CN116955738A - User behavior prediction system based on network footprint analysis - Google Patents
User behavior prediction system based on network footprint analysis Download PDFInfo
- Publication number
- CN116955738A CN116955738A CN202311206662.XA CN202311206662A CN116955738A CN 116955738 A CN116955738 A CN 116955738A CN 202311206662 A CN202311206662 A CN 202311206662A CN 116955738 A CN116955738 A CN 116955738A
- Authority
- CN
- China
- Prior art keywords
- footprint
- data
- unit
- user
- invalid
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 45
- 238000012216 screening Methods 0.000 claims description 15
- 238000012098 association analyses Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 claims description 4
- 230000006399 behavior Effects 0.000 description 96
- 238000000034 method Methods 0.000 description 12
- 230000006872 improvement Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 4
- 230000003542 behavioural effect Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002360 preparation method Methods 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
- G06F16/903—Querying
- G06F16/9035—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- 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
- G06F16/951—Indexing; Web crawling techniques
-
- 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
Abstract
The invention relates to the technical field of behavior analysis. The invention relates to a user behavior prediction system based on network footprint analysis. The system comprises a footprint grabbing unit, an invalid time setting unit, an invalid data eliminating unit, a behavior data searching unit and a predicted data updating unit; the footprint grabbing unit is used for grabbing footprint data of a user on the network; the invalid time setting unit is used for carrying out frequency analysis on the footprint data grabbed by the footprint grabbing unit and setting the invalid time of the footprint data according to an analysis result; the invalid time is set through the invalid time setting unit and the invalid data eliminating unit, part of the footprint data is compared through the invalid time, the invalid footprint data is eliminated, the situation that the user favors are changed, the part of history record data of the user is not suitable for the user, invalidity is caused, and misleading of the predicted data is caused by the part of history record data when the user is predicted.
Description
Technical Field
The invention relates to the technical field of behavior analysis, in particular to a user behavior prediction system for analyzing based on network footprints.
Background
The analysis of user network behavior refers to that under the condition that relevant data of user operation behaviors are obtained, the relevant data are subjected to statistical analysis, so that group constitution and respective preference of network users are judged and found, at present, when the user behaviors are predicted, the user is subjected to behavior analysis through historical record data of the user, but the preference of people is changed along with things, when the preference is changed, part of the historical record data of the user is not suitable for the user, inefficiency is caused, and when the user is predicted, part of the historical record data misguide the predicted data, so that deviation of the predicted data is caused, and in order to reduce the situation, a user behavior prediction system based on network footprint analysis is proposed.
Disclosure of Invention
The invention aims to provide a user behavior prediction system based on network footprint analysis, so as to solve the problems in the background technology.
In order to achieve the above object, a user behavior prediction system for performing analysis based on a network footprint is provided, which comprises a footprint grabbing unit, an invalid time setting unit, an invalid data eliminating unit, a behavior data searching unit and a predicted data updating unit;
the footprint grabbing unit is used for grabbing footprint data of a user on a network;
the invalid time setting unit is used for performing frequent analysis on the footprint data grabbed by the footprint grabbing unit and setting the invalid time of the footprint data according to an analysis result;
the invalid data eliminating unit is used for dividing the footprint data grabbed by the footprint grabbing unit according to the time periods, comparing the footprint data of each time period with the invalid time set by the invalid time setting unit, and eliminating the footprint data exceeding the invalid time;
the behavior data searching unit is used for screening according to the footprint data removed by the invalid data removing unit, obtaining the characteristic data with the longest service time in each period, and searching the behavior data with the type association in the footprint data according to the characteristic data;
the prediction data updating unit is used for establishing a user behavior prediction model, taking the characteristic data of the time period as behavior prediction data of the user, combining the footprint grabbing unit to grab the latest footprint data for feedback, and updating the behavior prediction data according to the feedback result.
As a further improvement of the technical scheme, the footprint grabbing unit grabs the behavior trace data of the user on the network through a network footprint grabbing tool.
As a further improvement of the technical scheme, the footprint grabbing unit further comprises a user information collecting module, wherein the user information collecting module is used for sending an interest behavior hobby filling questionnaire to a user and acquiring footprint data of the user in a network according to the content of the user feedback questionnaire.
