CN112967139A - Security user quality assessment method based on equipment data - Google Patents
Security user quality assessment method based on equipment data Download PDFInfo
- Publication number
- CN112967139A CN112967139A CN202110312725.4A CN202110312725A CN112967139A CN 112967139 A CN112967139 A CN 112967139A CN 202110312725 A CN202110312725 A CN 202110312725A CN 112967139 A CN112967139 A CN 112967139A
- Authority
- CN
- China
- Prior art keywords
- user
- equipment
- data
- users
- natural
- 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
- 238000001303 quality assessment method Methods 0.000 title description 4
- 238000000034 method Methods 0.000 claims abstract description 24
- 238000012417 linear regression Methods 0.000 claims description 3
- 239000000523 sample Substances 0.000 claims description 3
- 230000014759 maintenance of location Effects 0.000 claims description 2
- 238000009933 burial Methods 0.000 claims 1
- 238000013441 quality evaluation Methods 0.000 abstract 1
- 230000006399 behavior Effects 0.000 description 23
- 230000008569 process Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a security user quality evaluation method based on equipment data, which comprises the following steps: acquiring behavior data and equipment data of user operation securities APP in any channel; establishing an equipment library which comprises equipment attributes of various types of intelligent equipment brands; comparing the collected equipment data with the equipment attributes of the corresponding models in the equipment library, and if the collected equipment data and the equipment attributes of the corresponding models in the equipment library are not matched, identifying the user corresponding to the equipment data as a false user; if the two are matched, identifying the user corresponding to the equipment data as a real user; acquiring behavior data of securities APP operated by natural users, acquiring commission contribution values of the natural users, and establishing a multivariate function of the commission contribution values and the behavior data; calculating commission contribution values of all real users in the channel users; a user quality coefficient of the channel. The method can guide security dealer operators to predict the quality coefficient of a channel in an early stage, and determine the return rate of advertisements and subsequent operation decisions.
Description
Technical Field
The invention belongs to the field of securities user channel quality assessment, and particularly relates to a securities user quality assessment method based on equipment data.
Background
The china securities industry is faced with a comprehensive online approach, where online securities trading accounts for 98% of all securities trades. As competition in the securities industry becomes more severe, one of the core services of each dealer is to acquire new online securities users through channels such as online advertisements, which costs a lot of channel fees. However, some channels have a large number of false accounts. In order to solve the false account opening problem, the inside of a security dealer defines the concept of valid accounts and uses the proportion of the valid accounts to assess the channel quality, wherein the valid accounts refer to accounts which can bind a third party deposit and manage a certain number of security transactions in a certain time.
However, with the development of black and grey products on the Chinese line, under the condition that the securities APP are popularized, the channel cost can be easily cheated by false account opening through a network at a long distance. Because the standard of the valid user is relatively public, the black products can be operated in batch, a large number of true users can be counterfeited, even false securities purchasing and the like can be carried out. Based on this, the traditional method for judging channel quality according to valid account and opening account to prevent fraud can not effectively judge channel quality.
Disclosure of Invention
In order to solve the problems, the invention adopts the client-side non-buried point probe to collect user behavior data and equipment data, the data is irrelevant to service, but the data is high-dimensional data, and the data is not service data but is strongly relevant to true and false of a user. According to the method, channel users are firstly classified into normal users and false users through clustering analysis according to intelligent equipment including mobile phones and tablet computers and user behaviors, and then channel quality of a certain group of users is calculated through multivariate nonlinear regression on a macroscopic scale. The channel in the invention refers to an information source, and the channel users refer to a group of users obtained from the information source, such as users obtained through advertisement of a certain website, ground promotion of a certain teacher, and the like.
In order to achieve the above object, the present invention provides a method for evaluating the quality of securities users based on device data, comprising the steps of:
s1: acquiring behavior data and equipment data of user operation securities APP in any channel, wherein the behavior data comprises n behavior characteristics;
s2, establishing an equipment library, wherein the equipment library comprises equipment attributes of various types of intelligent equipment brands; comparing the device data collected in the step S1 with the device attributes of the corresponding models in the device library, and if the device data and the device attributes are not matched, identifying the user corresponding to the device data as a false user; if the two are matched, identifying the user corresponding to the equipment data as a real user;
s3: collecting behavior data of securities APP operated by natural users, acquiring commission contribution values of the natural users, and establishing a multivariate function Qj＝q1Mj,1,q2Mj,2,...,qnMj,nWherein Q isjJ 1,2, p denotes the commission contribution value of the j-th natural user, p is the number of natural users, Mj,iN denotes the i-th behavior feature of the j-th natural user, qiRepresenting the weight value corresponding to the ith behavior characteristic; calculating and obtaining weight value q by utilizing multiple linear regression method1,q2,...,qn；
S4: calculating commission contribution value Q 'of each real user in channel users'j＝q1mj,1,q2mj,2,...,qnmj,nWherein, Q'jJ 1,2, p 'represents the commission contribution value of the j-th real user, p' is the number of real users, mj,i1,2, n represents the ith behavior characteristic of the jth real user;
s5: and calculating the average commission contribution value of the real user and the average commission contribution value of the natural user of the channel, calculating the ratio of the two values, and obtaining the user quality coefficient of the channel.
