CN110210507A - Detection method, device and the readable storage medium storing program for executing that machine is clicked - Google Patents

Detection method, device and the readable storage medium storing program for executing that machine is clicked Download PDF

Info

Publication number
CN110210507A
CN110210507A CN201811265801.5A CN201811265801A CN110210507A CN 110210507 A CN110210507 A CN 110210507A CN 201811265801 A CN201811265801 A CN 201811265801A CN 110210507 A CN110210507 A CN 110210507A
Authority
CN
China
Prior art keywords
click
vector
clicked
target type
type content
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
Application number
CN201811265801.5A
Other languages
Chinese (zh)
Other versions
CN110210507B (en
Inventor
翁家才
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN201811265801.5A priority Critical patent/CN110210507B/en
Publication of CN110210507A publication Critical patent/CN110210507A/en
Application granted granted Critical
Publication of CN110210507B publication Critical patent/CN110210507B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • 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
    • G06Q30/0241—Advertisements
    • G06Q30/0242—Determining effectiveness of advertisements

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Information Transfer Between Computers (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

This application discloses detection method, device and readable storage medium storing program for executing that a kind of machine is clicked, are related to field of computer technology.This method comprises: obtaining the click coordinate data of n target type content；Determine i-th vector and object vector to be detected；When the vector distance of i-th of vector sum object vector to be detected is greater than first distance threshold value, determine that i-th of target type content is to receive the content of machine click.By determining object vector and i-th of vector to be detected, wherein, object vector represent manually to can click on area click when click feature, i-th of vector to be detected is used to represent the click feature of i-th of target type content, by comparing the similarity between i-th of vector sum object vector to be detected, when vector distance is greater than threshold value, then it is assumed that i-th of target type content is to receive the content of machine click, improves the recognition accuracy clicked to machine.