As a further improvement of the technical scheme, the invalid time setting unit comprises an invalid setting module, wherein the invalid setting module is used for setting invalid time thresholds corresponding to different occupied times in one day, and meanwhile, frequent analysis is carried out according to the footprint data grabbed by the footprint grabbing unit, so that an average value of network footprint occupied times of a month user in each day is obtained, and the average value is compared with the invalid time thresholds, so that corresponding invalid time is set.
As a further improvement of the technical scheme, the invalid data eliminating unit comprises a time interval dividing module and a data screening module;
the time interval division module is used for dividing the footprint data captured by the footprint capturing unit according to the three time intervals of morning, evening and evening to obtain an early footprint database, a middle footprint database and a late footprint database;
the data screening module is used for comparing the early footprint database and the middle footprint database which are divided by the time interval dividing module with the invalid time set by the invalid setting module respectively, and eliminating the footprint data of which the record time exceeds the invalid time in each time interval.
As a further improvement of the technical scheme, the invalid data eliminating unit further comprises a user use analysis module, wherein the user use analysis module is used for carrying out frequent analysis according to the footprint data grabbed by the footprint grabbing unit, dividing the footprint data into the working time and the working time according to the analysis result, eliminating the footprint data used by the working time, and simultaneously sending the reserved data to the time interval dividing module for dividing.
As a further improvement of the technical scheme, the behavior data searching unit comprises a characteristic data acquisition module and an association matching module;
the characteristic data acquisition module is used for respectively carrying out data screening with the longest occupation time according to the early footprint database and the middle footprint database which are removed by the data screening module, and acquiring characteristic data with the longest occupation time corresponding to the late footprint database;
and the association matching module is used for performing behavior association analysis on the footprint data acquired by the footprint grabbing unit according to the characteristic data to acquire behavior data of the footprint data and the characteristic data.
As a further improvement of the technical scheme, the prediction data updating unit establishes a user behavior prediction model according to an early footprint database, a middle footprint database and a late footprint database through a neural network technology.
As a further improvement of the technical scheme, the prediction data updating unit comprises a behavior prediction module and a feedback updating module;
the behavior prediction module is used for sending the time period to the user behavior prediction model, and the user behavior prediction model is matched with the footprint database of the corresponding time according to the time period and sends the characteristic data of the footprint database as behavior prediction data;
the feedback updating module is used for carrying out difference comparison according to the latest footprint data captured by the footprint capturing unit and the behavior prediction data sent by the behavior prediction module, carrying out matching according to the latest footprint data in combination with the behavior data if the comparison result is wrong, sending the behavior prediction module to update by taking the behavior data as the behavior prediction data if the matching is correct, and continuing to predict if the comparison result is correct.
Compared with the prior art, the invention has the beneficial effects that:
in the user behavior prediction system based on network footprint analysis, the invalid time is set through the invalid time setting unit and the invalid data eliminating unit, part of data in footprint data is eliminated through the invalid time comparison, the invalid footprint data is avoided, the situation that part of history record data of a user is not suitable for the user due to the fact that the user prefers to change, invalidity occurs, the predicted data is misled by part of history record data when the user is predicted, the predicted data is offset, the user behavior is predicted according to time intervals through the behavior data searching unit and the predicted data updating unit, the predicted data is updated according to the user behavior, and the fact that the predicted data is inconsistent with the user behavior, and the quality of a product pushed by the user is reduced is avoided.
Drawings
Fig. 1 is an overall flow diagram of the present invention.
The meaning of each reference sign in the figure is:
10. footprint grabbing unit; 20. an invalid time setting unit; 30. an invalid data eliminating unit; 40. a behavior data searching unit; 50. and a prediction data updating unit.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1:
referring to fig. 1, the present embodiment is directed to a user behavior prediction system for performing analysis based on a network footprint, which includes a footprint grabbing unit 10, an invalid time setting unit 20, an invalid data eliminating unit 30, a behavior data searching unit 40, and a predicted data updating unit 50;
the footprint grabbing unit 10 is used for grabbing footprint data of a user on a network;
the footprint grabbing unit 10 grabs the behavior trace data of the user on the network through the network footprint grabbing tool. The method comprises the following steps:
defining a grabbing range: determining a network platform or website to be grabbed, such as a social media platform, a search engine, an online shopping platform and the like;
acquiring authorized access: if the behavior data of the user needs to be acquired, the authorized access rights of the user, such as a login account or an API key, need to be acquired;
setting grabbing parameters: determining the type and time range of the data to be grabbed, such as browsing records, searching records, purchasing records and the like;
writing a grabbing script: programming a program by using a programming language or a special tool, simulating a user to access a webpage or call an API (application program interface), and thus obtaining behavior data of the user;
data storage and processing: the captured data is stored in a database or file system for later analysis and processing.
