CN115408586B - Intelligent channel operation data analysis method, system, equipment and storage medium - Google Patents
Intelligent channel operation data analysis method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN115408586B CN115408586B CN202211025826.4A CN202211025826A CN115408586B CN 115408586 B CN115408586 B CN 115408586B CN 202211025826 A CN202211025826 A CN 202211025826A CN 115408586 B CN115408586 B CN 115408586B
- Authority
- CN
- China
- Prior art keywords
- user
- data
- channel
- path
- retention
- 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.)
- Active
Links
- 238000007405 data analysis Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000003860 storage Methods 0.000 title claims description 12
- 238000006243 chemical reaction Methods 0.000 claims abstract description 154
- 230000014759 maintenance of location Effects 0.000 claims abstract description 144
- 238000005457 optimization Methods 0.000 claims abstract description 46
- 238000012545 processing Methods 0.000 claims abstract description 45
- 230000002159 abnormal effect Effects 0.000 claims abstract description 42
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 238000004140 cleaning Methods 0.000 claims abstract description 11
- 230000006399 behavior Effects 0.000 claims description 106
- 238000012384 transportation and delivery Methods 0.000 claims description 54
- 238000012502 risk assessment Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 12
- 238000013439 planning Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000004138 cluster model Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 2
- 238000012549 training Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 9
- 238000004458 analytical method Methods 0.000 description 14
- 238000012544 monitoring process Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 230000009466 transformation Effects 0.000 description 6
- 239000000463 material Substances 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000001680 brushing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004992 fission Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- 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
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The method comprises the steps of obtaining channel operation data in a full life cycle in real time, carrying out data cleaning processing on the channel operation data to obtain effective user data, carrying out user portrait processing on the effective user data to obtain user portrait data, constructing a retention curve related to channel user retention according to the user portrait data to obtain a user retention curve, identifying channel abnormal data on the user retention curve, carrying out data marking on the channel abnormal data to obtain channel marking data, calculating channel user conversion rate according to the user retention curve, and generating a strategy optimization instruction for optimizing channel throwing strategies according to calculation results so as to optimize the channel marking data through the strategy optimization instruction. The method and the device have the effect of improving the accuracy of the channel operation data analysis result.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to an intelligent channel operation data analysis method, system, equipment and storage medium.
Background
At present, along with the rapid development of big data age, the release channels of enterprises are also performing rapid iteration, common release channels such as information stream advertisements, media popularization and search engines and the like, no matter the release channels are payment channels or free channels, certain operation cost is required to be paid by the enterprises, so that higher requirements are also provided for channel operation by complex and various popularization channels, and especially in the field of electronic commerce, sales of electronic commerce products are often determined by the selection of electronic commerce product popularization channels.
The existing channel operation data analysis mode is to pull all channel operation data of the channels through a third party statistics tool to conduct data analysis, so that whether the channels are good quality channels or not is judged according to analysis results, and corresponding advertisement delivery is conducted, however, in order to attract advertisers, part of channel operators often log in at fixed time and fixed points through network water armies to provide a flow false with high click rate, and false flow caused by channel brushing amount and channel cheating often brings error influence which is difficult to ignore to channel operation data analysis, and accuracy of channel analysis results is affected.
Aiming at the related technology, the inventor considers that the defect that false channel flow influences the accuracy of channel operation data analysis results exists.
Disclosure of Invention
In order to improve accuracy of channel operation data analysis results, the application provides an intelligent channel operation data analysis method, system, equipment and storage medium.
The first object of the present invention is achieved by the following technical solutions:
an intelligent channel operation data analysis method is provided, and the intelligent channel operation data analysis method comprises the following steps:
channel operation data of a full life cycle is obtained in real time, and data cleaning processing is carried out on the channel operation data to obtain effective user data;
performing user portrait processing on the effective user data to obtain user portrait data;
constructing a retention curve about channel user retention according to the user portrait data to obtain a user retention curve;
identifying channel abnormal data on the user retention curve, and carrying out data marking on the channel abnormal data to obtain channel marking data;
and calculating the conversion rate of the channel user according to the user retention curve, and generating a strategy optimization instruction for optimizing the channel delivery strategy according to a calculation result so as to optimize the channel marking data through the strategy optimization instruction.
By adopting the technical scheme, the channel traffic cheating is generally performed uniformly through the timing simulation starting behavior and the fixed browsing behavior, so that the random browsing behavior and the click conversion behavior of a real user are difficult to generate, and therefore, the data cleaning processing is performed on the channel operation data in the full life cycle, the invalid operation data with fixed user behavior and no real conversion rate is helped to be cleaned, so that effective user data is obtained, the user image is performed on the effective user data, the error influence of false channel traffic on the user image is helped to be reduced, the data marking is performed on channel abnormal data such as inflection points, breakpoints and the like through the user retention graph, the channel abnormal data is helped to be optimized in a targeted manner, the multi-dimensional optimization is performed on the channel release strategy according to the user retention condition and the channel user conversion rate, the optimized channel release strategy is helped to be more attached to the real channel traffic, and the accuracy of the analysis result of the channel operation data is improved.
The present application may be further configured in a preferred example to: the user portrait processing is carried out on the effective user data to obtain user portrait data, which comprises the following steps:
Acquiring an original channel release strategy which accords with an enterprise channel release target;
performing data analysis processing on the original channel release strategy to obtain a user characteristic index for drawing a user portrait;
identifying a user unique identification code and user behavior data in the channel operation data;
clustering the user characteristic indexes and the user behavior data according to the unique user identification codes to obtain a user clustering model;
and inputting the effective user data into the user clustering model to obtain user portrait data.
By adopting the technical scheme, the channel delivery purposes of each enterprise are different, for example, the E-commerce enterprise is used for promoting commodity sales, the news website is used for increasing news browsing quantity, the delivery purposes are different, and the targeted target user groups are different, so that in order to improve the fit degree of user portraits and the enterprise delivery purposes, the original channel delivery strategy is analyzed, so that the user characteristic indexes meeting the enterprise delivery targets are obtained, the user data with optimal fit degree can be screened out according to the user characteristic indexes, and because the unique user identification codes have uniqueness, the false flow is difficult to simulate the unique user identification codes, the user characteristic indexes and the user behavior data are clustered and associated according to the unique user identification codes, and therefore, the user portraits data can be obtained by rapidly correlating and gathering the effective user data meeting the user characteristic indexes according to a user clustering model, and the user portraits rate of channel operation data is improved.
The present application may be further configured in a preferred example to: after the effective user data is input into the user clustering model to obtain user portrait data, the method comprises the following steps:
performing path tracking processing on the user behavior data to obtain a user behavior path;
according to the user behavior path, path planning is carried out on channel users reaching a preset conversion target, and a user conversion path is obtained;
acquiring an actual behavior path of a channel user in real time in the channel operation process according to the user conversion path;
performing path fitting processing on the actual behavior path and the user conversion path to obtain path deviation data;
generating a path optimization instruction which corresponds to the path deviation data and is used for carrying out delivery path adjustment on the original channel delivery strategy according to the path deviation data.
