CN111738608A - Channel scoring method and system - Google Patents

Channel scoring method and system Download PDF

Info

Publication number
CN111738608A
CN111738608A CN202010599185.8A CN202010599185A CN111738608A CN 111738608 A CN111738608 A CN 111738608A CN 202010599185 A CN202010599185 A CN 202010599185A CN 111738608 A CN111738608 A CN 111738608A
Authority
CN
China
Prior art keywords
channel
index
scoring
activity
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010599185.8A
Other languages
Chinese (zh)
Inventor
刘颖慧
刘楠
蔡一欣
张溶芳
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.)
China United Network Communications Group Co Ltd
Original Assignee
China United Network Communications Group 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 China United Network Communications Group Co Ltd filed Critical China United Network Communications Group Co Ltd
Priority to CN202010599185.8A priority Critical patent/CN111738608A/en
Publication of CN111738608A publication Critical patent/CN111738608A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Game Theory and Decision Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a channel scoring method and a channel scoring system. The method comprises the following steps: collecting channel data based on channel dimensions; the channel data is generated by a user through channel operation; acquiring an index actual value of a grading index of a channel according to the channel data; the scoring indexes comprise activity scoring indexes and service scoring indexes; and obtaining the score of the channel based on the index actual value. The method and the system can provide scientific basis for the operator to select the activity release channel, assist the operator to improve the activity conversion rate and increase the marketing benefit.