Description

Detection method, device and the readable storage medium storing program for executing that machine is clicked
Technical field
The invention relates to field of computer technology, in particular to detection method that a kind of machine is clicked, device and Readable storage medium storing program for executing.
Background technique
Each pay-per-click (Cost Per Click, CPC) advertisement refer to product supplier according to the click volume received, Advertisement to the publication account number payt of releasing advertisements.User can in the user interface of the terminal check advertisement, And in advertisement can be after click on area clicks, the detailed content into the corresponding commodity of advertisement introduces interface, and user The number clicked in advertisement is more, issues the report that the publication account number of the advertisement obtains from the corresponding product supplier of the advertisement Reward also can be accordingly higher.
There are the feelings for the click volume for increasing advertisement in such a way that machine is clicked to obtain higher remuneration in publication account number Condition, wherein machine click refer to by automatized script or software manipulate multiple electronic equipment to can click on area carry out it is empty It is quasi- to click, and machine click be usually for can the same coordinate in click on area repeatedly clicked.In the related technology, it takes Business device is by determining the advertisement when detecting advertisement with the presence or absence of the case where increasing click volume in such a way that machine is clicked The entropy that the thermodynamic chart is calculated after thermodynamic chart indicates the area distribution that the advertisement is clicked when the entropy being calculated is higher than threshold value Excessively disperse or excessively concentrate, it is determined that the click volume of the advertisement is with the presence or absence of cheating situation, wherein the thermodynamic chart of the advertisement The quantity for the click event that each region for indicating in advertisement receives.
However, being only capable of indicating the advertisement by point since the entropy of thermodynamic chart is higher than threshold value when being detected through the above way The position hit is distributed in advertisement excessively to be dispersed or excessively concentrates (such as: concentrating on some seat that can be red in click on area Mark), and can not be clicked according to machine and it is artificial click between difference be judged, when the mode that machine is clicked changes, Such as: with certain rule when that can be clicked in click on area, then according to the entropy of thermodynamic chart be easy to cause misjudgment rate compared with High problem.
Summary of the invention
The embodiment of the present application provides a kind of detection method, device and readable storage medium storing program for executing that machine is clicked, and can solve The higher problem of misjudgment rate is easy to cause according to the entropy of thermodynamic chart.The technical solution is as follows:
In a first aspect, providing a kind of detection method that machine is clicked, which comprises
Obtain n target type content click coordinate data, the n target type content be n same type and Including can click on area content, the click coordinate data are that the click event that can receive in click on area according to is raw At data；
I-th of vector to be detected is determined according to the click coordinate data of i-th of target type content, described i-th Vector to be detected be used to indicate i-th of target type content can at least one pixel in click on area be clicked Number, 0 < i≤n；
Determine that object vector, the object vector are used for table according to the click coordinate data of the n target type content That shows the n target type content can the total degree that is clicked of at least one corresponding pixel of click on area；
When the vector distance of the object vector described in described i-th vector sum to be detected is greater than first distance threshold value, determine I-th of target type content is the content for receiving the machine and clicking.
On the other hand, a kind of detection device that machine is clicked is provided, described device includes:
Module is obtained, for obtaining the click coordinate data of n target type content, the n target type content is n A same type and including can click on area content, the click coordinate data can be received in click on area according to Click event generate data；
Determining module, for determining i-th of direction finding to be checked according to the click coordinate data of i-th of target type content Amount, what i-th of vector to be detected was used to indicate i-th of target type content can at least one of click on area picture The number that vegetarian refreshments is clicked, 0 < i≤n；
The determining module is also used to determine object vector according to the click coordinate data of the n target type content, The object vector be used for indicate the n target type content can at least one corresponding pixel of click on area be clicked Total degree；
Judgment module, the vector distance for the object vector described in described i-th vector sum to be detected be greater than first away from When from threshold value, determine that i-th of target type content is the content for receiving the machine and clicking.
On the other hand, a kind of server is provided, the server includes processor and memory, is deposited in the memory Contain at least one instruction, at least a Duan Chengxu, code set or instruction set, at least one instruction, an at least Duan Cheng Sequence, the code set or instruction set are loaded as the processor and are executed with the machine as described in above-mentioned the embodiment of the present application of realization The detection method that device is clicked.
On the other hand, a kind of computer readable storage medium is provided, at least one finger is stored in the storage medium Enable, at least a Duan Chengxu, code set or instruction set, at least one instruction, an at least Duan Chengxu, the code set or Instruction set is loaded as the processor and is executed the detection side to realize the click of the machine as described in above-mentioned the embodiment of the present application Method.
On the other hand, a kind of computer program product is provided, when the computer program product is run on computers When, so that computer executes the detection method that the machine as described in above-mentioned the embodiment of the present application is clicked.
Technical solution bring beneficial effect provided by the embodiments of the present application includes at least:
Object vector is determined by the click coordinate data of n target type content, and the object vector is artificial for representing To can click on area click when click feature, by the click data of i-th of target type content determine i-th it is to be checked Direction finding amount, the click that this i-th vector to be detected is used to represent the click event received in i-th of target type content are special Sign by comparing the vector distance between i-th of vector sum object vector to be detected, namely compares i-th of vector sum to be detected Similarity between object vector then illustrates the phase of i-th vector to be detected and object vector when vector distance is greater than threshold value It is lower like spending, namely the click feature of click event that is received in i-th of target type content of explanation with manually to can click Click feature similarity when region is clicked is lower, then it is assumed that i-th of target type content is to receive machine click Content, improve to machine click recognition accuracy.
Detailed description of the invention
In order to more clearly explain the technical solutions in the embodiments of the present application, make required in being described below to embodiment Attached drawing is briefly described, it should be apparent that, the drawings in the following description are only some examples of the present application, for For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other Attached drawing.
Fig. 1 is the implementation environment schematic diagram that the machine that one exemplary embodiment of the application provides clicks detection system；
Fig. 2 is the flow chart for the detection method that the machine that one exemplary embodiment of the application provides is clicked；
Fig. 3 be based on Fig. 2 shows embodiment provide by can click on area map to the schematic diagram of coordinate system；
Fig. 4 is the flow chart for the detection method that the machine that another exemplary embodiment of the application provides is clicked；
Fig. 5 is the flow chart for the detection method that the machine that another exemplary embodiment of the application provides is clicked；
Fig. 6 is the point for receiving the recommendation information that same type of machine is clicked provided based on the embodiment shown in Fig. 5 Hit coordinate distribution schematic diagram；
Fig. 7 is the user interface schematic diagram to polling option that one exemplary embodiment of the application provides；
Fig. 8 is the flow chart for the detection method that the machine that another exemplary embodiment of the application provides is clicked；
Fig. 9 is the structural block diagram for the detection device that the machine that one exemplary embodiment of the application provides is clicked；
Figure 10 is the structural block diagram for the detection device that the machine that another exemplary embodiment of the application provides is clicked；
Figure 11 is the structural block diagram for the server that one exemplary embodiment of the application provides.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with attached drawing to the application embodiment party Formula is described in further detail.
Firstly, simply being introduced noun involved in the application:
Thermodynamic chart: refer to and show corresponding some the different degrees of event of different zones in highlighted form.In the application In embodiment, which refers to the quantity that clicks event that each region receives is shown in highlighted form, or, to highlight Form show the quantity of click event that each pixel receives.Schematically, can in click on area, coordinate (10, 11) number that the pixel indicated is clicked is 20 times, and the number that the pixel that coordinate (10,1) indicates is clicked is 1 time, is sat The number that is clicked of pixel that mark (14,5) indicates is 8 times, then this can be in the thermodynamic chart of click on area, coordinate (10,11) It is shown in red at the pixel of expression, yellow is shown as at the pixel that coordinate (14,5) indicates, what coordinate (10,1) indicated Blue is shown as at pixel.
Machine is clicked: refer to by automatized script or software manipulate multiple electronic equipment to can click on area carry out a little Hit, that is, without manually to can click on area click, can to this can the click volume of click on area increase.By certainly Dynamicization script or software operation electronic equipment to can click on area click when, since automatized script or software can not all be done To random click truly, it will usually there is certain rule, such as: one click laterally is carried out every pre-determined distance, or, It is repeatedly clicked at some coordinate points.And when manually clicking, user would generally click on most to be convenient for touching apart from thumb Position, such as: can the number that is clicked of the position in click on area close to the lower right corner it is more, and time that other positions are clicked Number is more scattered.
Secondly, schematical, this application involves application scenarios include at least one of following scene:
The first: is after user can click the recommendation information issued on public platform, to the detailed of the recommendation information Carefully information is checked, and when the recommendation information is CPC recommendation information, after user clicks on recommendation information, the click thing Part will increase the click volume received on recommendation information, and the publication account number for issuing the recommendation information is received according on recommendation information Click volume obtain remuneration from the corresponding product supplier of the recommendation information, so, issue account number to obtain higher report There is the case where increasing the click volume of recommendation information in such a way that machine is clicked in reward；
Second: user can click to ballot activity is participated in polling option, the click event will increase to The votes received in polling option, and the triumph that the ballot activity can be then obtained to polling option that votes are most, institute To win the triumph of the ballot activity to the corresponding account number of polling option, exist and increase ballot by way of machine click Several situations；
The third: public's account number is the account number that is certified and can be concerned on public platform, and user is to public's account number After carrying out the bean vermicelli that concern becomes public's account number, the content in public platform can be issued to public's account number and looked into It sees, and the number of fans of public's account number can also be increase accordingly, since the number of fans of public's account number is more, generated influence power It is bigger, so, there is the case where increasing number of fans by way of machine click to increase number of fans in public's account number.
It is worth noting that, above-mentioned three kinds of application scenarios are all schematical citing, in actual operation, according to click thing The quantity of part generates the detection side that the application scenarios of subsequent affect can use the machine provided in the embodiment of the present application to click Method detects machine click, and the embodiment of the present application is not limited this.
Fig. 1 is the implementation environment schematic diagram that the machine that one exemplary embodiment of the application provides clicks detection system, with It is said for the first application scenarios (i.e. user clicked to recommendation information scene) in above-mentioned three kinds of application scenarios It is bright, as shown in Figure 1, the system include: terminal 11, access server 12, data storage server 13, accounting server 14 and Real-time calculation server 15；