By adopting the technical scheme, because the user behavior paths of users of each channel are different, when the user behavior data is prejudged, the user behavior is tracked through the user behavior paths, therefore, the user transformation paths are obtained by drawing the user behavior data distributed in scattered points into visual user behavior paths and carrying out path planning in combination with the expected transformation targets, wherein the expected transformation targets are obtained by analysis in the channel release strategy, the channel release strategy is adjusted through the user transformation paths, so that the channel release content is more suitable for the user requirements, the user experience is improved, the path deviation data is obtained through fitting of the actual behavior paths and the user transformation paths, timely optimization of the channel release materials according to the path deviation data is facilitated, the channel release paths are more suitable for the requirements of the users, the release time between the expected transformation targets is shortened through dynamic adjustment of the release channels, the channel release cost is reduced, and the channel release conversion efficiency is also improved.
The present application may be further configured in a preferred example to: constructing a retention curve about channel user retention according to the user portrait data to obtain a user retention curve, wherein the method specifically comprises the following steps of:
identifying user newly-added data and user login data in the channel operation data, wherein the user login data is next-day login data of the newly-added user;
calculating the retention rate according to the newly added data of the user and the login data of the user to obtain the retention rate of the user;
and drawing a user retention curve corresponding to the user portrait data according to the user retention rate to obtain a user retention curve.
By adopting the technical scheme, because the user retention curve of the real user is usually in exponential decay change, the false flow is usually in fixed machine setting, fixed login or browsing behavior is carried out at fixed points at fixed time, the generated user retention curve is usually smooth and unchanged, and the effect of random browsing change of the real flow is difficult to simulate, therefore, the user newly-added data and the user login data under the adjacent operation period are obtained for calculation, thereby obtaining the user retention rate, being beneficial to judging whether the user is the real user according to the user retention rate, carrying out user retention curve drawing according to the user retention rate and combining the user portrait data, obtaining the user retention curve, visually observing the channel release effect and the user retention change condition through the user retention curve, carrying out faithfulness evaluation on the channel user according to the user retention change condition, and being beneficial to carrying out multi-dimensional classification on the user, and improving the judgment accuracy of the user authenticity.
The present application may be further configured in a preferred example to: the identifying the channel abnormal data on the user retention curve and carrying out data marking on the channel abnormal data to obtain channel marking data specifically comprises the following steps:
acquiring natural flow data in the channel operation data, wherein the natural flow data comprises natural user login data and natural user newly-added data;
constructing a standard retention curve related to natural flow data according to the natural user login data and the natural user newly-added data;
comparing the characteristics of the standard retention curve and the user retention curve to obtain channel abnormal data;
and carrying out data marking on the channel abnormal data according to the user characteristic indexes in the user portrait data to obtain channel marking data.
By adopting the technical scheme, because the phenomena of abrupt change of the IP address or free increase of browsing duration and the like of the user can cause the data inflection point or break point of the user retention curve to form channel abnormal data, the user retention change condition of the channel abnormal data is too random and uncontrollable, and the error compensation reference meaning of the user portrait is not great, so that in order to reduce the influence of invalid user retention data on the data analysis accuracy, the standard retention curve is drawn through the user login condition and the user newly-increased condition of natural flow data, a comparison reference basis is provided for the user retention curve conforming to the channel release strategy, abnormal data which is different from the standard retention curve in the user retention curve is obtained according to the comparison result, the channel abnormal data is marked through the user characteristic index, and the accurate compensation is carried out on the channel abnormal data which does not conform to the user characteristic index according to the channel marking data, thereby improving the error compensation accuracy of channel data analysis.
The present application may be further configured in a preferred example to: calculating the conversion rate of the channel user according to the user retention curve, and generating a strategy optimization instruction for optimizing the channel delivery strategy according to a calculation result so as to optimize the channel marking data through the strategy optimization instruction, wherein the method specifically comprises the following steps:
acquiring conversion frequency information of channel users corresponding to the effective user data in real time;
calculating the conversion frequency information and user login data conforming to the user retention curve to obtain channel user conversion rate;
constructing a user funnel model conforming to the user retention curve according to the channel user conversion rate;
and inputting the channel operation data into the user funnel model, and generating a strategy optimization instruction which corresponds to the channel user conversion rate and is used for optimizing the channel delivery strategy.
By adopting the technical scheme, the false flow generated by channel cheating is usually simulated by machine setting, so that the simulation actions of starting simulation, clicking simulation or browsing simulation and other macroscopic indexes are performed, but certain technical difficulties exist in simulating the conversion rate of the user, the false flow is usually remained only on the macroscopic index expression level, when the macroscopic index of channel operation data is excellent, the actual conversion rate of the user is very low, the possibility of high flow falsification is indicated, therefore, the user conversion rate is calculated by acquiring the user conversion frequency information corresponding to the effective user data and combining with the user login data in a user retention curve, the effectiveness of the corresponding channel operation data is judged according to the user conversion rate, a user funnel model is constructed according to the channel user conversion rate, the deep analysis of the user conversion condition is facilitated according to the user funnel model, the supervision and management are facilitated on a plurality of links in the channel operation process, the multidimensional analysis is performed on the channel data according to the user funnel model, the further optimization of the channel release strategy is facilitated according to the analysis result, the release strategy is further optimized, the release strategy is further attached to the user conversion rate is enabled, and the actual operation result of the channel operation data is improved.
The present application may be further configured in a preferred example to: the method comprises the steps of calculating the conversion rate of channel users according to the user retention curve, generating a strategy optimization instruction for optimizing the channel delivery strategy according to a calculation result, and optimizing channel marking data through the strategy optimization instruction, and further comprises:
comparing the conversion rate of the channel user with an expected conversion index preset in the channel release strategy to obtain a conversion rate comparison result;
performing risk assessment on the channel users according to the conversion rate comparison result to obtain user risk assessment results;
judging whether the channel user is a risk user or not according to the user risk assessment result;
if yes, data clearing processing is carried out on the risk users so that the channel operation data can be more fit with the user conversion rate under the natural flow.
By adopting the technical scheme, because the false flow rate of the channel often exists on the macroscopic level of the channel, such as the page stay time, the access depth and the like, the risk channel with the false flow rate is usually represented as good in macroscopic data performance of the user, and the conversion rate and the retention rate are lower, so that the comparison between the conversion rate of the channel user and the preset expected conversion index is facilitated, whether the conversion rate of the channel user meets the expected conversion index is facilitated, whether the release effect of the channel meets the release requirement of an enterprise is further judged, whether the channel user is a risk user is judged according to the risk evaluation result of the user, such as the false user with good macroscopic data but low conversion rate, the data clearing processing is carried out on the risk user, the error influence caused by the invalid user on the analysis result of the channel operation data is reduced, and the data analysis result of the channel operation data is more attached to the user conversion rate under the natural flow rate.