Description

Channel scoring method and system
Technical Field
The invention relates to the technical field of communication, in particular to a channel scoring method and a channel scoring system.
Background
The operator will often target users to promote activities that require placement in a relevant channel, such as H5 page, public numbers or applets. It should be noted that different delivery channels have larger influence difference on the campaign conversion rate, and the campaign delivery performed by the operator in a suitable and high-quality channel can effectively improve the campaign conversion rate and increase the marketing benefit.
However, because a method for effectively scoring channels is lacked at present, operators are difficult to judge whether channels are suitable or high-quality channels, and cannot scientifically select suitable and high-quality channels for campaign delivery, so that campaign conversion rate is low and marketing benefit is poor.
Disclosure of Invention
Therefore, the invention provides a channel scoring method and a channel scoring system, which aim to solve the problem that in the prior art, an operator is difficult to scientifically select a channel for putting activities due to the lack of a method for effectively scoring the channel.
In order to achieve the above object, a first aspect of the present invention provides a channel scoring method, including:
collecting channel data based on channel dimensions; the channel data is generated by a user through channel operation;
acquiring an index actual value of a grading index of a channel according to the channel data; the scoring indexes comprise activity scoring indexes and service scoring indexes;
and obtaining the score of the channel based on the index actual value.
Preferably, the step of obtaining the channel score based on the actual value of the index includes:
obtaining an index score corresponding to the grading index according to the actual index value;
clustering all the channels based on the index scores to generate a target clustering result;
obtaining channel attributes corresponding to different target clustering results based on the target clustering results;
obtaining the correlation coefficient of the grading index and the target clustering result;
obtaining a score of the channel based on the indicator score of the scoring indicator, the correlation coefficient, and the channel attribute.
Preferably, the step of obtaining channel attributes corresponding to different target clustering results based on the target clustering results includes:
extracting an index actual value of a scoring index of a channel in the target clustering result;
calculating a difference value between the mean value of the index actual value and the median of the index actual value, and calculating a ratio of the difference value to the median;
sorting the scoring indexes from large to small according to the ratio;
counting the number of the activity scoring indexes and the number of the service scoring indexes in scoring indexes before presetting the ranking; when the number of the activity scoring indexes is larger than that of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as an activity channel; when the number of the activity scoring indexes is equal to the number of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as a comprehensive channel; and when the number of the activity scoring indexes is smaller than that of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as a service channel.
Preferably, the step of clustering all the channels based on the index scores includes:
and clustering all the channels by adopting a density-based noisy spatial clustering algorithm based on the index scores.
Preferably, after obtaining the score of the channel based on the actual value of the index, the method further includes:
receiving activity data for an activity;
and recommending a channel for delivering the activity to the user according to the activity data and the score.
Preferably, the channel dimensions include H5 page dimensions, kiosk dimensions, application dimensions, applet dimensions, and/or public dimensions; the channel data comprises a user mobile phone number, a network positioning address, access time and a channel identifier.
Preferably, the activity scoring indexes include total activity click rate, average activity conversion rate, average channel activity conversion cost and average active days; the service scoring indexes comprise a channel daily average activity index, a channel maintenance cost index, a channel satisfaction index and a channel external openness index.
A second aspect of the present invention provides a channel scoring system, including:
the data acquisition module is used for acquiring channel data based on the channel dimension; the channel data is generated by a user through channel operation;
the index analysis module is used for acquiring an index actual value of a grading index of a channel according to the channel data; the scoring indexes comprise activity scoring indexes and service scoring indexes;
and the channel scoring module is used for obtaining the score of the channel based on the actual value of the index.
Preferably, the system further comprises:
a first receiving module for receiving activity data of an activity;
and the channel recommending module is used for recommending a channel for delivering the activity to the user according to the activity data and the score.
Preferably, the system further comprises:
the first calculation module is used for obtaining an index score corresponding to the grading index according to the actual index value;
the clustering module is used for clustering all the channels based on the index scores to generate a target clustering result;
the attribute analysis module is used for obtaining channel attributes corresponding to different target clustering results based on the target clustering results;
the first acquisition module is used for acquiring the correlation coefficient of the scoring index and the target clustering result;
the channel scoring module is used for obtaining the score of the channel based on the index score, the correlation coefficient and the channel attribute of the scoring index.
The invention has the following advantages:
the invention provides a channel scoring method, which comprises the following steps of firstly, acquiring channel data based on channel dimensions, wherein the channel data are generated by a user through channel operation; and then, acquiring an index actual value of a scoring index of the channel according to the channel data and acquiring a score of the channel based on the index actual value, wherein the scoring index comprises an activity scoring index and a service scoring index, and the related scoring indexes are diversified and comprehensive, so that the scoring accuracy of the channel acquired based on the index actual value corresponding to the scoring index is high, a scientific basis can be provided for an operator to select an activity delivery channel, the operator is assisted to improve the activity conversion rate, and the marketing benefit is increased.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
Fig. 1 is a flowchart of a channel scoring method according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for obtaining a channel score based on an actual index value according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a channel scoring system according to an embodiment of the present invention.
In the drawings:
31: the data acquisition module 32: index analysis module