Further, the behavior data includes the following behavior characteristics: the method comprises the following steps of using duration, next-day access rate, login/registration ratio, three-day retention, access probability of a certain page, jump probability of a certain page to a second-level page and click rate of common controls.
Further, in step S2, the device attributes include: the device brand, the device model, the CPU core number, the CPU frequency, the ROM storage size, the RAM storage size, the screen size, the system year, month and day, and whether a specific system file directory exists.
Furthermore, a client-side non-buried point probe is adopted to collect behavior data and equipment data of the user operation securities APP in a certain channel and behavior data of the natural user operation securities APP.
Further, in step S3, the time range of collecting the natural user behavior data is one day to seven days, and the commission contribution value of each natural user in one year is obtained.
Further, in step S3, the number of collected natural users is greater than 10000 or occupies 1% or more of all natural users.
The invention has the beneficial effects that:
1) the method can efficiently and accurately identify the false users in the channel users;
2) the method can guide security dealer operators to predict the quality coefficient of a channel in an early stage, and determine the return rate of advertisements and subsequent operation decisions.
Drawings
FIG. 1 is a flow chart of a method for evaluating the quality of a security user based on device data according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings and examples, it being understood that the examples described below are intended to facilitate the understanding of the invention, and are not intended to limit it in any way.
As shown in fig. 1, the method for evaluating the quality of a security user based on device data of the present embodiment includes the following steps:
the first step is as follows: acquiring behavior data and equipment data of a user operating securities APP in a certain channel; the channel users in the present invention refer to a group of users who develop through a specific website or advertisement.
Firstly, a specific plug-in is developed in a program development tool of android and iOS, and all pages and controls in securities APP on a mobile phone can be automatically scanned and automatically numbered through the plug-in the compiling process. The present embodiment encodes ten views from the start to the landing page as a1 to a10, while encoding controls for all pages, buttons, input boxes, etc., the present embodiment encodes the first button of the start page as B1, and up to the "ok" button of the landing page as B120. Then adding a specific function library in the application program, and calling and activating the function library in the initialization process of the mobile phone application program so as to automatically code the page and the control of the application program. In the running process of the application program, the function library is activated, the program package monitors the program after activation, when the program generates events listed in a left column box in the following table 1, the monitoring program calls android and iOS system functions and collects information listed in a right column box in the following table 1, and the information is irrelevant to the privacy of a user.
TABLE 1 event and related information
After the information is collected, the information is temporarily stored in the local computer. And the data is sent to the server once when the storage is full of 20 pieces or the next time the computer is started.
The second step is that: individual man-machine identification;
first, a device library is created, which includes device attributes for each model of each device brand. For a model B of a certain mobile phone brand A, the device attributes can include a CPU model, a CPU core number, a CPU frequency, a ROM storage size, a RAM storage size, a screen size, a system year, month and day, whether a specific system file directory exists or not, and the like. Then, the server compares the received behavior data and the equipment data of the channel user with the equipment attributes in the equipment library, checks whether the hardware information of the mobile phone used by the channel user is matched with the hardware information of the mobile phone with the same model in the equipment library, if not, the mobile phone used by the channel user is a modified machine, the corresponding channel user operation is not a real user, and if the two are matched, the user corresponding to the equipment data is identified as the real user, so that the real user (person) and the modified equipment are identified.
The third step: integrally evaluating channel quality;
first, the method utilizes the extraction in the step oneThe collective method collects behavior data M of a natural user operating a security APP for a period of time (e.g., one day to one week) and calculates trading situations for a longer period of time (one year) to obtain a commission contribution Q of the natural user. The natural user of the present invention refers to a new user who is actively going to a dealer office based on service awareness for the user. In this embodiment, the behavior data M is a behavior feature matrix including 8 dimensions: m1Length of use, M2Access rate of day after day, M3In terms of login/registration ratio, M4Left after three days, M5Access probability for a certain page a, M6Access probability for a certain page B, M7Probability of a jump from a page a to a secondary page C, M8Common control click rate.