The second object of the present invention is achieved by the following technical solutions:
there is provided an intelligent channel operation data analysis system including:
the data acquisition module is used for acquiring channel operation data in a full life cycle in real time, and carrying out data cleaning processing on the channel operation data to obtain effective user data;
the user portrait module is used for carrying out user portrait processing on the effective user data to obtain user portrait data;
the retention curve construction module is used for constructing a retention curve related to channel user retention according to the user portrait data to obtain a user retention curve graph;
the data marking module is used for identifying channel abnormal data on the user retention curve and marking the channel abnormal data to obtain channel marking data;
and the strategy optimization module is used for calculating the conversion rate of the channel user according to the user retention curve, and generating a strategy optimization instruction for optimizing the channel delivery strategy according to a calculation result so as to optimize the channel marking data through the strategy optimization instruction.
By adopting the technical scheme, the channel traffic cheating is generally performed uniformly through the timing simulation starting behavior and the fixed browsing behavior, so that the random browsing behavior and the click conversion behavior of a real user are difficult to generate, and therefore, the data cleaning processing is performed on the channel operation data in the full life cycle, the invalid operation data with fixed user behavior and no real conversion rate is helped to be cleaned, so that effective user data is obtained, the user image is performed on the effective user data, the error influence of false channel traffic on the user image is helped to be reduced, the data marking is performed on channel abnormal data such as inflection points, breakpoints and the like through the user retention graph, the channel abnormal data is helped to be optimized in a targeted manner, the multi-dimensional optimization is performed on the channel release strategy according to the user retention condition and the channel user conversion rate, the optimized channel release strategy is helped to be more attached to the real channel traffic, and the accuracy of the analysis result of the channel operation data is improved.
The third object of the present application is achieved by the following technical solutions:
a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the above-described intelligent channel operation data analysis method when the computer program is executed.
The fourth object of the present application is achieved by the following technical solutions:
a computer readable storage medium storing a computer program which when executed by a processor implements the steps of the above-described intelligent channel operation data analysis method.
In summary, the present application includes at least one of the following beneficial technical effects:
1. the method has the advantages that the channel operation data in the whole life cycle are subjected to data cleaning processing, invalid operation data with fixed user behaviors and no real conversion rate are helped to be cleaned, so that effective user data are obtained, user image is carried out through the effective user data, error influence on user image caused by false channel flow is helped to be reduced, channel abnormal data such as inflection points and breakpoints are marked through a user retention graph, the channel abnormal data are helped to be optimized in a targeted mode, the channel user conversion rate is calculated, multidimensional optimization is helped to be carried out on channel delivery strategies according to the user retention condition and the channel user conversion rate, and therefore the optimized channel delivery strategies are enabled to be more fit with the real channel flow, and accuracy of analysis results of the channel operation data is improved;
2. The user characteristic indexes meeting the enterprise throwing targets are obtained by analyzing the original channel throwing strategies, the user data with optimal fitting degree can be screened out according to the user characteristic indexes, because the unique user identification codes have uniqueness, the false flow is difficult to simulate the unique user identification codes, the user characteristic indexes and the user behavior data are clustered and associated according to the unique user identification codes, and effective user data meeting the user characteristic indexes can be rapidly associated and aggregated according to a user clustering model, so that user portrait data are obtained, and the user portrait rate of channel operation data is improved;
3. the method comprises the steps of drawing user behavior data distributed by scattered points into user behavior paths, and carrying out path planning by combining with expected conversion targets, so that user conversion paths are obtained, wherein the expected conversion targets are obtained by analysis in channel release strategies, the channel release strategies are adjusted through the user conversion paths, so that channel release contents are more matched with user requirements, user experience is improved, and path deviation data are obtained through fitting of actual behavior paths and user conversion paths, so that timely optimization of channel release materials according to the path deviation data is facilitated, the channel release paths are more matched with the user requirements, and release time between the actual behavior paths of users reaching the expected conversion targets is shortened through dynamic adjustment of release channels, so that channel release cost is reduced, and conversion efficiency of channel release is improved.
Drawings
Fig. 1 is a flowchart of an implementation of an intelligent channel operation data analysis method in an embodiment of the present application.
Fig. 2 is a flowchart of the implementation of step S20 of the intelligent channel operation data analysis method in an embodiment of the present application.
Fig. 3 is a flowchart of another implementation of step S20 of the intelligent channel operation data analysis method in an embodiment of the present application.
Fig. 4 is a flowchart of implementation of step S30 of the intelligent channel operation data analysis method in an embodiment of the present application.
Fig. 5 is a flowchart of implementation of step S40 of the intelligent channel operation data analysis method in an embodiment of the present application.
Fig. 6 is a flowchart of implementation of step S50 of the intelligent channel operation data analysis method in an embodiment of the present application.
Fig. 7 is a flowchart of another implementation of step S50 of the intelligent channel operation data analysis method in an embodiment of the present application.
Fig. 8 is a schematic structural diagram of an intelligent channel operation data analysis system according to an embodiment of the present application.
Fig. 9 is a schematic diagram of an internal structure of a computer device for implementing the intelligent channel operation data analysis method according to an embodiment of the present application.
Detailed Description
The present application is described in further detail below with reference to the accompanying drawings.
In an embodiment, as shown in fig. 1, the application discloses an intelligent channel operation data analysis method, which specifically includes the following steps:
s10: channel operation data of the whole life cycle is obtained in real time, and data cleaning processing is carried out on the channel operation data to obtain effective user data.
Specifically, the developer pulls channel operation data according to preset channel data packets, for example, when a user clicks corresponding product connection, short links between the user and the product are established, so that the channel operation data are pulled in real time according to the short links between each channel and the user, the channel operation data comprise user IP addresses, user personal information, user retention conditions, user conversion conditions and the like, the channel operation data in the whole life cycle comprise all user behavior data during the period that the user registers to the user logout, and part of suspicious or unnecessary traffic is filtered through preset anti-cheating rules, such as suspicious traffic with frequently replaced IP addresses, invalid traffic with liveness lower than preset standards, invalid traffic with incomplete data packets and the like, so that effective user data after the invalid traffic is filtered are obtained through data cleaning.
S20: and carrying out user portrait processing on the effective user data to obtain user portrait data.
Specifically, in order to make the user portrait data more fit with the delivery target of the enterprise, the original channel delivery strategy is analyzed, and the user characteristic index and the user behavior data are associated by combining with the user unique identification code, so as to obtain the user portrait data, as shown in fig. 2, step S20 specifically includes:
s101: and acquiring an original channel release strategy which accords with the enterprise channel release target.