33: channel scoring module 34: first receiving module
35: channel recommendation module
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
The operator will often target users to promote activities that require placement in a relevant channel, such as H5 page, public numbers or applets. Different delivery channels have larger influence difference on the activity conversion rate, and the activity delivery of operators in a proper and high-quality channel can effectively improve the activity conversion rate and increase the marketing benefit.
However, due to the lack of a method for effectively scoring channels, operators cannot scientifically select appropriate and high-quality channels for campaign delivery, which results in low campaign conversion rate and poor marketing benefit.
In order to solve the above problem, the present embodiment provides a channel scoring method, as shown in fig. 1, the method includes the following steps:
step S101, channel data are collected based on channel dimensions.
The channel is a contact point for business handling, activity participation or information browsing of a user, and comprises an H5 page, a self-service terminal, an application program, an applet, a public number and/or the like, namely the channel dimension in the embodiment comprises an H5 page dimension, a self-service terminal dimension, an application program dimension, an applet dimension and/or a public number dimension; the channel data is generated by the user through channel operation, and comprises the user mobile phone number, the accessed network positioning address, the accessed time and the channel identification.
In one embodiment, the method for collecting channel data based on H5 page dimension by the channel scoring system comprises the following steps: firstly, filtering historical network data to obtain historical access data; wherein the historical network data is data stored in a data storage subsystem of the internet server. Channel data is then collected from the historical access data. The channel data comprises a user mobile phone number, a network location address, access time and a channel identifier, wherein the network location address is represented by a Uniform Resource Locator (URL). It should be noted that, if the user jumps to the H5 page through other pages, the collected channel data further includes a user source, wherein the user source is a short message, an application program or other network location address. It should be further noted that, if the network location address accessed by the user corresponds to a page of an activity released by the operator, the collected channel data further includes an activity identifier, and the activity identifier is a unique identifier of the activity.
In another embodiment, a method for a channel scoring system to collect channel data based on a kiosk dimension, an application dimension, a applet dimension, and/or a public dimension includes: firstly, extracting historical access data from a background management system corresponding to a self-service terminal, an application program, an applet and/or a public number; then, channel data is collected from the historical access data, wherein the channel data comprises a user mobile phone number, a network location address, access time and a channel identifier, the network location address is represented by a Uniform Resource Locator (URL) or a page code, and the channel identifier is a self-service terminal, an application program, an applet and/or a public number. It should be noted that, if the user jumps to the page of the application program, the applet and/or the public number through other pages, the collected channel data further includes a user source, where the user source is a short message, an application program or other network location address. It should be further noted that, if the network location address accessed by the user corresponds to a page of an activity released by the operator, the channel data further includes an activity identifier, and the activity identifier is a unique identifier of the activity.
And S102, acquiring an index actual value of a grading index of the channel according to the channel data.
The scoring indexes comprise activity scoring indexes and business scoring indexes, wherein the activity scoring indexes comprise total activity click rate, average activity conversion rate, average channel activity conversion cost and average active days; the service scoring indexes comprise a channel daily average activity index, a channel maintenance cost index, a channel satisfaction index and a channel external openness index.
In one embodiment, a channel puts N (N is a positive integer) campaigns for the preset time period, labeled campaign Ei(i∈[1，N]) Wherein, activity EiThere are M (M is a positive integer) active pages, marked as active page Kij(i∈[1，N]，j∈[1，M]) I.e. active page KijIndicating that the active page is the jth active page of the ith activity. The channel not only contains a movable page KijAnd further comprises P (P is a positive integer)) Channel page Ws(s∈[1，P]). In this embodiment, the channel data includes a user mobile phone number, a network location address, access time, a channel identifier, an activity identifier, and a user source; the preset time period comprises an integer number of months; the step of the channel scoring system obtaining the index actual value of the scoring index of the channel according to the channel data comprises:
first, the channel scoring system obtains an actual value of an activity scoring index. The activity scoring indexes comprise total click rate of channel activities, average activity conversion rate of channels, average conversion cost of channel activities and average active days of channel activities. The method comprises the following specific steps:
firstly, a channel scoring system counts access activity pages K in a preset time periodijNumber R of mobile phone numbers of userij(i∈[1，N]，j∈[1，M])。
Second step, based on RijAnd obtaining the total click rate B of the channel activities in the preset time period. Wherein, the total click rate of channel activities is defined as the sum of the click rates of all activities in a channel, and it should be noted that the click rate of activities is defined as the access activity EiFirst active page K ofi1Number R of mobile phone numbers of useri1I.e. total click rate of channel campaign
Figure BDA0002558542510000061
i∈[1，N]，j∈[1，M]。
Thirdly, the channel scoring system is based on RijAnd obtaining the channel average activity conversion rate A in a preset time period. Wherein, activity EiActivity conversion of equal to RiMAnd Ri1Ratio of (A) to (B), RiMRepresenting access activity EiLast active page K ofiMThe number of the mobile phone numbers of the users is equivalent to successfully participating in the activity EiThe number of users of (1); ri1Representing access activity EiFirst active page K ofi1The number of mobile phone numbers of the user is equivalent to activity EiThe total number of users covered. Since the channel puts N (N is a positive integer) activities in total within a preset time period, therefore:
average activity conversion rate of channel
Figure BDA0002558542510000071
i∈[1，N]。
Fourth, the channel scoring system is based on activity EiThe activity conversion rate of the channel obtains the average conversion cost C of the channel activity in the preset time period. Specifically, the channel scoring system obtains an activity E corresponding to an activity identifier from an operator management system according to the activity identifier contained in the channel dataiWherein the cost data comprises a total cost of activity CiTotal cost of activity CiConsists of activity investment cost and activity labor cost. Total cost of activity CiObtaining the average conversion cost of channel activities:
average conversion cost of channel campaign
Figure BDA0002558542510000072
i∈[1，N]。
Fifthly, the channel scoring system is based on RijObtaining the average active days of channel activities in a preset time period
Figure BDA0002558542510000076
Specifically, the number of months in a preset time period is Z, and an activity E is selected according to the month from the preset time periodiActive click volume per day Ri1Obtaining Z sample sets as sample sets; obtaining the average value of the click rate samples of each sample set based on the sample sets, and counting the days D that the active click rate in the sample sets is greater than the corresponding sample average valuelWherein l ∈ [1, Z]And then:
average number of active days of channel activity
Figure BDA0002558542510000073
Secondly, the channel scoring system obtains the actual value of the service scoring index. The service scoring indexes comprise a channel daily average activity index, a channel maintenance cost index, a channel satisfaction index and a channel external openness index. The method comprises the following specific steps:
firstly, a channel scoring system accesses a channel page W in a preset time period according to a channel data systemsNumber R of mobile phone numbers of users(s∈[1，P]) And a channel page W in a preset time periodsNumber of clicks F (F is an integer) and is based on RsOr F, acquiring the actual value of the channel daily average activity index in the preset time period. The number of days contained in the preset time period is Q (Q is an integer), the channel daily average activity index comprises channel daily average activity times U1 and channel daily average activity user number U2, and then:
daily number of channel activation
Figure BDA0002558542510000074
Number of active channel users per day
Figure BDA0002558542510000075
s∈[1，P]。
And secondly, extracting channel maintainer data corresponding to the channel by the channel scoring system according to the channel identification contained in the channel data, and calculating the actual value of the channel maintenance cost index.
And thirdly, the channel scoring system sets sampling investigation according to the channel identification contained in the channel data to obtain the index actual value of the channel satisfaction index. The actual value of the channel satisfaction index is expressed in percentage.
And fourthly, the channel scoring system obtains an index actual value of the channel external openness degree index according to the user source contained in the channel data. And the channel external openness degree index comprises the channel daily average skip times and the channel daily average skip user number.
It should be noted that, in this embodiment, the scoring index for measuring the performance of the channel includes an activity scoring index and a service scoring index, and the scoring index is diversified and comprehensive, and can measure not only the performance of the channel but also the performance of the channel combined with the activity, so that scoring of the channel based on the scoring index can improve scoring accuracy, and provide a scientific basis for an operator to select an activity delivery channel.
And step S103, scoring the channel based on the index actual value.
The index actual value comprises an index actual value of the activity evaluation index and an index actual value of the business evaluation index.
In one embodiment, as shown in fig. 2, the scoring system obtains the score of the channel based on the actual value of the index, and includes:
step S201, according to the obtained index score corresponding to the grading index.
The index score corresponding to the scoring index is the ratio of the actual index value to the optimal index value, and the optimal index value is a value which is expected to be reached by the scoring index and is preset by a channel scoring system. In one embodiment, in order to observe and utilize the actual values of the indexes more intuitively and conveniently, the channel scoring system constructs a scoring matrix according to the actual values of the indexes, wherein the scoring matrix is an h x t-order matrix, rows of the scoring matrix are the actual values of the indexes of different scoring indexes belonging to the same channel, and columns of the scoring matrix are the actual values of the indexes of the same scoring index belonging to different channels; and obtaining an index score matrix of corresponding index scores according to the scoring matrix.
And step S202, clustering all channels based on the index scores to generate a target clustering result. It should be noted that the target clustering result includes a plurality of channels. In one embodiment, when the channel scoring system performs clustering, a Density-Based Spatial clustering of applications with Noise (DBSCAN) algorithm may be used to cluster all channels Based on the index scores. In another embodiment, the channel scoring system clusters all channels and uses the profile coefficients to select the optimal clustering result as the target clustering result.
Step S203, channel attributes corresponding to different target clustering results are obtained based on the target clustering results.
It should be noted that the target clustering result includes multiple channels, channels included in the same target clustering result have the same channel attribute, and channels in different target clustering results have different channel attributes. The channel attributes comprise an activity channel, a business channel and a comprehensive channel.
In order to accurately obtain channel attributes corresponding to different target clustering results based on the target clustering results, the embodiment provides a method for obtaining channel attributes corresponding to the target clustering results, which comprises the following steps:
firstly, extracting an index actual value of a scoring index of a channel in a target clustering result; and secondly, calculating the difference value between the mean value of the index actual value and the median of the index actual value, and calculating the ratio of the difference value to the median. Then, the scoring indexes are sorted from large to small according to the ratio. Finally, counting the number of activity scoring indexes and the number of service scoring indexes in scoring indexes before the preset name; when the number of the activity scoring indexes is larger than that of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as an activity channel; when the number of the activity scoring indexes is equal to the number of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as a comprehensive channel; and when the number of the activity scoring indexes is smaller than that of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as a service channel.
It should be noted that, since the mean value refers to each of the index actual values of the evaluation index, the median does not take into account the extreme values in the index actual values. Therefore, the mean value minus the median value obtains a difference value, and the ratio obtained by the ratio of the difference value to the median value can mark the influence of the extreme value in the index actual value on the overall index actual value. The larger the influence, the more the larger the data in the actual value of the index corresponding to the evaluation index, that is, the more important the evaluation index is.