Terminal 11 for showing above-mentioned recommendation information, and receive user in the recommendation information can point in click on area Event is hit, optionally, terminal 11 is also used to report to access server 12 to the click event that can be received in click on area, In, this can after click on area refers to and is clicked, this can the click event in click on area be reported as effective click event To the region of access server 12, optionally, this can click on area refer to be clicked after can be shown in the display screen of terminal The region of the details user interface of the recommendation information being clicked, wherein the recommendation information pair is shown in details user interface The details for the product answered.
Optionally, terminal 11 includes at least one of following information into the click event that access server 12 reports: To can the account number mark of the account number clicked of click on area, the recommendation information being clicked message identification, issue the recommendation The account number mark of the publication account number of breath, the coordinate for the position that can be clicked in click on area, terminal 11 Internet protocol The generation moment of the address (Internet Protocol, IP) and the click event.Schematically, terminal 11 is to server 12 It include " abc123 in the click event reported；product_A；def456；(15,12)；192.168.31.1；10:12 ", In, abc123 be to can the account number of account number clicked of click on area identify；Product_A is the recommendation information being clicked Message identification；Def456 is the account number mark for the publication account number for issuing the recommendation information；(15,12) for can be in click on area by point The coordinate for the position hit；It 192.168.31.1 is the IP address of terminal 11,10:12 is the generation moment of click event.Optionally, All or part of data in click event can be used as the click coordinate data of the click event.
Optionally, terminal 11 is reported to access server 12 in the point that can be received in click on area by communication network 16 Hit event, wherein communication network 16 can be cable network, be also possible to wireless network.
Click event is reported to access server 12 it is worth noting that, illustrating only a terminal 11 in the present embodiment, In actual operation, it is connected with access server 12, and the number of the terminal 11 of click event can be reported to access server 12 Amount can be more.
Optionally, the click event that the terminal 11 received reports is reported to data storage server by access server 12 13, data storage server 13 is for storing the click event that access server 12 reports, optionally, data storage clothes The click event that access server 12 reports can be stored preset duration by business device 13, access server 12 can also be reported Click event carries out permanent storage.
Optionally, after access server 12 receives the click event that terminal 11 reports, sending to accounting server 14 should Click event, the click event that accounting server 14 is reported according to access server 12 carry out charging to the secondary click event, In, which refers to the expense that the corresponding product supplier of record recommendation information needs the publication account number to publication recommendation information to pay With.Schematically, product supplier and publication account number agreement recommendation information are often clicked once, and 0.2 yuan of charging, then, and charging clothes After business device 14 receives the click event that access server 12 reports, when the click event matches billing conditions, charging 0.2 Member.
Optionally, accounting server 14 needs to determine whether the click event received accords with by real-time calculation server 15 Close billing conditions.Optionally, accounting server 14 sends to real-time calculation server 15 and inquires according to the click event received Request.Schematically, include the account number mark for the publication account number for issuing the recommendation information in the inquiry request, calculate service in real time The case where device 15 identifies the recommendation information for determining publication account number publication according to account with the presence or absence of machine click is received, Or, including the message identification for the recommendation information being clicked in the inquiry request, real-time calculation server 15 is according to the message identification Determine whether the recommendation information is the information for receiving machine click.
Optionally, real-time calculation server 15 by from data storage server 12 obtain data to determine query result, And query result is sent to accounting server 14.Schematically, when the publication account in inquiry request including publication recommendation information Number account number mark when, real-time calculation server 15 obtains number corresponding with account mark from data storage server 12 According to calculate the case where recommendation information of publication account number publication is with the presence or absence of machine click is received；When being wrapped in inquiry request When including the message identification for the recommendation information being clicked, real-time calculation server 15 is obtained from data storage server 12 and is pushed away with this The corresponding data of information are recommended, to calculate whether the recommendation information is the information for receiving machine click.
Optionally, above-mentioned access server 12, data storage server 13, accounting server 14 and calculating service in real time Device 15 respectively can be realized independently on a different server, can realize on multiple servers in any combination, such as: access clothes The function that business device 12 and data storage server 13 are realized is realized by two modules respectively on server, can also be connect Enter server 12, data storage server 13, accounting server 14 and calculation server 15 is embodied as same service in real time Disparate modules on device can also be realized the function that server is realized by one or one group of server apparatus, such as: passing through Server a, server b and server c combination realize that this is not added in the function that access server 12 is realized, the embodiment of the present application To limit.
In conjunction with the detection system that above-mentioned application scenarios and machine are clicked, to the invention relates to the inspection clicked of machine Survey method is illustrated, as shown in Fig. 2, this method can be executed by server, which can be detection system shown in Fig. 1 In real-time calculation server 15, this method comprises:
Step 201, the click coordinate data of n target type content are obtained.
Optionally, the n target type content be n same type and including can click on area content, the click sit Mark data are the data generated according to the click event that can be received in click on area.
Optionally, the click coordinate data include target type content mark and the target type content on be clicked Pixel.
Optionally, which can be the click coordinate data in nearest preset duration, be also possible to deposit All click coordinate datas of storage, can also be the click coordinate data of the default size obtained at random, the embodiment of the present application pair This is not limited.
Optionally, it is serviced by the real-time calculating in detection system as shown in Figure 1 when obtain click coordinate data the step of Device 15 is completed, and above-mentioned access server 12, data storage server 13, accounting server 14 and real-time calculation server 15 When respectively independently realizing on a different server, real-time calculation server 15 can be obtained from data storage server 13 should Click coordinate data；When above-mentioned access server 12, data storage server 13, accounting server 14 and calculating service in real time Device 15 is realized when on same server, is stored with click coordinate data on the server, which can directly acquire The click coordinate data of storage.
Optionally, the type of the n target type content includes any one in following situation:
First, which is the recommendation information of n publication account number publication, wherein i-th of target type Content be i-th publication account number publication recommendation information set, wherein can click on area recommendation information is checked Region；
Second, the n target type content be n participate in ballot activity to polling option, wherein can click on area be Treat the control that polling option is voted；
Third, the n target type content are the recommendation informations that n is released to common platform, can click on area be pair The region that recommendation information is checked；
4th, which is the corresponding concern request message of n public's account number, wherein can click on area For the control paid close attention to public's account number.
Optionally, which is used to indicate can be clicked in click on area in n target type content Pixel.Optionally, which can be indicated in the form of coordinate in target-based coordinate system.Schematically, as shown in figure 3, Include recommendation information 32 in user interface 31, the recommendation information 32 can click on area be the region 33 that outlines of dotted line, such as sit Mark system 34 will be that a pixel be in 33 corresponding body present coordinate system 34 of region as it can be seen that including x-axis and y-axis in the coordinate system 34 The region of m*n.Schematically, the click coordinate data of above-mentioned recommendation information 32 include: (6,1), (3,2), (5,3), (6,1), (3,1), (6,1), (3,2), wherein the pixel that coordinate (6,1) indicates has been clicked three times, the pixel that coordinate (3,2) indicates Point has been clicked twice and the pixel of coordinate (5,3), (3,1) expression has respectively been clicked once.
Step 202, i-th of vector to be detected is determined according to the click coordinate data of i-th of target type content, 0 < i≤ n。
Optionally, what this i-th vector to be detected was used to indicate i-th of target type content can be in click on area at least The number that one pixel is clicked.
Optionally, when pixel is indicated in the form of coordinate in target-based coordinate system, this i-th vector to be detected is used for Indicate i-th of target type content can the number that is clicked of the corresponding coordinate of click on area, wherein target-based coordinate system is for leading to Cross the form of coordinate to pixel can the position in click on area be indicated.Optionally, the target-based coordinate system can refer to as Coordinate system 34 shown in Fig. 3.
Optionally, this i-th vector to be detected is the vector of p*q dimension, and the value of xth dimension is that the xth ties up corresponding coordinate The expression way of the touching quantity received, this i-th vector to be detected is as follows:
Wherein, CiIndicate i-th of vector to be detected,The touching quantity k that indicates coordinate (a, b) receives, wherein k's Value is changed according to the touching quantity being an actually-received on coordinate.
Schematically, with the click coordinate data of recommendation information 32 include: (3,1), (3,2), (1,3), (3,1), (2, 1), (3,3), (3,2), with C32Indicate the corresponding vector to be detected of the recommendation information 32, then the recommendation information 32 is corresponding to be checked Direction finding amount C32It is as follows:
Wherein,It is not clicked on for indicates coordinate (1,1),It is not clicked on for indicates coordinate (1,2),It is clicked once for indicates coordinate (1,3),It is clicked once for indicates coordinate (2,1),For indicates coordinate (2,2) are not clicked on,It is not clicked on for indicates coordinate (2,3),Two are clicked for indicates coordinate (3,1) It is secondary,It is clicked twice for indicates coordinate (3,2),It is clicked once for indicates coordinate (3,3).
Step 203, object vector is determined according to the click coordinate data of n target type content.
Optionally, what which was used to indicate n target type content can at least one corresponding picture of click on area The total degree that vegetarian refreshments is clicked.
Optionally, when pixel is indicated in the form of coordinate in target-based coordinate system, the object vector is for indicating n Target type content can the total degree that is clicked of the corresponding coordinate of click on area.
Optionally, which is the vector that p*q is tieed up, and the value of xth dimension is that the xth ties up corresponding and n target The total number of clicks amount that the corresponding coordinate of type content receives, the expression way of the object vector are as follows: C=(C1,1,C1,2..., Ca,b,…,Cp,q)
Wherein, C indicates object vector, Ca,bThe click that indicates coordinate (a, b) receives in n target type content is total Quantity.
Schematically, it is illustrated by taking recommendation information A, recommendation information B and tri- recommendation informations of recommendation information C as an example, Wherein, the click coordinate data of recommendation information A include: (3,1), (3,2), (1,3), the click coordinate data packet of recommendation information B Include: (3,1), (2,1), (3,3), the click coordinate data of recommendation information C include: (2,1), (3,3), (3,2), then recommendation information A, the corresponding object vector C of recommendation information B and recommendation information C is as follows:
C=(C1,1,C1,2,C1,3,C2,1,C2,2,C2,3,C3,1,C3,2,C3,3)
In conjunction with the click coordinate data of above-mentioned recommendation information A, recommendation information B and recommendation information C, in object vector C It is as follows per one-dimensional value difference:
C=(0,0,1,2,0,0,2,2,2)
Wherein, C1,1Corresponding value is 0, C1,2Corresponding value is 0, C1,3Corresponding value is 1, C2,1Corresponding value is 2, C2,2 Corresponding value is 0, C2,3Corresponding value is 0, C3,1Corresponding value is 2, C3,2Corresponding value is 2, C3,3Corresponding value is 2, then (0, 0,1,2,0,0,2,2,2) it is respectively used to indicates coordinate (1,1) to be not clicked in three recommendation informations, coordinate (1,2) is three It is not clicked in a recommendation information, coordinate (1,3) is clicked once in three recommendation informations, and coordinate (2,1) is pushed away at three It recommends and is clicked in information twice, coordinate (2,2) is not clicked in three recommendation informations, and coordinate (2,3) is in three recommendations It is not clicked in breath, coordinate (3,1) is clicked twice in three recommendation informations, and coordinate (3,2) is in three recommendation informations It is clicked twice, coordinate (3,3) is clicked twice in three recommendation informations.