Specifically, the channel delivery strategy is formulated according to the product labels and the user group positioning, and the delivery behaviors of the products are analyzed to obtain the original channel delivery strategy, for example, the original channel delivery strategy is obtained by comprehensively analyzing according to the product popularization mode selection of the users, the target user group selection, the popularization effect selection, the payment mode selection and the like, and accordingly, the selection of multiple delivery modes such as search engine delivery, social media delivery, information flow advertisement delivery and the like is obtained.
S102: and carrying out data analysis processing on the original channel release strategy to obtain a user characteristic index for drawing the user portrait.
Specifically, the data analysis is performed on the original channel release strategy, including analysis on channel user characteristics, channel release scenes and channel information, so as to obtain user characteristic indexes such as target crowd ages, hobbies, channel active time, purchasing power and the like.
S103: the unique user identification code and the user behavior data in the channel operation data are identified.
Specifically, according to channel operation data, a user mobile phone number or real-name authentication information when a user registers is obtained as a mobile phone unique identification code, and a pre-packaged monitoring data packet detects specific user behaviors in the channel operation data, wherein the specific user behaviors comprise user behaviors such as clicking, browsing and purchasing, and when the user behaviors reach preset triggering conditions, the user behavior data is automatically obtained through the monitoring data packet.
S104: and clustering the user characteristic indexes and the user behavior data according to the unique user identification codes to obtain a user clustering model.
Specifically, users are clustered according to each user characteristic index, for example, female users in 18-28 ages are classified into a class, users with channel active time exceeding a preset value such as 5 hours are classified into a class, a corresponding number of cluster centroids are set according to the number of the user characteristic indexes, repeated classification processing is carried out on all users in the same channel, and the user characteristic indexes and the user behavior data are associated according to classification results, so that a user clustering model is obtained.
S105: and inputting the effective user data into a user clustering model to obtain user portrait data.
Specifically, when effective user data is input into the user clustering model, the effective user data is classified according to user characteristic indexes, and user behavior data in the classified effective user data and the user characteristic indexes are associated, so that user portrait data about a unique user identification code is obtained, for example, the user portrait data corresponding to the associated unique user identification code is female users with ages 18-28, channel liveness is higher than average level, and behavior data such as browsing time, purchasing power and the like are higher than those of male users.
In this embodiment, in order to more accurately describe the user behavior, after the valid user data is input into the user cluster model to obtain the user portrait data, as shown in fig. 3, step S20 further includes:
s201: and carrying out path tracking processing on the user behavior data to obtain a user behavior path.
Specifically, user behavior data of each effective user behavior is obtained through a pre-packaged data monitoring package, such as user behavior data of actual clicking, browsing or purchasing behaviors, and the user behavior data distributed in scattered points are connected according to behavior generation time, so that trend of the user behavior data is obtained, and a user behavior path is obtained according to the trend of the user behavior data.
S202: and planning a path for the channel user to reach a preset conversion target according to the user behavior path to obtain a user conversion path.
Specifically, the user behavior paths are screened through a decision tree algorithm, an optimal path for reaching a preset conversion target at the current user position is obtained, for example, the next behavior of the user is predicted according to the user behavior paths, random sampling is carried out among a plurality of user expected behaviors according to the decision tree algorithm, and data training is carried out on the sampled samples so as to plan a conversion path which reaches the preset conversion target at the highest speed.
S203: and acquiring the actual behavior path of the channel user in real time in the channel operation process according to the user conversion path.
Specifically, accurate pushing is performed on the user according to the user conversion path, if a preset conversion target of the user conversion path is commodity achievement, when the user enters a pushing channel, the attention of the user is attracted through a popup window or a pushing material optimizing mode, order achievement is facilitated, actual behavior data of the user are obtained in real time in the process of pushing products through the user conversion path, and an actual behavior path is drawn.
S204: and carrying out path fitting processing on the actual behavior path and the user conversion path to obtain path deviation data.
Specifically, performing path fitting on the actual behavior path and the user conversion path, judging that the actual behavior path of the user accords with the user conversion path in an error range according to a fitting result, and if so, indicating that the user conversion path is an optimal user conversion path; if not, further optimizing the user conversion path according to the path deviation data between the user actual behavior path and the user conversion path, such as adjusting pushing material content, adjusting pushing mode, such as changing popup pushing into video page pushing, and the like.
S205: generating a path optimization instruction which corresponds to the path deviation data and is used for adjusting the delivery path of the original channel delivery strategy according to the path deviation data.
Specifically, according to path deviation data, such as an order forming stage on a user conversion path, a user only stays in a product browsing stage, so that the path deviation data between an actual behavior path and the user conversion path is caused, and according to the path deviation data and user portrait data, the reason of the path deviation data, such as low payment capability, is judged, so that a path optimization instruction corresponding to the path deviation data is generated, and the user adjusts the delivery path, such as adjusting a high-price product to a time period with higher purchasing power of the user for delivery, and adjusting a low-price product to a time period with lower purchasing power of the user for delivery.
S30: and constructing a retention curve about channel user retention according to the user portrait data to obtain a user retention curve.
Specifically, whether the user is a target crowd conforming to the channel promotion object is determined according to the user portrait data, and a user retention curve is constructed according to the user portrait data, such as user behavior, as shown in fig. 4, step S30 specifically includes:
s301: and identifying the user newly-added data and the user login data in the channel operation data, wherein the user login data is the next-day login data of the newly-added user.
Specifically, a pre-packaged data monitoring package is triggered according to a registration button when a user registers, new data of the user is obtained according to the activation condition of the data monitoring package, a short link between a user terminal and an operation background is built according to a unique user identification code, and the data monitoring package is triggered to record user login data through login operation of the user, so that the user login data is obtained.
S302: and calculating the retention rate according to the newly added data of the user and the login data of the user to obtain the retention rate of the user.
Specifically, according to a preset retention formula, if the quotient between the user login data and the newly added user data is used as the user retention, if the user login value is 200 times and the corresponding newly added user value is 1000 times, the user retention is 20%.
S303: and drawing a user retention curve corresponding to the user portrait data according to the user retention rate to obtain a user retention curve.
Specifically, according to the user retention rate in the adjacent operation time period, the adjacent user behavior data are associated, so that a user retention curve is drawn, and a user retention curve chart is obtained.
S40: identifying channel abnormal data on the user retention curve, and carrying out data marking on the channel abnormal data to obtain channel marking data.
Specifically, in order to reduce the error influence of the channel abnormal data on the analysis of the channel operation data, the channel marking data is obtained by identifying and marking the abnormal user data such as the data inflection point, the breakpoint and the like on the user retention curve, as shown in fig. 5, the step S40 specifically includes:
s401: and acquiring natural flow data in the channel operation data, wherein the natural flow data comprises natural user login data and natural user newly-added data.
Specifically, the natural flow data refers to product browsing or payment data which is automatically generated by a user through keyword searching, user fission and the like before the product is promoted without payment, and the natural flow data is obtained by recording the user access condition when the product is not promoted in a channel, wherein the number of user login in the non-promoted state is natural user login data, the number of user login in the non-promoted state is natural user newly added data,
S402: and constructing a standard retention curve related to the natural flow data according to the natural user login data and the natural user newly-added data.