And step S204, obtaining the correlation coefficient of the grading index and the target clustering result.
It should be noted that, since the target clustering result is obtained by clustering the index score corresponding to the index actual value of the score index by the channel scoring system, the score index and the target clustering result have a correlation, and the correlation is represented by a correlation coefficient.
Step S205, the score of the channel is obtained based on the index score of the score index, the correlation coefficient and the channel attribute.
Wherein the scoring indexes comprise activity scoring indexes and service scoring indexes; the channel attributes include an activity channel, a business channel, and a composite channel. In one embodiment, the set of scoring indexes is Y, and the set Y includes v (v is a positive integer) scoring indexes, where the set of activity scoring indexes is YE, the set of business scoring indexes is YP, and the index score corresponding to the scoring index Y is XyThe correlation coefficient corresponding to the score index y is RyAnd then:
rating of campaign channel { R ═ Ry*Xy*0.6|y∈YE}+{Ry*Xy0.4| y ∈ YP }, wherein 0.6 and 0.4 are weights preset by the channel scoring system.
Rating of a traffic channel { R ═ Ry*Xy*0.4|y∈YE}+{Ry*Xy0.6| y ∈ YP }, wherein 0.6 and 0.4 are weights preset by the channel scoring system.
Figure BDA0002558542510000101
In one embodiment, in order to maximize the channel value, assist the operator in improving the activity conversion rate and increasing the marketing benefit, the channel scoring system may further receive activity data of the activity, which is from the operator system and is data of the activity that the operator prepares to release; and after receiving the activity data of the activity, the channel scoring system recommends a channel for delivering the activity to the user according to the activity data and the score of the channel.
The embodiment provides a channel scoring method, which includes the steps that channel data are collected based on channel dimensions, and the channel data are generated by a user through channel operation; and then, acquiring an index actual value of a scoring index of the channel according to the channel data and acquiring a score of the channel based on the index actual value, wherein the scoring index comprises an activity scoring index and a service scoring index, and the related scoring indexes are diversified and comprehensive, so that the scoring accuracy of the channel acquired based on the index actual value corresponding to the scoring index is high, a scientific basis can be provided for an operator to select an activity delivery channel, the operator is assisted to improve the activity conversion rate, and the marketing benefit is increased.
The present embodiment further provides a channel scoring system, as shown in fig. 3, the system includes: a data collection module 31, an index analysis module 32, a channel scoring module 33, a first receiving module 34, and a channel recommendation module 35.
Wherein, the data collection module 31 is configured to collect channel data based on the channel dimension.
The channel is a contact point for business handling, activity participation or information browsing of a user, and comprises an H5 page, a self-service terminal, an application program, an applet, a public number and/or the like, namely the channel dimension in the embodiment comprises an H5 page dimension, a self-service terminal dimension, an application program dimension, an applet dimension and/or a public number dimension; the channel data is generated by the user through channel operation, and comprises the user mobile phone number, the accessed network positioning address, the accessed time and the channel identification.
In one embodiment, the data collection module 31 includes a filtering configuration sub-module and a first collection sub-module, wherein the filtering configuration sub-module and the first collection sub-module are configured in the internet server. When the data acquisition module 31 acquires channel data based on the H5 page dimension: firstly, filtering historical access data from historical network data by a filtering configuration submodule; wherein the historical network data is data stored in a data storage subsystem of the internet server. Then, the first collecting submodule collects channel data from the historical visiting data. The channel data comprises a user mobile phone number, a network location address, access time and a channel identifier, wherein the network location address is represented by a Uniform Resource Locator (URL). It should be noted that, if the user jumps to the H5 page through another page, the channel data collected by the first collecting sub-module further includes a user source, where the user source is a short message, an application program, or another network location address. It should be further noted that, if the network location address accessed by the user corresponds to a page of an activity released by the operator, the channel data collected by the first collection sub-module further includes an activity identifier, and the activity identifier is a unique identifier of the activity.
In another embodiment, the data acquisition module 31 includes a first extraction sub-module and a second acquisition sub-module. The method for acquiring channel data by the data acquisition module 31 channel scoring system based on the self-service terminal dimension, the application program dimension, the small program dimension and/or the public number dimension comprises the following steps: firstly, extracting historical access data from a background management system corresponding to a self-service terminal, an application program, an applet and/or a public number by a first extraction submodule; then, the second collecting submodule collects channel data from the historical access data, wherein the channel data comprises a user mobile phone number, a network location address, access time and a channel identifier, the network location address is represented by a Uniform Resource Locator (URL) or a page code, and the channel identifier is a self-service terminal, an application program, an applet and/or a public number. It should be noted that, if the user jumps to the page of the application program, the applet and/or the public number through another page, the channel data collected by the second collection sub-module further includes a user source, where the user source is a short message, an application program or another network location address. It should be further noted that, if the network location address accessed by the user corresponds to a page of an activity released by the operator, the channel data collected by the second collection sub-module further includes an activity identifier, and the activity identifier is a unique identifier of the activity.
And the index analysis module 32 is used for acquiring an index actual value of the grading index of the channel according to the channel data. The scoring indexes comprise activity scoring indexes and business scoring indexes, wherein the activity scoring indexes comprise total activity click rate, average activity conversion rate, average channel activity conversion cost and average active days; the service scoring indexes comprise a channel daily average activity index, a channel maintenance cost index, a channel satisfaction index and a channel external openness index.