Building a multivariate function Q based on collected behavioral data and commission contribution values of natural usersj＝q1Mj,1,q2Mj,2,...,q8Mj,8Wherein Q isjJ 1,2, p represents the commission contribution value of the j-th natural user, and p is the number of natural users, and the collected natural users should be more than 10000 or account for more than 1% of the actual users; mj,i1, 2., 8 denotes the i-th behavior feature of the j-th natural user, qiAnd representing the weight value corresponding to the ith behavior characteristic. Obtaining the weight value q with the minimum error through least square method multiple linear regression training1,q2,...,q8。
Then, the behavior data of the real users identified in the step two is collected by still using the collection method in the step one, and the commission contribution value of each real user and the average commission contribution value of a batch of real users are deduced according to the multivariate function obtained above. Specifically, for some real user, its commission contribution value is Q'j＝q1mj,1,q2mj,2,...,qnmj,nWherein, Q'jJ 1,2, p 'represents the commission contribution value of the j-th real user, p' is the number of real users, mj,iAnd n represents the ith behavior characteristic of the jth real user. Calculate the batchAnd after the commission contribution value of the real user is calculated, further calculating the overall average commission contribution value.
And finally, dividing the average commission contribution value of a batch of real users of the selected channel by the average commission contribution value of the natural users to obtain a user quality coefficient of the selected channel. The coefficient is a coefficient from 0, and by the coefficient, the dealer operator can be guided to predict a channel quality coefficient earlier, and determine the return rate of the advertisement and subsequent operation decision.
It will be apparent to those skilled in the art that various modifications and improvements can be made to the embodiments of the present invention without departing from the inventive concept thereof, and these modifications and improvements are intended to be within the scope of the invention.
Claims (6)
1. A method for evaluating the quality of securities users based on device data is characterized by comprising the following steps:
s1: acquiring behavior data and equipment data of user operation securities APP in any channel, wherein the behavior data comprises n behavior characteristics;
s2, establishing an equipment library, wherein the equipment library comprises equipment attributes of various types of intelligent equipment brands; comparing the device data collected in the step S1 with the device attributes of the corresponding models in the device library, and if the device data and the device attributes are not matched, identifying the user corresponding to the device data as a false user; if the two are matched, identifying the user corresponding to the equipment data as a real user;
s3: collecting behavior data of securities APP operated by natural users, acquiring commission contribution values of the natural users, and establishing a multivariate function Qj＝q1Mj,1,q2Mj,2,...,qnMj,nWherein Q isjJ 1,2, p denotes the commission contribution value of the j-th natural user, p is the number of natural users, Mj,iN denotes the i-th behavior feature of the j-th natural user, qiRepresenting the weight value corresponding to the ith behavior characteristic; calculating and obtaining weight value q by utilizing multiple linear regression method1,q2,...,qn；
S4: calculating commission contribution value Q 'of each real user in channel users'j＝q1mj,1,q2mj,2,...,qnmj,nWherein, Q'jJ 1,2, p 'represents the commission contribution value of the j-th real user, p' is the number of real users, mj,i1,2, n represents the ith behavior characteristic of the jth real user;
s5: and calculating the average commission contribution value of the real user and the average commission contribution value of the natural user of the channel, calculating the ratio of the two values, and obtaining the user quality coefficient of the channel.
2. The method of claim 1, wherein the behavior data comprises the following behavior characteristics: the method comprises the following steps of using duration, next-day access rate, login/registration ratio, three-day retention, access probability of a certain page, jump probability of a certain page to a second-level page and click rate of common controls.
3. The method according to claim 1, wherein in step S2, the device attributes include: the device brand, the device model, the CPU core number, the CPU frequency, the ROM storage size, the RAM storage size, the screen size, the system year, month and day, and whether a specific system file directory exists.
4. Method according to one of claims 1 to 3, characterized in that the client side burial point-free probe is used to collect the behavior data and equipment data of the user operated securities APP and the behavior data of the natural user operated securities APP in a certain channel.
5. The method according to any one of claims 1 to 3, wherein in step S3, the time for collecting the natural user behavior data ranges from one day to seven days, and the commission contribution value of each natural user in one year is obtained.