Specifically, the retention rate of the natural flow is calculated according to the natural user login data and the natural user newly-added data of the adjacent operation period, if the obtained number of the natural user newly-added people on the day is 1000, the natural user login number on the next day is 500, the retention rate of the natural flow is 50%, and the user data under the adjacent operation period are connected according to the retention rate of the natural flow data, so that a standard retention curve is drawn, and an effective data reference is conveniently provided for the user retention condition under the channel popularization state.
S403: and comparing the characteristics of the standard retention curve and the user retention curve to obtain channel abnormal data.
Specifically, with an operation period as a reference, comparing the characteristics of a user retention curve and a standard retention curve in the same operation period, and dividing the operation period into an activation stage, a retention stage and a loss stage, wherein if the average user browsing duration of the standard retention curve is 3 hours in the retention stage and the average user browsing duration in the user retention curve is 1 hour, the characteristic comparison of the average browsing duration of the two curves is greatly different, the average browsing duration abnormal data is 2 hours, and the judgment of the reason of the user retention characteristic abnormality according to the channel abnormal data is facilitated.
S404: and carrying out data marking on the channel abnormal data according to the user characteristic indexes in the user portrait data to obtain channel marking data.
Specifically, the channel abnormal data are marked according to user characteristic indexes in the user portrait data, such as user purchasing power, user activity, user preference and the like, for example, channel abnormal data with average user browsing time length lower than a standard retention curve are marked as low user activity, and channel abnormal data on the user retention curve are marked through each user characteristic index respectively, so that channel marked data which accords with the user portrait data are obtained.
S50: and calculating the conversion rate of the channel user according to the user retention curve, and generating a strategy optimization instruction for optimizing the channel delivery strategy according to the calculation result so as to optimize the channel marking data through the strategy optimization instruction.
Specifically, the channel delivery strategy is optimized in multiple dimensions by combining the user retention curve with the user conversion rate, so that the channel delivery strategy is more fit with the user demand of the real flow of the channel, as shown in fig. 6, step S50 specifically includes:
s501: and acquiring conversion frequency information of channel users corresponding to the effective user data in real time.
Specifically, real-time data monitoring is performed on the effective user data, conversion frequency information of channel users corresponding to the effective user data is counted, for example, according to an original channel delivery strategy, a conversion target is analyzed, for example, the channel delivery target of an e-commerce enterprise generates purchasing behavior for the users, for example, the channel delivery target of a news media enterprise generates page browsing behavior for the users, real purchasing frequency of the effective channel users is counted, or page browsing frequency of the effective channel users is counted, for example, browsing duration is more than 30 seconds, and effective browsing is performed, so that conversion frequency information of the channel users is obtained, for example, the frequency of generating purchasing behavior in 500 effective channel users is 300 times, and the conversion frequency information is 300 times.
S502: and calculating the conversion frequency information and user login data conforming to a user retention curve to obtain the conversion rate of the channel user.
Specifically, according to a preset channel conversion rate calculation formula, if the quotient of the conversion times information and the user login data is taken as the channel user conversion rate, if the user login data is that the number of people logged in is 500, the conversion times which meet the channel release target are 200 times, and the channel user conversion rate is 40%.
S503: and constructing a user funnel model conforming to a user retention curve according to the channel user conversion rate.
Specifically, according to the conversion rate of channel users, planning channel operation links from registration to channel delivery targets, for example, planning channel operation links taking commodity purchase as delivery targets, for example, setting users who click into commodity homepage as target users, setting the user who clicks into commodity homepage as first links, acquiring page browsing data of the target users as second links, setting commodity adding shopping carts as third links, setting payment pages as fourth links, setting successful payment as fifth links, respectively acquiring user behavior data of each link, and associating the user behavior data of each link according to unique identification codes of the users, thereby obtaining a user funnel model.
S504: and inputting channel operation data into the user funnel model to generate a strategy optimization instruction which corresponds to the channel user conversion rate and is used for optimizing the channel delivery strategy.
Specifically, channel operation data are input into a user funnel model, and user conversion rate under each link is calculated according to a unique user identification code, so that according to the conversion condition of each link, targeted optimization is performed on problem links with user loss or low conversion rate, for example, for users with short page browsing duration, channel delivery strategies can be optimized by optimizing product detail pages or changing product introduction modes and the like; and for the user with low payment conversion rate, the purchasing power of the user is analyzed, commodities with corresponding price are pushed to the user, and the channel delivery strategy is optimized.
In this embodiment, in order to better perform comprehensive evaluation on the channel delivery strategy, so as to screen out an effective channel for performing accurate marketing, as shown in fig. 7, after step S50, the method further includes:
s601: and comparing the conversion rate of the channel user with an expected conversion index preset in the channel release strategy to obtain a conversion rate comparison result.
Specifically, if the expected conversion target is that the user pays 50% of conversion rate and the actual conversion rate of the channel user is 20%, the conversion rate comparison result is that the conversion rate difference between the conversion rate of the channel user and the expected conversion target is 30%.
S602: and carrying out risk assessment on the channel users according to the conversion rate comparison result to obtain user risk assessment results.
Specifically, the preset conversion rate difference threshold value is 10% in the channel operation strategy, namely, in the error range, the conversion rate difference between the actual channel user conversion rate and the expected conversion target is smaller than 10%, the conversion rate comparison result is larger than or equal to 10%, the existence of false flow in the corresponding channel is judged, the macroscopic base number of the user conversion rate is overlarge, the actual user conversion rate is far smaller than the expected conversion index, and the user risk assessment result is high risk; if the conversion rate comparison result is smaller than 10%, the possibility that the corresponding channel user is the real flow is judged to be high, and the user risk assessment result is low risk.
S603: and judging whether the channel user is a risk user or not according to the user risk assessment result.
Specifically, if the user risk assessment result is high risk, the phenomenon that the corresponding channel is fake is indicated, and the corresponding channel user is a risk user; if the risk assessment result of the user is low risk, the corresponding channel flow is indicated to be real flow, and the corresponding channel user is a non-risk user.
S604: if yes, data clearing processing is carried out on the risk users so that channel operation data can be more fit with the user conversion rate under the natural flow.
Specifically, if the channel user is a risk user, the corresponding channel operation data is not in accordance with the actual operation requirement, the effect of achieving the enterprise release target is not great, and the user data determined to be the risk user is subjected to data clearing, so that the error influence of false flow data on the data analysis accuracy is reduced.
S605: if not, continuing to monitor the data of the non-risk user.
Specifically, if the channel user is a non-risk user, the corresponding channel operation data is the user data under the real flow, and continuous data monitoring is performed on the non-risk user through the short link constructed during user registration.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic of each process, and should not limit the implementation process of the embodiment of the present application in any way.