It is worth noting that, being directed to above-mentioned steps 202 and step 203, step 202 can be first carried out, then execute step 203, step 203 can also be first carried out, then execute step 202, can be performed simultaneously with step 202 and step 203, the application is real It applies example and this is not limited.
Step 204, it when the vector distance of i-th of vector sum object vector to be detected is greater than first distance threshold value, determines I-th of target type content is to receive the content of machine click.
Optionally, when calculating the vector distance of i-th of vector sum object vector to be detected by cosine similarity formula, Actually being calculated is similarity between i-th of vector sum object vector to be detected, that is, passes through cosine similarity formula meter The similarity between i-th of vector sum object vector to be detected is calculated, when similar between i-th of vector sum object vector to be detected When degree is lower than similarity threshold, determine that i-th of target type content is to receive the content of machine click.
Schematically, the calculation formula of the similarity between i-th of vector sum object vector to be detected are as follows:
Wherein, CkFor indicating that i-th of vector to be detected, C are used to indicate object vector,For indicating i-th A target type content can each coordinate is clicked in click on area number summation, wherein the value of abscissa from 1 to P, the value of ordinate is from 1 to q, the number that k indicates coordinate (a, b) is clicked,It is being sat for i-th of target type content The total degree being clicked on mark (a, b) is k times, Ca,bTotal time be clicked on coordinate (a, b) for n target type content Number, wherein p >=1, q >=1, a >=1, b >=1, k >=0.
Optionally, the mode for calculating the vector distance of i-th of vector sum object vector to be detected includes as under type It is at least one:
First, i-th of vector sum mesh to be detected is calculated by Euclidean distance (English: Euclidean distance) formula Mark the vector distance between vector；
Schematically, i-th of vector to be detectedObject vector C=(C1,1, C1,2,…,Cp,q), then when calculating the distance between i-th of vector sum object vector to be detected d by Euclidean distance formula, calculate Formula is as follows:
Second, i-th of vector sum to be detected is calculated by mahalanobis distance (English: Mahalanobis Distance) formula Vector distance between object vector；
Schematically, i-th of vector C to be detected is calculated by horse formula range formulakThe distance between object vector C d When, calculation formula is as follows:
Wherein, S-1For indicating the covariance between two each elements of vector.
Third calculates the vector distance between i-th of vector sum object vector to be detected by manhatton distance formula.
Schematically, i-th of vector to be detectedObject vector C=(C1,1, C1,2,…,Cp,q), then when calculating the distance between i-th of vector sum object vector to be detected d by Euclidean distance formula, calculate Formula is as follows:
Optionally, when the vector distance of i-th of vector sum object vector to be detected is greater than first distance threshold value, illustrate the The difference of i vector sum object vectors to be detected is larger, it is determined that i-th of target type content is receive machine click interior Hold.
Optionally, it when the vector distance of i-th of vector sum object vector to be detected is not more than first distance threshold value, determines I-th of target type content is not receive the content of machine click.
In conclusion the detection method that machine provided in this embodiment is clicked, is sat by the click of n target type content Mark data determine object vector, the object vector be used for represent manually to can click on area click when click feature, lead to The click data for crossing i-th of target type content determines i-th of vector to be detected, this i-th vector to be detected is for representing i-th The click feature of the click event received in a target type content, by comparing i-th of vector sum object vector to be detected Between vector distance, namely compare the similarity between i-th of vector sum object vector to be detected, when vector distance is greater than threshold When value, then illustrate that the similarity of i-th vector and object vector to be detected is lower, namely in i-th of target type content of explanation The click feature of the click event received with manually to can click on area click when click feature similarity it is lower, then Think that i-th of target type content is to receive the content of machine click, improves the recognition accuracy clicked to machine.
In an alternative embodiment, n target type content can the size of click on area can be different, scheme 4 be the flow chart for the detection method that the machine that another exemplary embodiment of the application provides is clicked, and this method can be by servicing Device executes, which can be the real-time calculation server 15 in detection system shown in Fig. 1, this method comprises:
Step 401, the click coordinate data of n target type content are obtained.
Optionally, before obtaining n target type content, when access server 12, the number in detection system shown in FIG. 1 On a different server according to storage server 13, accounting server 14 and the respective independent realization of calculation server 15 in real time When, real-time calculation server 15 receives the inquiry request that accounting server 14 is sent, and the inquiry request is for requesting i-th of inquiry Whether target type content is the content for receiving machine click；When above-mentioned access server 12, data storage server 13, meter Take server 14 and calculation server 15 realizes that when on same server, server receives the point that terminal reports in real time After hitting event, the click coordinate data of n target type content are obtained.
Optionally, the n target type content be n same type and including can click on area content, the click sit Mark data are the data generated according to the click event that can be received in click on area.
Optionally, in the present embodiment by the n target type content be n publication account number issue recommendation information for into Row explanation, wherein i-th of target type content is the set of the recommendation information of i-th of publication account number publication, wherein can be clicked Region is the region checked to recommendation information；
Step 402, by miRecommendation information can the pixel in click on area map to target-based coordinate system.
Optionally, the target-based coordinate system for coordinate by way of to pixel can position progress in click on area It indicates.
Optionally, the miRecommendation information can the size of click on area can be identical, be also possible to different.
Optionally, by recommendation information can click on area map to target-based coordinate system mode can refer to it is shown in Fig. 3 It can click on area 33 and coordinate system 34.
Optionally, which the terminal 11 in detection system can also carry out as shown in Figure 1.
Step 403, according to can the pixel that is clicked of click on area determine the coordinate being clicked in target-based coordinate system.
Optionally, the pixel that can be clicked in click on area is mapped in target-based coordinate system, obtains target-based coordinate system In the coordinate that is clicked.
Step 404, by miThe coordinate that can be clicked in click on area of recommendation information is normalized.
Optionally, which is by miRecommendation information can click on area normalizing be target size region, And by can the Unitary coordinate in click on area be coordinate under target size.
Schematically, the size of target size 5*5, recommendation information A in target-based coordinate system is 10*15, this is pushed away Recommending the coordinate being clicked in information A includes (6,3), (8,9), then when recommendation information A being normalized to the size of 5*5, normalization The coordinate that is clicked afterwards includes (3,1), (4,3)；Size of the recommendation information B in target-based coordinate system is 15*15, the recommendation The coordinate being clicked in information B includes (9,3) and (6,9), then when recommendation information B being normalized to target size, after normalization Be clicked coordinate include (3,1) and (2,3).
Step 405, i-th of vector to be detected is obtained according to the total degree that the coordinate after normalization is clicked.
Optionally, what this i-th vector to be detected was used to indicate i-th of target type content can be in click on area at least The number that one pixel is clicked.
Optionally, when pixel is indicated in the form of coordinate in target-based coordinate system, this i-th vector to be detected is used for Indicate i-th of target type content can the number that is clicked of the corresponding coordinate of click on area, wherein target-based coordinate system is for leading to Cross the form of coordinate to pixel can the position in click on area be indicated.
Schematically, in conjunction with the example above, when recommendation information A is normalized to the size of 5*5, being clicked after normalization Coordinate include (3,1), (4,3), when recommendation information B is normalized to target size, after normalization be clicked coordinate include (3, 1) with (2,3), then after normalizing, the number that coordinate (3,1) is clicked is 1 time, the number that coordinate (4,3) and (2,3) are clicked It is 1 time.Optionally, the number that other coordinates indicated in vector to be detected are clicked all is 0 time.
Step 406, by recommendation information that n is issued account number publication can the pixel in click on area map to target and sit Mark system.
Optionally, by recommendation information that n is issued account number publication can the pixel in click on area map to target and sit When mark system, need in each recommendation information for issuing n account number publication can the pixel in click on area map to target Coordinate system.Wherein, mapping mode please refer to it is shown in Fig. 3 can click on area 33 and coordinate system 34.
Step 407, according to can the pixel that is clicked of click on area determine the coordinate being clicked in target-based coordinate system.
Step 408, the coordinate that can be clicked in click on area for the recommendation information that n is issued account number publication is subjected to normalizing Change.
Optionally, the detailed process of the normalization step can refer to above-mentioned steps 404, and details are not described herein again.
Step 409, object vector is obtained according to the total degree that the coordinate after normalization is clicked.
Optionally, the object vector be used for indicate n publication account number issue recommendation information can click on area it is corresponding The total degree that at least one pixel is clicked.
Optionally, when pixel is indicated in the form of coordinate in target-based coordinate system, the object vector is for indicating n Target type content can the total degree that is clicked of the corresponding coordinate of click on area.
It schematically, is (88,14) after the Unitary coordinate that user 1 clicks on the recommendation information of publication account number a publication, It is (99,39) after the Unitary coordinate that user 1 clicks on the recommendation information of publication account number b publication, user 2 is in publication account number a It is (77,15) that user 3 is on the recommendation information of publication account number a publication after the Unitary coordinate clicked on the recommendation information of publication It is (88,14) Unitary coordinate that user 4 clicks on the recommendation information of publication account number a publication after the Unitary coordinate of click Be afterwards (88,16), then after summarizing by publication account number, include in the corresponding vector to be detected of publication account number 1 following item (88,14, 2), (88,16,1) and (77,15,1), issuing in the corresponding vector to be detected of account number 2 includes following item (99,39,1).
It is worth noting that, above-mentioned steps 402 can first carry out step 402 to step 405 and step 406 to step 409 To step 405, then execute step 406 to step 409, step 406 can also be first carried out to step 409, then execute step 402 to Step 405, it may also be performed simultaneously step 402 to step 405 and step 406 to step 409, this is not added in the embodiment of the present application To limit.
Step 410, it when the vector distance of i-th of vector sum object vector to be detected is greater than first distance threshold value, determines I-th of target type content is to receive the content of machine click.
Optionally, when calculating the vector distance of i-th of vector sum object vector to be detected by cosine similarity formula, Actually being calculated is similarity between i-th of vector sum object vector to be detected, that is, passes through cosine similarity formula meter The similarity between i-th of vector sum object vector to be detected is calculated, when similar between i-th of vector sum object vector to be detected When degree is lower than similarity threshold, determine that i-th of target type content is to receive the content of machine click.
Optionally, the mode for calculating the vector distance of i-th of vector sum object vector to be detected includes as under type It is at least one:
First, the vector distance between i-th of vector sum object vector to be detected is calculated by Euclidean distance formula；
Second, the vector distance between i-th of vector sum object vector to be detected is calculated by mahalanobis distance formula；
Third calculates the vector distance between i-th of vector sum object vector to be detected by manhatton distance formula.
Optionally, when the vector distance of i-th of vector sum object vector to be detected is greater than first distance threshold value, illustrate the The difference of i vector sum object vectors to be detected is larger, it is determined that i-th of target type content is receive machine click interior Hold.