And a channel scoring module 33 for scoring the channel based on the index actual value. The index actual value comprises an index actual value of the activity evaluation index and an index actual value of the business evaluation index. In one embodiment, the channel scoring system further comprises: the device comprises a first calculation module, a clustering module, an attribute analysis module and a first acquisition module.
The first calculation module is used for obtaining an index score corresponding to the grading index according to the actual index value. The index score corresponding to the scoring index is the ratio of the actual index value to the optimal index value, and the optimal index value is a value which is expected to be reached by the scoring index and is preset by a channel scoring system. In one embodiment, in order to observe and utilize the index actual value more intuitively and conveniently, a first calculation module of the channel scoring system constructs a scoring matrix according to the index actual value, wherein the scoring matrix is an h x t-order matrix, rows of the scoring matrix are index actual values of different scoring indexes belonging to the same channel, and columns of the scoring matrix are index actual values of the same scoring index belonging to different channels; and obtaining an index score matrix of corresponding index scores according to the scoring matrix.
And the clustering module is used for clustering all channels based on the index scores to generate a target clustering result. It should be noted that the target clustering result includes a plurality of channels. In one embodiment, when the channel scoring system performs clustering, a Density-Based Spatial clustering of applications with Noise (DBSCAN) algorithm may be used to cluster all channels Based on the index scores. In another embodiment, the channel scoring system clusters all channels and uses the profile coefficients to select the optimal clustering result as the target clustering result.
In one embodiment, in order to accurately obtain channel attributes corresponding to different target clustering results based on the target clustering results, the clustering module further includes: the device comprises a cluster extraction sub-module, a cluster calculation sub-module, a cluster sequencing sub-module, a cluster counting sub-module and a cluster analysis sub-module. Firstly, extracting an index actual value of a scoring index of a channel in a target clustering result by a clustering extraction submodule; secondly, the cluster calculation submodule calculates the difference between the mean value of the actual index values and the median of the actual index values and calculates the ratio of the difference to the median, and it needs to be noted that the mean value relates to each value in the actual index values of the evaluation indexes, and the median does not consider the extreme value in the actual index values, so that the ratio obtained in the step can mark the influence of the extreme value in the actual index values on the whole actual index values, and the larger the influence, the more the larger the data in the actual index values corresponding to the evaluation indexes, the higher the importance of the evaluation indexes; then, the clustering and sorting submodule sorts the scoring indexes from large to small according to the ratio; finally, the cluster statistic submodule counts the number of the activity scoring indexes and the number of the service scoring indexes in the scoring indexes before the preset ranking; when the number of the activity scoring indexes is larger than that of the service scoring indexes, the cluster analysis submodule determines the channel attribute corresponding to the target clustering result as an activity channel; when the number of the activity scoring indexes is equal to that of the service scoring indexes, the cluster analysis submodule determines the channel attribute corresponding to the target clustering result as a comprehensive channel; and when the number of the activity scoring indexes is smaller than that of the service scoring indexes, the cluster analysis submodule determines the channel attribute corresponding to the target clustering result as a service channel.
And the attribute analysis module is used for obtaining channel attributes corresponding to different target clustering results based on the target clustering results. It should be noted that the target clustering result includes multiple channels, channels included in the same target clustering result have the same channel attribute, and channels in different target clustering results have different channel attributes. The channel attributes comprise an activity channel, a business channel and a comprehensive channel.
And the first acquisition module is used for acquiring the correlation coefficient of the scoring index and the target clustering result. It should be noted that, since the target clustering result is obtained by clustering the index score corresponding to the index actual value of the score index by the channel scoring system, the score index and the target clustering result have a correlation, and the correlation is represented by a correlation coefficient.
The channel scoring module 33 obtains a score of the channel based on the index score of the scoring index, the correlation coefficient obtained by the first obtaining module, and the channel attribute obtained by the attribute analyzing module.
In one embodiment, in order to maximize the channel value, assist the operator to improve the activity conversion rate and increase the marketing benefit, the channel assessment system receives the activity data of the activity through the first receiving module 34, it should be noted that the activity data is from the operator system and is the data of the activity that the operator prepares to release; after the first receiving module 34 receives the campaign data for the campaign, the channel recommending module 35 recommends the channel for delivering the campaign to the user based on the campaign data and the channel's score.
The working modes of the modules of the channel scoring system provided by this embodiment correspond to the steps of the channel scoring method, and therefore, the detailed working modes of the modules in the channel scoring system can be referred to the channel scoring method provided by this embodiment.
The embodiment provides a channel scoring system, which includes a data acquisition module 31 for acquiring channel data based on channel dimensions, wherein the channel data is generated by a user operating through a channel; then, the index analysis module 32 obtains an index actual value of a scoring index of the channel according to the channel data, and the channel scoring module 33 obtains a score of the channel based on the index actual value, it should be noted that the scoring index includes an activity scoring index and a service scoring index, and the related scoring indexes are diversified and comprehensive, so that the scoring accuracy of the channel obtained by the channel scoring module 33 based on the index actual value corresponding to the scoring index is high, a scientific basis can be provided for an operator to select an activity delivery channel, and the operator is assisted to improve the activity conversion rate and increase the marketing benefit.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.