In an embodiment, an intelligent channel operation data analysis system is provided, where the intelligent channel operation data analysis system corresponds to the intelligent channel operation data analysis method in the above embodiment one by one. As shown in FIG. 8, the intelligent channel operation data analysis system comprises a data acquisition module, a user portrayal module, a retention curve construction module, a data marking module and a strategy optimization module. The functional modules are described in detail as follows:
the data acquisition module is used for acquiring channel operation data in a full life cycle in real time, and performing data cleaning processing on the channel operation data to obtain effective user data.
And the user portrait module is used for carrying out user portrait processing on the effective user data to obtain user portrait data.
And the retention curve construction module is used for constructing a retention curve related to channel user retention according to the user portrait data to obtain a user retention curve graph.
And the data marking module is used for identifying channel abnormal data on the user retention curve and marking the channel abnormal data to obtain channel marking data.
The strategy optimization module is used for calculating the conversion rate of the channel user according to the user retention curve, and generating a strategy optimization instruction for optimizing the channel delivery strategy according to the calculation result so as to optimize the channel marking data through the strategy optimization instruction.
Optionally, the user portrait module includes:
and the strategy acquisition sub-module is used for acquiring an original channel release strategy which accords with the enterprise channel release target.
And the index acquisition sub-module is used for carrying out data analysis processing on the original channel delivery strategy to obtain a user characteristic index for drawing the user portrait.
And the data identification sub-module is used for identifying the unique user identification code and the user behavior data in the channel operation data.
And the clustering model construction submodule is used for carrying out clustering processing on the user characteristic indexes and the user behavior data according to the unique user identification codes to obtain a user clustering model.
And the user portrait data acquisition sub-module is used for inputting the effective user data into the user clustering model to obtain the user portrait data.
In this embodiment, in order to improve accuracy of user portrayal and shorten an operation period for a user to reach a preset conversion target, after valid user data is input into a user cluster model to obtain user portrayal data, the user portrayal module further includes:
And the path tracking sub-module is used for carrying out path tracking processing on the user behavior data to obtain a user behavior path.
And the path planning sub-module is used for planning the path of the channel user reaching the preset conversion target according to the user behavior path to obtain the user conversion path.
The actual behavior path acquisition sub-module is used for acquiring the actual behavior path of the channel user in real time in the channel operation process according to the user conversion path.
And the path fitting sub-module is used for carrying out path fitting processing on the actual behavior path and the user conversion path to obtain path deviation data.
The path optimization instruction generation sub-module is used for generating a path optimization instruction which corresponds to the path deviation data and is used for carrying out delivery path adjustment on the original channel delivery strategy according to the path deviation data.
Optionally, the retention curve construction module includes:
the data identification sub-module is used for identifying the user newly-added data and the user login data in the channel operation data, wherein the user login data is the next-day login data of the newly-added user.
And the retention rate calculation sub-module is used for calculating the retention rate according to the newly added data of the user and the login data of the user to obtain the retention rate of the user.
And the retention curve drawing sub-module is used for drawing a user retention curve corresponding to the user portrait data according to the user retention rate to obtain a user retention curve.
Optionally, the data marking module includes:
the natural flow obtaining sub-module is used for obtaining natural flow data in channel operation data, wherein the natural flow data comprises natural user login data and natural user newly-added data.
And the standard retention curve drawing submodule is used for constructing a standard retention curve related to the natural flow data according to the natural user login data and the natural user newly-added data.
And the characteristic comparison sub-module is used for comparing the characteristics of the standard retention curve and the user retention curve to obtain channel abnormal data.
And the data marking sub-module is used for marking the channel abnormal data according to the user characteristic indexes in the user portrait data to obtain channel marking data.
Optionally, the policy optimization module includes:
the conversion times acquisition sub-module is used for acquiring conversion times information of channel users corresponding to the effective user data in real time.
And the conversion rate calculation sub-module is used for calculating the conversion times information and the user login data which accords with the user retention curve to obtain the channel user conversion rate.
And the funnel model construction submodule is used for constructing a user funnel model which accords with the user retention curve according to the channel user conversion rate.
The strategy optimization instruction generation sub-module is used for inputting channel operation data into the user funnel model, and generating strategy optimization instructions which correspond to the channel user conversion rate and are used for optimizing the channel delivery strategy.
In this embodiment, in order to comprehensively manage channel operation data, and in order to screen out an effective channel to perform accurate delivery, the policy optimization module further includes:
and the conversion index comparison sub-module is used for comparing the conversion rate of the channel user with an expected conversion index preset in the channel release strategy to obtain a conversion rate comparison result.
And the risk assessment processing sub-module is used for carrying out risk assessment on the channel users according to the conversion rate comparison result to obtain user risk assessment results.
And the risk user judging sub-module is used for judging whether the channel user is a risk user or not according to the user risk assessment result.
And the data clearing processing sub-module is used for carrying out data clearing processing on the risk users if the risk users are in the natural flow, so that channel operation data can be more attached to the user conversion rate under the natural flow.
The specific limitation of the intelligent channel operation data analysis system can be referred to the limitation of the intelligent channel operation data analysis method hereinabove, and will not be described herein. The modules in the intelligent channel operation data analysis system can be all or partially realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 9. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing channel operation data and analysis data generated in the process of carrying out data analysis on the channel operation data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a method of intelligent channel operation data analysis.
In one embodiment, a computer readable storage medium is provided, on which a computer program is stored which, when executed by a processor, implements the steps of the intelligent channel operation data analysis method described above.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the system is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.
Claims (9)
1. The intelligent channel operation data analysis method is characterized by comprising the following steps of:
channel operation data of a full life cycle is obtained in real time, and data cleaning processing is carried out on the channel operation data to obtain effective user data;
Performing user portrait processing on the effective user data to obtain user portrait data;
constructing a retention curve about channel user retention according to the user portrait data to obtain a user retention curve;
identifying channel abnormal data on the user retention curve, and carrying out data marking on the channel abnormal data to obtain channel marking data;
calculating the conversion rate of the channel user according to the user retention curve, and generating a strategy optimization instruction for optimizing the channel release strategy according to a calculation result so as to optimize the channel marking data through the strategy optimization instruction;
the user portrait processing is carried out on the effective user data to obtain user portrait data, which comprises the following steps:
acquiring an original channel release strategy which accords with an enterprise channel release target;
performing data analysis processing on the original channel release strategy to obtain a user characteristic index for drawing a user portrait; the data analysis processing is carried out on the original channel release strategy to obtain a user characteristic index for drawing the user portrait, which comprises the following steps: analyzing channel user characteristics, channel putting scenes and channel information to obtain user characteristic indexes;
Identifying a user unique identification code and user behavior data in the channel operation data;
setting a corresponding number of clustering centroids according to the number of the user characteristic indexes, repeatedly classifying all users in the same channel according to the unique user identification codes, and associating the user characteristic indexes with the user behavior data according to classification results to obtain a user clustering model;
and inputting the effective user data into the user clustering model, classifying the effective user data according to the user characteristic index, and correlating the user behavior data in the classified effective user data with the user characteristic index to obtain the user portrait data about the user unique identification code.