In conclusion the detection method that machine provided in this embodiment is clicked, is sat by the click of n target type content Mark data determine object vector, the object vector be used for represent manually to can click on area click when click feature, lead to The click data for crossing i-th of target type content determines i-th of vector to be detected, this i-th vector to be detected is for representing i-th The click feature of the click event received in a target type content, by comparing i-th of vector sum object vector to be detected Between vector distance, namely compare the similarity between i-th of vector sum object vector to be detected, when vector distance is greater than threshold When value, then illustrate that the similarity of i-th vector and object vector to be detected is lower, namely in i-th of target type content of explanation The click feature of the click event received with manually to can click on area click when click feature similarity it is lower, then Think that i-th of target type content is to receive the content of machine click, improves the recognition accuracy clicked to machine.
Method provided in this embodiment, by by miThe described of recommendation information can click on area and can be in click on area The coordinate being clicked is normalized, and avoids due to miRecommendation information it is described can click on area size it is different and cause Difference between i-th of vector sum object vector to be detected also will increase difference namely i-th of vector sum to be detected in size The problem of difference between object vector is influenced by size factor, is also decreased to the recognition accuracy that machine is clicked.
In an alternative embodiment, by calculate i-th k-th of vector sum vector to be detected to be detected to span From may determine that whether i-th of target type content and k-th of target type content receive same type of machine and click.Figure 5 be the flow chart for the detection method that the machine that another exemplary embodiment of the application provides is clicked, as shown in figure 5, this method It can be executed by server, which can be the real-time calculation server 15 in detection system shown in Fig. 1, this method packet It includes:
Step 501, the inquiry request that accounting server is sent is received.
It optionally, include the corresponding mark of i-th of target type content in the inquiry request, the inquiry request is for requesting Determine whether i-th of target type content is the content for receiving machine click.
Step 502, the click coordinate data of n target type content are obtained.
Optionally, the n target type content be n same type and including can click on area content, the click sit Mark data are the data generated according to the click event that can be received in click on area.
Optionally, the type of the n target type content includes any one in following situation:
First, which is the recommendation information of n publication account number publication, wherein i-th of target type Content be i-th publication account number publication recommendation information set, wherein can click on area recommendation information is checked Region；
Second, the n target type content be n participate in ballot activity to polling option, wherein can click on area be Treat the control that polling option is voted；
Third, the n target type content are the recommendation informations that n is released to common platform, can click on area be pair The region that recommendation information is checked；
4th, which is the corresponding concern request message of n public's account number, wherein can click on area For the control paid close attention to public's account number.
Optionally, the click coordinate data of n target type content are obtained according to the inquiry request that accounting server is sent, Schematically, which is the recommendation information of n publication account number publication, then includes i-th of hair in inquiry request The account number of cloth account number identifies.
Step 503, i-th of vector to be detected is determined according to the click coordinate data of i-th of target type content, 0 < i≤ n。
Optionally, what this i-th vector to be detected was used to indicate i-th of target type content can be in click on area at least The number that one pixel is clicked.
Optionally, when pixel is indicated in the form of coordinate in target-based coordinate system, this i-th vector to be detected is used for Indicate i-th of target type content can the number that is clicked of the corresponding coordinate of click on area, wherein target-based coordinate system is for leading to Cross the form of coordinate to pixel can the position in click on area be indicated.Optionally, the target-based coordinate system can refer to as Coordinate system 34 shown in Fig. 3.
Step 504, object vector is determined according to the click coordinate data of n target type content.
Optionally, what which was used to indicate n target type content can at least one corresponding picture of click on area The total degree that vegetarian refreshments is clicked.
Optionally, when pixel is indicated in the form of coordinate in target-based coordinate system, the object vector is for indicating n Target type content can the total degree that is clicked of the corresponding coordinate of click on area.
It is worth noting that, being directed to above-mentioned steps 503 and step 504, step 503 can be first carried out, then execute step 504, step 504 can also be first carried out, then execute step 503, can be performed simultaneously with step 503 and step 504, the application is real It applies example and this is not limited.
Step 505, it when the vector distance of i-th of vector sum object vector to be detected is greater than first distance threshold value, determines I-th of target type content is to receive the content of machine click.
Optionally, when calculating the vector distance of i-th of vector sum object vector to be detected by cosine similarity formula, Actually being calculated is similarity between i-th of vector sum object vector to be detected, that is, passes through cosine similarity formula meter The similarity between i-th of vector sum object vector to be detected is calculated, when similar between i-th of vector sum object vector to be detected When degree is lower than similarity threshold, determine that i-th of target type content is to receive the content of machine click.
Optionally, the mode for calculating the vector distance of i-th of vector sum object vector to be detected includes as under type It is at least one:
First, the vector distance between i-th of vector sum object vector to be detected is calculated by Euclidean distance formula；
Second, the vector distance between i-th of vector sum object vector to be detected is calculated by mahalanobis distance formula；
Third calculates the vector distance between i-th of vector sum object vector to be detected by manhatton distance formula.
Optionally, when the vector distance of i-th of vector sum object vector to be detected is greater than first distance threshold value, illustrate the The difference of i vector sum object vectors to be detected is larger, it is determined that i-th of target type content is receive machine click interior Hold.
Step 506, query result is sent to accounting server.
Optionally, it is to receive the content of machine click which, which is used to indicate i-th of target type content,.
Optionally, when the vector distance of i-th of vector sum object vector to be detected is less than or equal to first distance threshold value, then I-th of target type content is not receive the content of machine click, then is used for the query result that accounting server is sent Indicate that i-th of target type content is not receive the content of machine click.
Step 507, according to the click coordinate data of k-th of target type content determine k-th of 0 < k of vector to be detected≤ n。
Optionally, what this k-th vector to be detected was used to indicate k-th of target type content can be in click on area at least The number that one pixel is clicked.
Optionally, when pixel is indicated in the form of coordinate in target-based coordinate system, this k-th vector to be detected is used for Indicate k-th of target type content can the number that is clicked of the corresponding coordinate of click on area.
Step 508, when the vector distance of i-th of k-th of vector sum vector to be detected to be detected is less than second distance threshold value When, determine that k-th of target type content and the machine of i-th of target type content reception to same type are clicked.
Schematically, referring to FIG. 6, Fig. 6 shows the click coordinate distribution map of n recommendation information, wherein mainly show The click coordinate point of recommendation information A click coordinate distribution map 61, the click coordinate distribution map 62 of recommendation information B, recommendation information C The click coordinate distribution map 64 of cloth Figure 63 and recommendation information N, according to the click distribution map of n recommendation information obtains comprehensive point Cloth Figure 65, wherein in the click coordinate distribution map 61 of recommendation information A, click coordinate is more concentrated in region 611, recommendation In the click coordinate distribution map 62 for ceasing B, click coordinate is more concentrated in region 612, and is integrated in distribution map 65, in region Click coordinate is more concentrated in 651, through the above steps 502 to step 505 calculating recommendation information A vector to be detected and target The distance between vector is greater than distance threshold, then confirms that recommendation information A is the content for receiving machine click, by recommendation information A Click coordinate distribution map and recommendation information B click coordinate distribution map normalize and obtains two vectors to be detected after, calculating Obtain the vector distance of the corresponding vector to be detected of recommendation information A vector to be detected corresponding with recommendation information B less than second away from From threshold value, therefore recommendation information A and recommendation information B are the content for receiving the machine of same type and clicking.
In conclusion the detection method that machine provided in this embodiment is clicked, is sat by the click of n target type content Mark data determine object vector, the object vector be used for represent manually to can click on area click when click feature, lead to The click data for crossing i-th of target type content determines i-th of vector to be detected, this i-th vector to be detected is for representing i-th The click feature of the click event received in a target type content, by comparing i-th of vector sum object vector to be detected Between vector distance, namely compare the similarity between i-th of vector sum object vector to be detected, when vector distance is greater than threshold When value, then illustrate that the similarity of i-th vector and object vector to be detected is lower, namely in i-th of target type content of explanation The click feature of the click event received with manually to can click on area click when click feature similarity it is lower, then Think that i-th of target type content is to receive the content of machine click, improves the recognition accuracy clicked to machine.
Method provided in this embodiment, by the click data by k-th of target type content determine k-th it is to be detected Vector, this k-th vector to be detected are used to represent the click feature of the click event received in k-th of target type content, By comparing the vector distance between i-th of k-th of vector sum vector to be detected to be detected, namely compare i-th of direction finding to be checked Measure the similarity between k-th of vector to be detected, when vector distance is less than threshold value, then illustrate i-th vector to be detected with The similarity of k-th of vector to be detected is higher, namely the click of the click event received in i-th of target type content of explanation The click feature similarity of the click event received in feature and k-th of target type content is higher, then it is assumed that i-th of mesh Mark type content and k-th of target type content reception to be same type machine click, improve to same type machine point The recognition accuracy hit.
It is to participate in the n of ballot activity to polling option with n target type content in a schematical embodiment For be illustrated, as shown in fig. 7, shown in the user interface 71 of ballot activity to polling option A can click on area 711, to polling option B can click on area 712, to polling option C can click on area 713 and to polling option D can Click on area 714, for user after can the clicking in click on area of polling option, selecting ballot control 72 can be to by point That hits votes to polling option.Schematically, user treat polling option A can click on area 711 clicked, and After clicking ballot control 72, treats polling option A and vote, then terminal waits for this in secondary click event of polling option A Report is to background server, such as: access server 12 shown in FIG. 1, and accounting server 14 shown in FIG. 1 in the present embodiment can be with It is embodied as count of votes server, after access server 12 sends the secondary click event to count of votes server, count of votes server is to real-time Calculation server 15 sends inquiry request.Real-time calculation server 15 to this wait for polling option A whether receive machine click into Row detection.
Realization process as shown in figure 8, this method comprises:
Step 801, terminal reports click coordinate data after receiving click event.
It optionally, include at least one of following information in the click event that terminal to server reports: to can click The account number mark of the account number that region is clicked, the recommendation information being clicked message identification, issue the publication of the recommendation information The life of the account number mark, the coordinate of position, the IP address of terminal and the click event that can be clicked in click on area of account number At the moment.
Optionally, terminal reports click coordinate data to above-mentioned access server 12, and optionally, terminal is to access server 12 report click event, include click coordinate data in the click event.
Step 802, server obtains the corresponding click coordinate data of each option in preset duration, and is aggregated into respectively to be checked Direction finding amount.
Optionally, it will click on and had been carried out in above-mentioned steps 202 in detail the step of coordinate data summarizes for vector to be detected Describe in detail bright, details are not described herein again.
Step 803, server generates object vector according to the click coordinate data of each option, and determines vector sum to be detected Vector distance between object vector.