Claims (10)

1. A channel scoring method, the method comprising:
collecting channel data based on channel dimensions; the channel data is generated by a user through channel operation;
acquiring an index actual value of a grading index of a channel according to the channel data; the scoring indexes comprise activity scoring indexes and service scoring indexes;
and obtaining the score of the channel based on the index actual value.
2. The method of claim 1, wherein the step of deriving the channel score based on the actual value of the indicator comprises:
obtaining an index score corresponding to the grading index according to the actual index value;
clustering all the channels based on the index scores to generate a target clustering result;
obtaining channel attributes corresponding to different target clustering results based on the target clustering results;
obtaining the correlation coefficient of the grading index and the target clustering result;
obtaining a score of the channel based on the indicator score of the scoring indicator, the correlation coefficient, and the channel attribute.
3. The method of claim 2, wherein the step of obtaining channel attributes corresponding to different target clustering results based on the target clustering results comprises:
extracting an index actual value of a scoring index of a channel in the target clustering result;
calculating a difference value between the mean value of the index actual value and the median of the index actual value, and calculating a ratio of the difference value to the median;
sorting the scoring indexes from large to small according to the ratio;
counting the number of the activity scoring indexes and the number of the service scoring indexes in scoring indexes before presetting the ranking; when the number of the activity scoring indexes is larger than that of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as an activity channel; when the number of the activity scoring indexes is equal to the number of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as a comprehensive channel; and when the number of the activity scoring indexes is smaller than that of the service scoring indexes, determining the channel attribute corresponding to the target clustering result as a service channel.
4. The method of claim 2, wherein the step of clustering all of the channels based on the indicator scores comprises:
and clustering all the channels by adopting a density-based noisy spatial clustering algorithm based on the index scores.
5. The method of claim 1, wherein after obtaining the channel score based on the actual indicator value, the method further comprises:
receiving activity data for an activity;
and recommending a channel for delivering the activity to the user according to the activity data and the score.
6. The method of claim 1, wherein the channel dimensions include H5 page dimensions, kiosk dimensions, application dimensions, applet dimensions, and/or public dimensions; the channel data comprises a user mobile phone number, a network positioning address, access time and a channel identifier.
7. The method of claim 1, wherein the campaign scoring metrics comprise total campaign click-through, average campaign conversion rate, average channel campaign conversion cost, and average number of active days; the service scoring indexes comprise a channel daily average activity index, a channel maintenance cost index, a channel satisfaction index and a channel external openness index.
8. A channel scoring system, the system comprising:
the data acquisition module is used for acquiring channel data based on the channel dimension; the channel data is generated by a user through channel operation;
the index analysis module is used for acquiring an index actual value of a grading index of a channel according to the channel data; the scoring indexes comprise activity scoring indexes and service scoring indexes;
and the channel scoring module is used for obtaining the score of the channel based on the actual value of the index.
9. The system of claim 8, further comprising:
a first receiving module for receiving activity data of an activity;
and the channel recommending module is used for recommending a channel for delivering the activity to the user according to the activity data and the score.
10. The system of claim 8, further comprising:
the first calculation module is used for obtaining an index score corresponding to the grading index according to the actual index value;
the clustering module is used for clustering all the channels based on the index scores to generate a target clustering result;
the attribute analysis module is used for obtaining channel attributes corresponding to different target clustering results based on the target clustering results;
the first acquisition module is used for acquiring the correlation coefficient of the scoring index and the target clustering result;
the channel scoring module is used for obtaining the score of the channel based on the index score, the correlation coefficient and the channel attribute of the scoring index.
CN202010599185.8A 2020-06-28 2020-06-28 Channel scoring method and system Pending CN111738608A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010599185.8A CN111738608A (en) 2020-06-28 2020-06-28 Channel scoring method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010599185.8A CN111738608A (en) 2020-06-28 2020-06-28 Channel scoring method and system

