is black list user.
- 4. the bad booking behavior management method according to claim 1 based on deep learning, which is characterized in thatN=2, M=1.
- 5. the bad booking behavior management method according to claim 1 based on deep learning, which is characterized in thatThe registered user of blacklist is added into, booking can not be carried out again.
- 6. the bad booking behavior management method according to claim 1 based on deep learning, which is characterized in thatTicket-booking system will by force return ticket the booking of this user, and user behavior data is added to black list database In, it is used for model training.
- 7. the bad booking behavior management method according to claim 1 based on deep learning, which is characterized in thatThe user that pipes off applies releasing, and need to manually be audited.
- 8. the bad booking behavior management method according to claim 1 based on deep learning, which is characterized in thatIf the user that pipes off does not apply releasing blacklist, it is black list user to confirm this user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810040220.5A CN108229749A (en) | 2018-01-16 | 2018-01-16 | Bad booking behavior management method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810040220.5A CN108229749A (en) | 2018-01-16 | 2018-01-16 | Bad booking behavior management method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108229749A true CN108229749A (en) | 2018-06-29 |
Family
ID=62640488
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810040220.5A Pending CN108229749A (en) | 2018-01-16 | 2018-01-16 | Bad booking behavior management method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108229749A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109377301A (en) * | 2018-08-27 | 2019-02-22 | 中国民航信息网络股份有限公司 | A kind of Feature Extraction Method based on Airline reservation behavioral data |
| CN110310407A (en) * | 2019-06-05 | 2019-10-08 | 上海车轮互联网服务有限公司 | Anti- brush ticket method and device based on user behavior monitoring |
| CN110675228A (en) * | 2019-09-27 | 2020-01-10 | 支付宝(杭州)信息技术有限公司 | User ticket buying behavior detection method and device |
| CN110751536A (en) * | 2019-09-28 | 2020-02-04 | 同程网络科技股份有限公司 | Risk control method and system |
| CN111147441A (en) * | 2019-11-12 | 2020-05-12 | 恒大智慧科技有限公司 | Method and device for automatically detecting fraud behaviors of online ticket purchasing and readable storage medium |
| CN111435507A (en) * | 2019-01-11 | 2020-07-21 | 腾讯科技（北京）有限公司 | Advertisement anti-cheating method and device, electronic equipment and readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103279868A (en) * | 2013-05-22 | 2013-09-04 | 兰亭集势有限公司 | Method and device for automatically identifying fraud order form |
| CN106779126A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | Malice accounts for the processing method and system of an order |
| US20170206462A1 (en) * | 2016-01-14 | 2017-07-20 | International Business Machines Corporation | Method and apparatus for detecting abnormal contention on a computer system |
-
2018
- 2018-01-16 CN CN201810040220.5A patent/CN108229749A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103279868A (en) * | 2013-05-22 | 2013-09-04 | 兰亭集势有限公司 | Method and device for automatically identifying fraud order form |
| US20170206462A1 (en) * | 2016-01-14 | 2017-07-20 | International Business Machines Corporation | Method and apparatus for detecting abnormal contention on a computer system |
| CN106779126A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | Malice accounts for the processing method and system of an order |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109377301A (en) * | 2018-08-27 | 2019-02-22 | 中国民航信息网络股份有限公司 | A kind of Feature Extraction Method based on Airline reservation behavioral data |
| CN111435507A (en) * | 2019-01-11 | 2020-07-21 | 腾讯科技（北京）有限公司 | Advertisement anti-cheating method and device, electronic equipment and readable storage medium |
| CN110310407A (en) * | 2019-06-05 | 2019-10-08 | 上海车轮互联网服务有限公司 | Anti- brush ticket method and device based on user behavior monitoring |
| CN110675228A (en) * | 2019-09-27 | 2020-01-10 | 支付宝(杭州)信息技术有限公司 | User ticket buying behavior detection method and device |
| TWI740507B (en) * | 2019-09-27 | 2021-09-21 | 大陸商支付寶（杭州）信息技術有限公司 | Method and device for detecting ticket purchase behavior of user |
| CN110751536A (en) * | 2019-09-28 | 2020-02-04 | 同程网络科技股份有限公司 | Risk control method and system |
| CN111147441A (en) * | 2019-11-12 | 2020-05-12 | 恒大智慧科技有限公司 | Method and device for automatically detecting fraud behaviors of online ticket purchasing and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108229749A (en) | Bad booking behavior management method based on deep learning | |
| Plantinga et al. | Agricultural land values and the value of rights to future land development | |
| Gillenwater | What is additionality | |
| CN108932585B (en) | Merchant operation management method and equipment, storage medium and electronic equipment thereof | |
| CN104834983B (en) | Business data processing method and device | |
| CN107392755A (en) | Credit risk merges appraisal procedure and system | |
| CN108985347A (en) | Training method, the method and device of shop classification of disaggregated model | |
| CN108133013A (en) | Information processing method, device, computer equipment and storage medium | |
| CN112529716B (en) | Method, device and computer readable storage medium for predicting credit | |
| CN110503565A (en) | Behaviorist risk recognition methods, system, equipment and readable storage medium storing program for executing | |
| CN107895324A (en) | Insurance examination ＆ verification apparatus and method | |
| Gupta et al. | Adoption of mobile telephony in rural India: An empirical study | |
| CN109064357B (en) | Real-time tour guide recommendation system and method | |
| CN108648091A (en) | Declaration form methods of risk assessment, device, equipment and computer storage media | |
| CN105868878A (en) | Method and device for MAC (Media Access Control) address risk identification | |
| CN107871203A (en) | Business personnel's behaviorist risk screens management method, application server and computer-readable recording medium | |
| CN109150894A (en) | A kind of method and system for identifying malicious user | |
| Mason et al. | The application of a system of simultaneous equations to an innovation diffusion model | |
| CN109492863A (en) | The automatic generation method and device of financial document | |
| CN108776913A (en) | Invoice source tracing method and device | |
| CN110751395A (en) | Passenger travel state determination method, device and server | |
| CN110457601A (en) | The recognition methods and device of social account, storage medium and electronic device | |
| CN109726931A (en) | Seat manages method, apparatus, computer equipment and storage medium | |
| CN113535848A (en) | Block chain-based credit investigation grade determination method, device, equipment and storage medium | |
| CN113743816A (en) | Vehicle rental risk control method and device, computer equipment and storage medium |
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: 20180629 |