6. The method according to any one of claims 1 to 3, wherein in step S3, the number of collected natural users is greater than 10000 or more than 1% of all natural users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110312725.4A CN112967139A (en) | 2021-03-24 | 2021-03-24 | Security user quality assessment method based on equipment data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110312725.4A CN112967139A (en) | 2021-03-24 | 2021-03-24 | Security user quality assessment method based on equipment data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112967139A true CN112967139A (en) | 2021-06-15 |
Family
ID=76279391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110312725.4A Pending CN112967139A (en) | 2021-03-24 | 2021-03-24 | Security user quality assessment method based on equipment data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112967139A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102710770A (en) * | 2012-06-01 | 2012-10-03 | 汪德嘉 | Identification method for network access equipment and implementation system for identification method |
| CN107133734A (en) * | 2017-04-28 | 2017-09-05 | 浙江极赢信息技术有限公司 | A kind of Channel Quality evaluation method and system |
| CN109711129A (en) * | 2018-12-15 | 2019-05-03 | 深圳壹账通智能科技有限公司 | Login validation method, device, equipment and storage medium based on Application on Voiceprint Recognition |
| CN109784095A (en) * | 2018-11-29 | 2019-05-21 | 武汉极意网络科技有限公司 | A kind of user equipment model authenticating method and system |
| CN110691090A (en) * | 2019-09-29 | 2020-01-14 | 武汉极意网络科技有限公司 | Website detection method, device, equipment and storage medium |
| CN111612366A (en) * | 2020-05-27 | 2020-09-01 | 中国联合网络通信集团有限公司 | Channel quality evaluation method and device, electronic equipment and storage medium |
| CN112235564A (en) * | 2020-09-10 | 2021-01-15 | 当趣网络科技(杭州)有限公司 | Data processing method and device based on delivery channel |
| CN112487376A (en) * | 2020-12-07 | 2021-03-12 | 北京明略昭辉科技有限公司 | Man-machine verification method and device |
-
2021
- 2021-03-24 CN CN202110312725.4A patent/CN112967139A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102710770A (en) * | 2012-06-01 | 2012-10-03 | 汪德嘉 | Identification method for network access equipment and implementation system for identification method |
| CN107133734A (en) * | 2017-04-28 | 2017-09-05 | 浙江极赢信息技术有限公司 | A kind of Channel Quality evaluation method and system |
| CN109784095A (en) * | 2018-11-29 | 2019-05-21 | 武汉极意网络科技有限公司 | A kind of user equipment model authenticating method and system |
| CN109711129A (en) * | 2018-12-15 | 2019-05-03 | 深圳壹账通智能科技有限公司 | Login validation method, device, equipment and storage medium based on Application on Voiceprint Recognition |
| CN110691090A (en) * | 2019-09-29 | 2020-01-14 | 武汉极意网络科技有限公司 | Website detection method, device, equipment and storage medium |
| CN111612366A (en) * | 2020-05-27 | 2020-09-01 | 中国联合网络通信集团有限公司 | Channel quality evaluation method and device, electronic equipment and storage medium |
| CN112235564A (en) * | 2020-09-10 | 2021-01-15 | 当趣网络科技(杭州)有限公司 | Data processing method and device based on delivery channel |
| CN112487376A (en) * | 2020-12-07 | 2021-03-12 | 北京明略昭辉科技有限公司 | Man-machine verification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rashidi et al. | Subsistence over symbolism: the role of transnational municipal networks on cities’ climate policy innovation and adoption | |
| US7200607B2 (en) | Data analysis system for creating a comparative profile report | |
| US20020143562A1 (en) | Automated legal action risk management | |
| CN111861732A (en) | Risk assessment system and method | |
| CN101194286A (en) | Risk based data assessment | |
| CN104834983A (en) | Business data processing method and device | |
| CA2927849A1 (en) | Intelligent collections models | |
| CN111383094A (en) | Product service full-chain driving method, equipment and readable storage medium | |
| CN111222784A (en) | Security monitoring method and system based on population big data | |
| CN112907356A (en) | Overdue collection method, device and system and computer readable storage medium | |
| CN111401447A (en) | Artificial intelligence-based flow cheating identification method and device and electronic equipment | |
| CN110689443A (en) | Insurance data processing method and device, storage medium and server | |
| CN106713267A (en) | Network security assessment method and system | |
| CN111611519A (en) | Method and device for detecting personal abnormal behaviors | |
| CN112000862B (en) | Data processing method and device | |
| CN111582757B (en) | Method, device, equipment and computer readable storage medium for analyzing fraud risk | |
| CN111008887B (en) | Skill service bidding system | |
| CN112967139A (en) | Security user quality assessment method based on equipment data | |
| KR20130083053A (en) | Warning system for bad project of overseas constructions | |
| CN112712270B (en) | Information processing method, device, equipment and storage medium | |
| CN113285970B (en) | Information answering platform based on digital television and mobile phone terminal | |
| CN116112273B (en) | Anti-fraud man-machine identification method and system | |
| CN111488539B (en) | Page adjustment method and device | |
| KR20070035659A (en) | Brand management service method | |
| CN116932628A (en) | Flow main detection method and device and computer readable 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 |