2. The intelligent channel operation data analysis method according to claim 1, wherein after the effective user data is input into the user cluster model to obtain user portrayal data, comprising:
performing path tracking processing on the user behavior data to obtain a user behavior path; the path tracking processing of the user behavior data comprises the following steps: connecting the user behavior data distributed by the scattered points according to behavior generation time to obtain the trend of the user behavior data, and obtaining the user behavior path according to the trend of the user behavior data;
According to the user behavior path, path planning is carried out on channel users reaching a preset conversion target, and a user conversion path is obtained; and planning a path for the channel user to reach a preset conversion target according to the user behavior path to obtain a user conversion path, wherein the method comprises the following steps: predicting the user expected behavior of the user according to the user behavior path; randomly sampling in a plurality of expected behaviors of the user according to a decision tree algorithm, and performing data training on the sampled samples so as to plan the user conversion path which reaches the preset conversion target at the fastest speed;
acquiring an actual behavior path of a channel user in real time in the channel operation process according to the user conversion path;
performing path fitting processing on the actual behavior path and the user conversion path to obtain path deviation data;
generating a path optimization instruction which corresponds to the path deviation data and is used for carrying out delivery path adjustment on the original channel delivery strategy according to the path deviation data.
3. The method for analyzing intelligent channel operation data according to claim 1, wherein the constructing a retention curve for channel user retention according to the user portrait data to obtain a user retention curve specifically comprises:
Identifying user newly-added data and user login data in the channel operation data, wherein the user login data is next-day login data of the newly-added user;
calculating the retention rate according to the newly added data of the user and the login data of the user to obtain the retention rate of the user;
and drawing a user retention curve corresponding to the user portrait data according to the user retention rate to obtain a user retention curve.
4. The method for analyzing intelligent channel operation data according to claim 1, wherein the identifying the channel anomaly data on the user retention curve and performing data marking on the channel anomaly data to obtain channel marking data specifically comprises:
acquiring natural flow data in the channel operation data, wherein the natural flow data comprises natural user login data and natural user newly-added data;
constructing a standard retention curve related to natural flow data according to the natural user login data and the natural user newly-added data;
comparing the characteristics of the standard retention curve and the user retention curve to obtain channel abnormal data;
and carrying out data marking on the channel abnormal data according to the user characteristic indexes in the user portrait data to obtain channel marking data.
5. The intelligent channel operation data analysis method according to claim 1, wherein the calculating the conversion rate of the channel user according to the user retention curve, and generating a policy optimization instruction for optimizing the channel delivery policy according to the calculation result, so as to optimize the channel marking data through the policy optimization instruction, specifically comprises:
acquiring conversion frequency information of channel users corresponding to the effective user data in real time;
calculating the conversion frequency information and user login data conforming to the user retention curve to obtain channel user conversion rate;
constructing a user funnel model conforming to the user retention curve according to the channel user conversion rate;
and inputting the channel operation data into the user funnel model, and generating a strategy optimization instruction which corresponds to the channel user conversion rate and is used for optimizing the channel delivery strategy.
6. The method for analyzing intelligent channel operation data according to claim 5, wherein the calculating the conversion rate of the channel user according to the user retention curve, and generating a policy optimization instruction for optimizing the channel delivery policy according to the calculation result, so that after optimizing the channel marking data by the policy optimization instruction, the method further comprises:
Comparing the conversion rate of the channel user with an expected conversion index preset in the channel release strategy to obtain a conversion rate comparison result;
performing risk assessment on the channel users according to the conversion rate comparison result to obtain user risk assessment results;
judging whether the channel user is a risk user or not according to the user risk assessment result;
if yes, data clearing processing is carried out on the risk users so that the channel operation data can be more fit with the user conversion rate under the natural flow.
7. An intelligent channel operation data analysis system, characterized in that the intelligent channel operation data analysis system comprises:
the data acquisition module is used for acquiring channel operation data in a full life cycle in real time, and carrying out data cleaning processing on the channel operation data to obtain effective user data;
the user portrait module is used for carrying out user portrait processing on the effective user data to obtain user portrait data;
the retention curve construction module is used for constructing a retention curve related to channel user retention according to the user portrait data to obtain a user retention curve graph;
the data marking module is used for identifying channel abnormal data on the user retention curve and marking the channel abnormal data to obtain channel marking data;
The strategy optimization module is used for calculating the conversion rate of the channel user according to the user retention curve, and generating a strategy optimization instruction for optimizing the channel release strategy according to a calculation result so as to optimize the channel marking data through the strategy optimization instruction;
the user portrait module comprises:
the strategy acquisition sub-module is used for acquiring an original channel release strategy which accords with the enterprise channel release target;
the index acquisition sub-module is used for carrying out data analysis processing on the original channel delivery strategy to obtain a user characteristic index for drawing a user portrait; the data analysis processing is carried out on the original channel release strategy to obtain a user characteristic index for drawing the user portrait, which comprises the following steps: analyzing channel user characteristics, channel putting scenes and channel information to obtain user characteristic indexes;
the data identification sub-module is used for identifying the unique user identification code and the user behavior data in the channel operation data;
the clustering model construction submodule is used for setting a corresponding number of clustering centroids according to the number of the user characteristic indexes, carrying out repeated classification processing on all users in the same channel according to the unique user identification codes, and associating the user characteristic indexes with the user behavior data according to classification results to obtain a user clustering model;
And the user portrait data acquisition sub-module is used for inputting the effective user data into the user clustering model, classifying the effective user data according to the user characteristic index, and correlating the user behavior data in the classified effective user data with the user characteristic index so as to obtain the user portrait data about the user unique identification code.