Publications (1)

Publication Number Publication Date
CN111738608A true CN111738608A (en) 2020-10-02

Family

ID=72651483

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010599185.8A Pending CN111738608A (en) 2020-06-28 2020-06-28 Channel scoring method and system

Country Status (1)

Country Link
CN (1) CN111738608A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107133734A (en) * 2017-04-28 2017-09-05 浙江极赢信息技术有限公司 A kind of Channel Quality evaluation method and system
CN108764332A (en) * 2018-05-25 2018-11-06 北京证大向上金融信息服务有限公司 A kind of Channel Quality analysis method, computing device and storage medium
CN109583721A (en) * 2018-11-16 2019-04-05 北京奇虎科技有限公司 A kind of appraisal procedure of channel, device and electronic equipment
CN109670855A (en) * 2018-11-02 2019-04-23 北京奇虎科技有限公司 The methods of marking and device of information flow platform author

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107133734A (en) * 2017-04-28 2017-09-05 浙江极赢信息技术有限公司 A kind of Channel Quality evaluation method and system
CN108764332A (en) * 2018-05-25 2018-11-06 北京证大向上金融信息服务有限公司 A kind of Channel Quality analysis method, computing device and storage medium
CN109670855A (en) * 2018-11-02 2019-04-23 北京奇虎科技有限公司 The methods of marking and device of information flow platform author
CN109583721A (en) * 2018-11-16 2019-04-05 北京奇虎科技有限公司 A kind of appraisal procedure of channel, device and electronic equipment

Similar Documents

Publication Publication Date Title
CN107451861B (en) Method for identifying user internet access characteristics under big data
CN106790487B (en) Method, device and system for displaying help information
CN106529714A (en) Method and system predicting user loss
CN111666351A (en) Fuzzy clustering system based on user behavior data
CN111737646B (en) Advertisement promotion effect evaluation data processing method, system and storage medium
CN103150696A (en) Method and device for selecting potential customer of target value-added service
CN106843941B (en) Information processing method, device and computer equipment
CN106375369A (en) Mobile Web service recommendation method and collaborative recommendation system based on user behavior analysis
CN107896153B (en) Traffic package recommendation method and device based on mobile user internet surfing behavior
US20090259525A1 (en) Internet Probability Sampling
CN107093092B (en) Data analysis method and device
CN109995549B (en) Method and device for evaluating flow value
CN117352114B (en) Recommendation method and system of psychological assessment scale based on clustering algorithm
CN107016561B (en) Information processing method and device
CN113411453A (en) Intelligent management method and device of outbound object, medium and computing equipment
CN111738608A (en) Channel scoring method and system
JP4308683B2 (en) User activity history visualization / analysis method, user activity history visualization / analysis device, and program
CN116340374A (en) Personalized task recommendation method and system
CN115719183A (en) Power customer self-feedback service evaluation method and system based on weight dynamic grading
CN111752991B (en) Meter data statistics method and device and meter statistics data acquisition method
CN112396313A (en) Method for optimizing telephone sales performance by using smart watch
CN112612934A (en) User charging behavior portrait processing method and device
CN109710876B (en) Information recommendation method and device and computer readable storage medium
CN112863231B (en) Method, system and device for calibrating remaining parking spaces of parking lot and storage medium
CN111400568B (en) Behavior intention analysis method and device, electronic equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20201002

RJ01 Rejection of invention patent application after publication