Optionally, the step of generating object vector according to click coordinate data has been carried out in detail in above-mentioned steps 203 Illustrate, details are not described herein again.
Step 804, judge whether the vector distance of the vector and object vector to be detected of option A exceeds threshold value.
Step 805, when the vector distance of the vector and object vector to be detected of option A exceeds threshold value, determine that option A connects Receive machine click.
Step 806, when the vector distance of the vector and object vector to be detected of option A is without departing from threshold value, option A is determined Machine click is not received.
In conclusion the detection method that machine provided in this embodiment is clicked, true by the click coordinate data of each option Set the goal vector, the object vector be used for represent manually to can click on area click when click feature, pass through option A's Click data determines the vector to be detected of option A, and the vector to be detected of option A is for representing the point received in option A The click feature for hitting event by comparing the vector distance between the vector sum object vector to be detected of option A, namely compares choosing Similarity between the vector sum object vector to be detected of item A then illustrates that option A's is to be checked when vector distance is greater than threshold value Direction finding amount and the similarity of object vector are lower, namely illustrate the click feature of the click event received in option A and artificial To can click on area click when click feature similarity it is lower, then it is assumed that option A be receive machine click it is interior Hold, improves the recognition accuracy clicked to machine.
Fig. 9 is the structural block diagram for the detection device that the machine that one exemplary embodiment of the application provides is clicked, such as Fig. 9 institute Show, which can be by special hardware circuit, alternatively, software and hardware combining is implemented as real-time calculation server shown in FIG. 1 15 all or part of, which includes: to obtain module 91, determining module 92 and judgment module 93；
Module 91 is obtained, for obtaining the click coordinate data of n target type content, the n target type content Be n same type and including can click on area content, the click coordinate data can be connect in click on area according to The data that the click event received generates；
Determining module 92, for according to the click coordinate data of i-th of target type content determine i-th it is to be detected Vector, what i-th of vector to be detected was used to indicate i-th of target type content can at least one of click on area The number that pixel is clicked, 0 < i≤n；
The determining module 92, be also used to be determined according to the click coordinate data of the n target type content target to Amount, what the object vector was used to indicate the n target type content can at least one corresponding pixel quilt of click on area The total degree of click；
Judgment module 93, the vector distance for working as object vector described in described i-th vector sum to be detected are greater than first When distance threshold, determine that i-th of target type content is the content for receiving the machine and clicking.
In an alternative embodiment, as shown in Figure 10, when the n target type content is n publication account number hair When the recommendation information of cloth, i-th of publication account number publication has miRecommendation information includes described in the click coordinate data miIn recommendation information, every recommendation information can the pixel that is clicked of click on area；
Determining module 92, including
Mapping submodule 921 is used for the miRecommendation information it is described can the pixel in click on area map to Target-based coordinate system, the target-based coordinate system for by way of coordinate to the pixel it is described can position in click on area It sets and is indicated；
Determine submodule 922, for according to can the pixel that is clicked of click on area determine the target-based coordinate system In the coordinate that is clicked；
Submodule 923 is normalized, is used for the miThe described of recommendation information can be clicked described in click on area Coordinate is normalized；
The determining submodule 922 is also used to be obtained according to the total degree that the coordinate after normalization is clicked described I-th of vector to be detected.
In an alternative embodiment, the n target type content is the recommendation information of n publication account number publication；
Determining module 92, comprising:
Mapping submodule 921, can be in click on area described in the recommendation information of account number publication for issuing described n Pixel maps to target-based coordinate system, the target-based coordinate system be used for the pixel it is described can position in click on area It is indicated；
Determine submodule 922, for according to can the pixel that is clicked of click on area determine the target-based coordinate system In the coordinate that is clicked；
Submodule 923 is normalized, it can be in click on area described in the recommendation information that account number is issued for issuing described n The coordinate being clicked is normalized；
The determining submodule 922 is also used to be obtained according to the total degree that the coordinate after normalization is clicked described Object vector.
In an alternative embodiment, the n target type content is the recommendation information of n publication account number publication, I-th of target type content be i-th publication account number publication recommendation information set, it is described can click on area for institute State the region that recommendation information is checked；
Or,
The n target type content be n participate in ballot activity to polling option, it is described can click on area for institute State the control voted to polling option；
Or,
The n target type content is the recommendation informations that n is released to common platform, it is described can click on area be pair The region that the recommendation information is checked；
Or,
The n target type content be the corresponding concern request message of n public's account number, it is described can click on area for pair The control that public's account number is paid close attention to.
In an alternative embodiment, the determining module 92, is also used to the click according to k-th of target type content Coordinate data determines k-th of vector to be detected, 0 < k≤n；
The determining module 92 is also used to work as the vector of k-th of vector to be detected described in described i-th vector sum to be detected When distance is less than second distance threshold value, determine that k-th of target type content is arrived with i-th of target type content reception The machine of same type is clicked.
In an alternative embodiment, the judgment module 93 is also used to by described in the calculating of cosine similarity formula Similarity between object vector described in i-th of vector sum to be detected, the object vector described in described i-th vector sum to be detected Similarity when being lower than similarity threshold, determine that i-th of target type content is the content for receiving the machine and clicking.
In an alternative embodiment, the judgment module 93 is also used to calculate described i-th by Euclidean distance formula The vector distance between object vector described in a vector sum to be detected；
Or,
The judgment module 93 is also used to calculate mesh described in i-th of vector sum to be detected by mahalanobis distance formula Mark the vector distance between vector；
Or,
The judgment module 93 is also used to calculate described in i-th of vector sum to be detected by manhatton distance formula The vector distance between object vector.
In an alternative embodiment, described device, further includes:
Receiving module 94 includes described i-th in the inquiry request for receiving the inquiry request of accounting server transmission The corresponding mark of a target type content, the inquiry request for request to determine i-th of target type content whether be Receive the content that the machine is clicked；
Sending module 95, for sending query result to the accounting server, the query result is used to indicate described I-th of target type content is the content for receiving the machine and clicking.
In conclusion the detection device that machine provided in this embodiment is clicked, is sat by the click of n target type content Mark data determine object vector, the object vector be used for represent manually to can click on area click when click feature, lead to The click data for crossing i-th of target type content determines i-th of vector to be detected, this i-th vector to be detected is for representing i-th The click feature of the click event received in a target type content, by comparing i-th of vector sum object vector to be detected Between vector distance, namely compare the similarity between i-th of vector sum object vector to be detected, when vector distance is greater than threshold When value, then illustrate that the similarity of i-th vector and object vector to be detected is lower, namely in i-th of target type content of explanation The click feature of the click event received with manually to can click on area click when click feature similarity it is lower, then Think that i-th of target type content is to receive the content of machine click, improves the recognition accuracy clicked to machine.
Present invention also provides a kind of server, which includes processor and memory, be stored in memory to A few instruction, at least one instruction load by processor and is executed with realize above-mentioned each embodiment of the method offer in machine The detection method of click.It should be noted that the server can be server provided by following Figure 11.
Figure 11 is please referred to, it illustrates the structural schematic diagrams for the server that one exemplary embodiment of the application provides.Tool For body: the server 1100 includes 1102 He of central processing unit (CPU) 1101 including random access memory (RAM) The system storage 1104 of read-only memory (ROM) 1103, and connection system storage 1104 and central processing unit 1101 System bus 1105.The server 1100 further includes that the substantially defeated of information is transmitted between each device helped in computer Enter/output system (I/O system) 1106, and is used for storage program area 1113, application program 1114 and other program modules 1115 mass-memory unit 1107.
The basic input/output 1106 includes display 1108 for showing information and inputs for user The input equipment 1109 of such as mouse, keyboard etc of information.Wherein the display 1108 and input equipment 1109 all pass through The input and output controller 1110 for being connected to system bus 1105 is connected to central processing unit 1101.The basic input/defeated System 1106 can also include input and output controller 1110 to touch for receiving and handling from keyboard, mouse or electronics out Control the input of multiple other equipment such as pen.Similarly, input and output controller 1110 also provide output to display screen, printer or Other kinds of output equipment.
The mass-memory unit 1107 (is not shown by being connected to the bulk memory controller of system bus 1105 It is connected to central processing unit 1101 out).The mass-memory unit 1107 and its associated computer-readable storage medium Matter is that server 1100 provides non-volatile memories.That is, the mass-memory unit 1107 may include such as hard The computer readable storage medium (not shown) of disk or CD-ROI driver etc.
Without loss of generality, the computer readable storage medium may include computer storage media and communication media.Meter Calculation machine storage medium is believed including computer readable instructions, data structure, program module or other data etc. for storage The volatile and non-volatile of any method or technique realization of breath, removable and irremovable medium.Computer storage medium Including RAM, ROM, EPROM, EEPROM, flash memory or other solid-state storages its technologies, CD-ROM, DVD or other optical storages, magnetic Tape drum, tape, disk storage or other magnetic storage devices.Certainly, skilled person will appreciate that computer storage is situated between Matter is not limited to above-mentioned several.Above-mentioned system storage 1104 and mass-memory unit 1107 may be collectively referred to as memory.
Memory is stored with one or more programs, and one or more programs are configured to by one or more central processings Unit 1101 executes, and one or more programs include the instruction for the detection method clicked for realizing above-mentioned machine, central processing Unit 1101 executes the detection method that the machine that the one or more program realizes that above-mentioned each embodiment of the method provides is clicked.
According to various embodiments of the present invention, the server 1100 can also be arrived by network connections such as internets Remote computer operation on network.Namely server 1100 can be connect by the network being connected on the system bus 1105 Mouth unit 1111 is connected to network 1112, in other words, it is other kinds of to be connected to that Network Interface Unit 1111 also can be used Network or remote computer system (not shown).
The memory further includes that one or more than one program, the one or more programs are stored in In memory, the one or more programs include the detection for carrying out machine click provided in an embodiment of the present invention The step as performed by server in method.
The embodiment of the present application also provides a kind of computer readable storage medium, and at least one finger is stored in the storage medium Enable, at least a Duan Chengxu, code set or instruction set, at least one instruction, an at least Duan Chengxu, the code set or Instruction set is loaded by the processor 1110 and is executed to realize such as the machine click as described in Fig. 2, Fig. 4, Fig. 5 and Fig. 8 are any Detection method.
Present invention also provides a kind of computer program products to make when computer program product is run on computers It obtains computer and executes the detection method that the machine that above-mentioned each embodiment of the method provides is clicked.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware It completes, relevant hardware can also be instructed to complete by program, the program can store in a kind of computer-readable In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.