8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the intelligent channel operation data analysis method according to any one of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the intelligent channel operation data analysis method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211025826.4A CN115408586B (en) | 2022-08-25 | 2022-08-25 | Intelligent channel operation data analysis method, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211025826.4A CN115408586B (en) | 2022-08-25 | 2022-08-25 | Intelligent channel operation data analysis method, system, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115408586A CN115408586A (en) | 2022-11-29 |
| CN115408586B true CN115408586B (en) | 2024-01-23 |
Family
ID=84162177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211025826.4A Active CN115408586B (en) | 2022-08-25 | 2022-08-25 | Intelligent channel operation data analysis method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115408586B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117217828B (en) * | 2023-08-18 | 2024-05-10 | 上海数禾信息科技有限公司 | Method, device, computer equipment and storage medium for verifying conversion return data |
| CN117556192A (en) * | 2023-11-16 | 2024-02-13 | 南京小裂变网络科技有限公司 | User growth type operation system and method based on artificial intelligence |
| CN117593055A (en) * | 2023-11-21 | 2024-02-23 | 广州吴凡科技服务有限公司 | Advertisement effect intelligent analysis method and system based on big data |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106612202A (en) * | 2015-10-27 | 2017-05-03 | 网易（杭州）网络有限公司 | Method and system for pre-estimate and judgment of amount brushing of online game channel |
| JP2018037009A (en) * | 2016-09-02 | 2018-03-08 | 株式会社日立製作所 | Cluster analysis system, cluster analysis device, and cluster analysis method |
| CN107944059A (en) * | 2017-12-29 | 2018-04-20 | 深圳市中润四方信息技术有限公司西安分公司 | A kind of user behavior analysis method and system based on stream calculation |
| CN109086787A (en) * | 2018-06-06 | 2018-12-25 | 平安科技（深圳）有限公司 | User's portrait acquisition methods, device, computer equipment and storage medium |
| CN109284380A (en) * | 2018-09-25 | 2019-01-29 | 平安科技（深圳）有限公司 | Illegal user's recognition methods and device, electronic equipment based on big data analysis |
| CN109815269A (en) * | 2018-12-24 | 2019-05-28 | 苏州蜗牛数字科技股份有限公司 | A kind of data increment information calculation method and system |
| CN111090822A (en) * | 2019-12-18 | 2020-05-01 | 浙江口碑网络技术有限公司 | Business object pushing method and device |
| CN111476375A (en) * | 2020-03-30 | 2020-07-31 | 中移动信息技术有限公司 | Method and device for determining recognition model, electronic equipment and storage medium |
| CN112633715A (en) * | 2020-12-28 | 2021-04-09 | 四川新网银行股份有限公司 | Method for analyzing loss of online service user |
| CN113743995A (en) * | 2021-09-09 | 2021-12-03 | 平安科技（深圳）有限公司 | Multi-channel advertisement delivery method, device, equipment and storage medium |
| CN113781095A (en) * | 2021-07-29 | 2021-12-10 | 深圳思为科技有限公司 | Method for optimizing product delivery channel and related device |
| CN113836124A (en) * | 2021-07-23 | 2021-12-24 | 青岛海尔科技有限公司 | User portrait generation method, device, equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220101044A1 (en) * | 2020-09-29 | 2022-03-31 | International Business Machines Corporation | Artificial intelligence model generation using data with desired diagnostic content |
-
2022
- 2022-08-25 CN CN202211025826.4A patent/CN115408586B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106612202A (en) * | 2015-10-27 | 2017-05-03 | 网易（杭州）网络有限公司 | Method and system for pre-estimate and judgment of amount brushing of online game channel |
| JP2018037009A (en) * | 2016-09-02 | 2018-03-08 | 株式会社日立製作所 | Cluster analysis system, cluster analysis device, and cluster analysis method |
| CN107944059A (en) * | 2017-12-29 | 2018-04-20 | 深圳市中润四方信息技术有限公司西安分公司 | A kind of user behavior analysis method and system based on stream calculation |
| CN109086787A (en) * | 2018-06-06 | 2018-12-25 | 平安科技（深圳）有限公司 | User's portrait acquisition methods, device, computer equipment and storage medium |
| CN109284380A (en) * | 2018-09-25 | 2019-01-29 | 平安科技（深圳）有限公司 | Illegal user's recognition methods and device, electronic equipment based on big data analysis |
| CN109815269A (en) * | 2018-12-24 | 2019-05-28 | 苏州蜗牛数字科技股份有限公司 | A kind of data increment information calculation method and system |
| CN111090822A (en) * | 2019-12-18 | 2020-05-01 | 浙江口碑网络技术有限公司 | Business object pushing method and device |
| CN111476375A (en) * | 2020-03-30 | 2020-07-31 | 中移动信息技术有限公司 | Method and device for determining recognition model, electronic equipment and storage medium |
| CN112633715A (en) * | 2020-12-28 | 2021-04-09 | 四川新网银行股份有限公司 | Method for analyzing loss of online service user |
| CN113836124A (en) * | 2021-07-23 | 2021-12-24 | 青岛海尔科技有限公司 | User portrait generation method, device, equipment and storage medium |
| CN113781095A (en) * | 2021-07-29 | 2021-12-10 | 深圳思为科技有限公司 | Method for optimizing product delivery channel and related device |
| CN113743995A (en) * | 2021-09-09 | 2021-12-03 | 平安科技（深圳）有限公司 | Multi-channel advertisement delivery method, device, equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Autoencoder-based feature construction for IoT attacks clustering;Junaid Haseeb 等;《Future Generation Computer Systems》;487-502 * |
| 按风格划分数据的模糊聚类算法;沈浩 等;《模式识别与人工智能》;204-213 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115408586A (en) | 2022-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115408586B (en) | Intelligent channel operation data analysis method, system, equipment and storage medium | |
| CN107665444B (en) | Network advertisement instant effect evaluation method and system based on user online behavior | |
| CN106022834B (en) | Advertisement anti-cheating method and device | |
| US20190122258A1 (en) | Detection system for identifying abuse and fraud using artificial intelligence across a peer-to-peer distributed content or payment networks | |
| CN104050178B (en) | A kind of anti-cheat method of Internet surveillance and device | |
| US20170186030A1 (en) | Advertisement click-through rate correction method and advertisement push server | |
| US8255273B2 (en) | Evaluating online marketing efficiency | |
| US20100161385A1 (en) | Method and System for Content Based Demographics Prediction for Websites | |
| US10037546B1 (en) | Honeypot web page metrics | |
| US20160285672A1 (en) | Method and system for processing network media information | |
| AU2006332534A1 (en) | Predicting ad quality | |
| CN103136695A (en) | Analytical method and system of media placement | |
| CN105183873A (en) | Malicious clicking behavior detection method and device | |
| Galán et al. | Large-scale analysis of user exposure to online advertising on Facebook | |
| CN105260913A (en) | CTR estimation method and system, and DSP server used for Internet advertisement putting | |
| CN111859234A (en) | Illegal content identification method and device, electronic equipment and storage medium | |
| CN113869931A (en) | Advertisement putting strategy determining method and device, computer equipment and storage medium | |
| KR20070094264A (en) | Method for targeting web advertisement clickers based on click pattern by using a collaborative filtering system with neural networks and system thereof | |
| US9306958B2 (en) | Methods, systems and media for detecting non-intended traffic using co-visitation information | |
| Miranda et al. | On the flow of software security advisories | |
| CN114065090A (en) | Method and system for updating classification database, storage medium and computer equipment | |
| EP2469462A2 (en) | A method and a system for analysing traffic on a website | |
| WO2023205713A2 (en) | Systems and methods for improved user experience participant selection | |
| KR20120069938A (en) | Online marketing analysis method and online marketing analysis system | |
| TWI810339B (en) | Keyword Ad Malicious Click Analysis System |
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 |