Claims (15)

1. the detection method that a kind of machine is clicked, which is characterized in that the described method includes:
Obtain n target type content click coordinate data, the n target type content be n same type and including Can click on area content, the click coordinate data are that the click event that can be received in click on area according to generates Data；
I-th of vector to be detected is determined according to the click coordinate data of i-th of target type content, and described i-th to be checked What direction finding amount was used to indicate i-th of target type content can time that is clicked of at least one pixel in click on area Number, 0 < i≤n；
Object vector is determined according to the click coordinate data of the n target type content, and the object vector is for indicating institute That states n target type content can the total degree that is clicked of at least one corresponding pixel of click on area；
When the vector distance of the object vector described in described i-th vector sum to be detected is greater than first distance threshold value, determine described in I-th of target type content is the content for receiving the machine and clicking.
2. the method according to claim 1, wherein when the n target type content is n publication account number hair When the recommendation information of cloth, i-th of publication account number publication has miRecommendation information includes described in the click coordinate data miEvery recommendation information can the pixel that is clicked of click on area in recommendation information；
The click coordinate data according to i-th of target type content determine i-th of vector to be detected, comprising:
By the miRecommendation information it is described can the pixel in click on area map to target-based coordinate system, the coordinates of targets System for by way of coordinate to the pixel it is described can the position in click on area be indicated；
According to it is described can the pixel that is clicked of click on area determine the coordinate being clicked in the target-based coordinate system；
By the miThe coordinate that can be clicked described in click on area of recommendation information is normalized；
Described i-th vector to be detected is obtained according to the total degree that the coordinate after normalization is clicked.
3. the method according to claim 1, wherein the n target type content is n publication account number publication Recommendation information；
The click coordinate data according to the n target type content determine object vector, comprising:
By described n issue described in the recommendation information of account number publication can the pixel in click on area map to coordinates of targets System, the target-based coordinate system be used for the pixel it is described can the position in click on area be indicated；
According to it is described can the pixel that is clicked of click on area determine the coordinate being clicked in the target-based coordinate system；
The coordinate that can be clicked in click on area described in the recommendation information of account number publication is issued by described n to be normalized；
The object vector is obtained according to the total degree that the coordinate after normalization is clicked.
4. method according to any one of claims 1 to 3, which is characterized in that
The n target type content is the recommendation information of n publication account number publication, and i-th of target type content is i-th It is a publication account number publication recommendation information set, it is described can click on area be the region checked to the recommendation information；
Or,
The n target type content is n participation ballot activity to polling option, it is described can click on area for it is described to The control that polling option is voted；
Or,
The n target type content is the recommendation informations that n is released to common platform, it is described can click on area be to described The region that recommendation information is checked；
Or,
The n target type content is the corresponding concern request message of n public's account number, it is described can click on area for described The control that public's account number is paid close attention to.
5. method according to any one of claims 1 to 3, which is characterized in that determination i-th of target type content After content to receive the machine click, further includes:
K-th of vector to be detected, 0 < k≤n are determined according to the click coordinate data of k-th of target type content；
When the vector distance of k-th of vector to be detected described in described i-th vector sum to be detected is less than second distance threshold value, really The machine of fixed k-th of target type content and i-th of target type content reception to same type is clicked.
6. method according to any one of claims 1 to 3, which is characterized in that described to work as described i-th vector sum institute to be detected When stating the vector distance of object vector greater than first distance threshold value, it is described to receive to determine i-th of target type content The content that machine is clicked, comprising:
The similarity between object vector described in described i-th vector sum to be detected is calculated by cosine similarity formula, works as institute When stating the similarity of object vector described in i-th of vector sum to be detected lower than similarity threshold, i-th of target type is determined Content is the content for receiving the machine and clicking.
7. method according to any one of claims 1 to 3, which is characterized in that described to work as described i-th vector sum institute to be detected When stating the vector distance of object vector greater than first distance threshold value, it is described to receive to determine i-th of target type content Before the content that machine is clicked, further includes:
The vector distance between object vector described in described i-th vector sum to be detected is calculated by Euclidean distance formula；
Or,
The vector distance between object vector described in described i-th vector sum to be detected is calculated by mahalanobis distance formula；
Or,
It is calculated by manhatton distance formula described to span between object vector described in described i-th vector sum to be detected From.
8. method according to any one of claims 1 to 3, which is characterized in that described according to i-th of target type content Click coordinate data determine i-th of vector to be detected before, further includes:
The inquiry request that accounting server is sent is received, includes that i-th of target type content is corresponding in the inquiry request Mark, the inquiry request is for requesting to determine whether i-th of target type content is to receive the machine to click Content；
Determination i-th of target type content is after receiving the content that the machine is clicked, further includes:
Query result is sent to the accounting server, the query result is used to indicate i-th of target type content and is Receive the content that the machine is clicked.
9. the detection device that a kind of machine is clicked, which is characterized in that described device includes:
Module is obtained, for obtaining the click coordinate data of n target type content, the n target type content is n phase Same type and including can click on area content, the click coordinate data are the points that can be received in click on area according to Hit the data of event generation；
Determining module, for determining i-th of vector to be detected according to the click coordinate data of i-th of target type content, What i-th of vector to be detected was used to indicate i-th of target type content can at least one pixel in click on area The number that point is clicked, 0 < i≤n；
The determining module is also used to determine object vector according to the click coordinate data of the n target type content, described Object vector be used for indicate the n target type content can at least one corresponding pixel of click on area be clicked it is total Number；
Judgment module, the vector distance for working as object vector described in described i-th vector sum to be detected are greater than first distance threshold When value, determine that i-th of target type content is the content for receiving the machine and clicking.
10. device according to claim 9, which is characterized in that when the n target type content is n publication account number When the recommendation information of publication, i-th of publication account number publication has miRecommendation information includes institute in the click coordinate data State miIn recommendation information, every recommendation information can the pixel that is clicked of click on area；
The determining module, comprising:
Mapping submodule is used for the miRecommendation information it is described can the pixel in click on area map to coordinates of targets System, the target-based coordinate system for coordinate by way of to the pixel it is described can position progress table in click on area Show；
Determine submodule, for according to can the pixel that is clicked of click on area determine in the target-based coordinate system and be clicked Coordinate；
Submodule is normalized, is used for the miThe coordinate that can be clicked described in click on area of recommendation information carries out Normalization；
The determining submodule, be also used to be obtained according to the total degree that the coordinate after normalization is clicked described i-th to Detect vector.
11. device according to claim 9, which is characterized in that the n target type content is n publication account number hair The recommendation information of cloth；
The determining module, comprising:
Mapping submodule, can pixel in click on area for issuing described n described in the recommendation information of account number publication Map to target-based coordinate system, the target-based coordinate system be used for the pixel it is described can the position in click on area carry out table Show；
Determine submodule, for according to can the pixel that is clicked of click on area determine in the target-based coordinate system and be clicked Coordinate；
Submodule is normalized, can be clicked in click on area described in the recommendation information of account number publication for issuing described n Coordinate be normalized；
The determining submodule, be also used to be obtained according to the total degree that the coordinate after normalization is clicked the target to Amount.
12. according to any device of claim 9 to 11, which is characterized in that the determining module is also used to according to kth The click coordinate data of a target type content determine k-th of vector to be detected, 0 < k≤n；
The determining module, the vector distance for being also used to work as k-th of vector to be detected described in described i-th vector sum to be detected are small When second distance threshold value, determine k-th of target type content and i-th of target type content reception to same type The machine click.
13. according to any device of claim 9 to 11, which is characterized in that the judgment module is also used to pass through cosine Similarity formula calculates the similarity between object vector described in described i-th vector sum to be detected, when described i-th it is to be detected When the similarity of object vector described in vector sum is lower than similarity threshold, determine that i-th of target type content is to receive The content that the machine is clicked.
14. a kind of server, which is characterized in that the server includes processor and memory, is stored in the memory At least one instruction, at least a Duan Chengxu, code set or instruction set, at least one instruction, an at least Duan Chengxu, institute Code set or instruction set is stated to be loaded by the processor and executed to realize that machine as described in any of the claims 1 to 8 is clicked Detection method.
15. a kind of computer readable storage medium, which is characterized in that be stored at least one instruction, extremely in the storage medium A few Duan Chengxu, code set or instruction set, at least one instruction, an at least Duan Chengxu, the code set or instruction Collection is loaded by the processor and is executed the detection method to realize machine click as described in any of the claims 1 to 8.
CN201811265801.5A 2018-10-29 2018-10-29 Method and device for detecting machine click and readable storage medium Active CN110210507B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811265801.5A CN110210507B (en) 2018-10-29 2018-10-29 Method and device for detecting machine click and readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811265801.5A CN110210507B (en) 2018-10-29 2018-10-29 Method and device for detecting machine click and readable storage medium

Publications (2)

Publication Number Publication Date
CN110210507A true CN110210507A (en) 2019-09-06
CN110210507B CN110210507B (en) 2023-06-02

Family

ID=67779861

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811265801.5A Active CN110210507B (en) 2018-10-29 2018-10-29 Method and device for detecting machine click and readable storage medium

Country Status (1)

Country Link
CN (1) CN110210507B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111722982A (en) * 2020-06-19 2020-09-29 广州万唯邑众信息科技有限公司 Method and device for detecting whether machine clicks or not
CN111953557A (en) * 2020-07-08 2020-11-17 北京明略昭辉科技有限公司 Method and device for identifying abnormal traffic of advertisement point positions
CN112540904A (en) * 2020-12-15 2021-03-23 北京百度网讯科技有限公司 Machine operation behavior recognition method and device, electronic equipment and computer medium
CN117041064A (en) * 2023-10-10 2023-11-10 深圳安天网络安全技术有限公司 Topology node connection method, storage medium and electronic equipment

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090299967A1 (en) * 2008-06-02 2009-12-03 Microsoft Corporation User advertisement click behavior modeling
US20110113388A1 (en) * 2008-04-22 2011-05-12 The 41St Parameter, Inc. Systems and methods for security management based on cursor events
CN106447383A (en) * 2016-08-30 2017-02-22 杭州启冠网络技术有限公司 Cross-time multi-dimensional abnormal data monitoring method and system
CN106649372A (en) * 2015-10-30 2017-05-10 北京国双科技有限公司 Display method and device for advertisement clicks in thermodynamic diagram
CN106817235A (en) * 2015-11-30 2017-06-09 北京国双科技有限公司 The detection method and device of website abnormal visit capacity
CN107330731A (en) * 2017-06-30 2017-11-07 北京京东尚科信息技术有限公司 It is a kind of to recognize that advertisement position clicks on abnormal method and apparatus
CN107566192A (en) * 2017-10-18 2018-01-09 中国联合网络通信集团有限公司 A kind of abnormal flow processing method and Network Management Equipment
CN108400972A (en) * 2018-01-30 2018-08-14 北京兰云科技有限公司 A kind of method for detecting abnormality and device

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110113388A1 (en) * 2008-04-22 2011-05-12 The 41St Parameter, Inc. Systems and methods for security management based on cursor events
US20090299967A1 (en) * 2008-06-02 2009-12-03 Microsoft Corporation User advertisement click behavior modeling
CN106649372A (en) * 2015-10-30 2017-05-10 北京国双科技有限公司 Display method and device for advertisement clicks in thermodynamic diagram
CN106817235A (en) * 2015-11-30 2017-06-09 北京国双科技有限公司 The detection method and device of website abnormal visit capacity
CN106447383A (en) * 2016-08-30 2017-02-22 杭州启冠网络技术有限公司 Cross-time multi-dimensional abnormal data monitoring method and system
CN107330731A (en) * 2017-06-30 2017-11-07 北京京东尚科信息技术有限公司 It is a kind of to recognize that advertisement position clicks on abnormal method and apparatus
CN107566192A (en) * 2017-10-18 2018-01-09 中国联合网络通信集团有限公司 A kind of abnormal flow processing method and Network Management Equipment
CN108400972A (en) * 2018-01-30 2018-08-14 北京兰云科技有限公司 A kind of method for detecting abnormality and device

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111722982A (en) * 2020-06-19 2020-09-29 广州万唯邑众信息科技有限公司 Method and device for detecting whether machine clicks or not
CN111953557A (en) * 2020-07-08 2020-11-17 北京明略昭辉科技有限公司 Method and device for identifying abnormal traffic of advertisement point positions
CN112540904A (en) * 2020-12-15 2021-03-23 北京百度网讯科技有限公司 Machine operation behavior recognition method and device, electronic equipment and computer medium
CN112540904B (en) * 2020-12-15 2023-06-23 北京百度网讯科技有限公司 Machine operation behavior recognition method, device, electronic equipment and computer medium
CN117041064A (en) * 2023-10-10 2023-11-10 深圳安天网络安全技术有限公司 Topology node connection method, storage medium and electronic equipment
CN117041064B (en) * 2023-10-10 2023-12-08 深圳安天网络安全技术有限公司 Topology node connection method, storage medium and electronic equipment

Also Published As

Publication number Publication date
CN110210507B (en) 2023-06-02

Similar Documents

Publication Publication Date Title
CN110210507A (en) Detection method, device and the readable storage medium storing program for executing that machine is clicked
US11328331B2 (en) Methods, devices, and systems for sending and receiving virtual goods
CN104378341B (en) Template acquisition methods, template provider method, Apparatus and system
JP5901640B2 (en) System, method and computer readable medium for distributing target data using anonymous profile
CN108111554B (en) Control method and device for access queue
CN110163474A (en) A kind of method and apparatus of task distribution
WO2019137230A1 (en) Location information processing method and apparatus
CN110225104A (en) Data capture method, device and terminal device
CN109345417B (en) Online assessment method and terminal equipment for business personnel based on identity authentication
CN110830445B (en) Method and device for identifying abnormal access object
CN108711085A (en) A kind of response method and its equipment of transaction request
CN110390584A (en) A kind of recognition methods of abnormal user, identification device and readable storage medium storing program for executing
CN109376078A (en) Test method, terminal device and the medium of mobile application
CN106330837A (en) Suspicious network user identification method and device
US20130282583A1 (en) Fraud detection system rule profile interaction
CN110049041A (en) The interface call method and interface calling system of recruitment website open platform
US20140180765A1 (en) Web-based survey verification
US20190080360A1 (en) Technologies for propogating advertisements
CN112181628B (en) Resource transfer method, device, system and electronic equipment
US11107139B1 (en) Computing system learning of a merchant category code
CN109902698A (en) Information generating method and device
CN107748988A (en) A kind of method and system that red packet is obtained by POS
CN114626915A (en) Commodity recommendation activity processing method
CN111105301B (en) Information processing method, terminal, server and storage medium
CN109313771A (en) The degree of reiability of real-time auction

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