The footprint grabbing unit 10 further comprises a user information collecting module, wherein the user information collecting module is used for sending an interest behavior hobby filling out questionnaire to the user, and acquiring footprint data of the user in the network according to the content of the user feedback questionnaire. The method comprises the following steps:
designing a questionnaire: information content to be collected is determined, including questions in terms of interests, behaviors, hobbies, and the like of the user. Ensuring that the questions are clear and providing appropriate options for the user to choose or freely fill in;
selecting a transmission mode: determining a method for sending a questionnaire to a user, which can be performed by means of e-mail, social media, website popup and the like;
sending a questionnaire: the questionnaire links or the questionnaire contents are sent to the user, so that a concise and clear description is provided, and the explanation is why the user is required to fill in the questionnaire and how to ensure the privacy and the data security of the user;
questionnaire data was collected: after the user fills out the questionnaire, the user's answer information is collected, recorded and stored. An online survey tool or questionnaire platform may be used to simplify the data collection and management process;
data analysis: statistical analysis and interpretation of the collected questionnaire data is performed to learn the interests and preferences of the user. Statistical software or data analysis tools can be used to process data and generate reports
The invalid time setting unit 20 is used for performing frequent analysis on the footprint data captured by the footprint capturing unit 10, and setting the invalid time of the footprint data according to the analysis result;
the invalid time setting unit 20 includes an invalid setting module, where the invalid setting module is configured to set an invalid time threshold corresponding to different occupied times in one day, and perform frequency analysis according to footprint data captured by the footprint capturing unit 10, obtain an average value of network footprint occupied times of a month user in each day, and compare the average value with the invalid time threshold according to the average value, so as to set a corresponding invalid time. The method comprises the following steps:
daily footprint occupancy time was analyzed: from the collected footprint data, a footprint occupancy time of the user on the network per day is calculated. The duration of each visit can be accumulated to obtain the total occupation time of each day;
the average value over one month was calculated: for each day of the month, an average of the daily footprint occupancy times was calculated. Dividing the total daily occupation time by the total number of days in one month to obtain an average value;
setting an invalid time threshold: and setting invalid time thresholds corresponding to different occupied times according to the requirements and acceptable definitions. Can be determined according to user habits, behavior patterns and business requirements; if the invalid time threshold is set to be one month, the invalid data is regarded as invalid data after exceeding one month;
comparing the average value to a threshold value: the average footprint occupancy time per day is compared to a set dead time threshold. If the average value exceeds the threshold value, the footprint time within the day is considered to be invalid;
adjusting the threshold according to the result: according to the comparison result, the invalid time threshold value can be adjusted according to the actual situation so as to better reflect the requirements and the behavior mode of the user.
The invalid data eliminating unit 30 is configured to divide the footprint data grabbed by the footprint grabbing unit 10 according to the time periods, compare the footprint data of each time period with the invalid time set by the invalid time setting unit 20, and eliminate the footprint data exceeding the invalid time;
the invalid data eliminating unit 30 includes a period dividing module and a data filtering module;
the time interval division module is used for dividing the footprint data captured by the footprint capturing unit 10 according to the three time intervals of morning, evening and evening to obtain an early footprint database, a middle footprint database and a late footprint database; the method comprises the following steps:
footprint data collection: collecting footprint data of a user on a network by using a network monitoring tool or an application analysis tool;
dividing time periods: each piece of footprint data is divided into three periods of early, medium and late according to time information in the footprint data. The time frame can be divided according to specific requirements and time ranges, such as 6:0011:59 in the morning, 12:0017:59 in the noon and 18:0023:59 in the evening;
early footprint database creation: storing footprint data divided into morning periods into an early footprint database;
creating a middle footprint database: storing footprint data divided into midday periods into a midday footprint database;
late footprint database creation: the footprint data divided into night periods is stored in a late footprint database.
The data screening module is used for comparing the early footprint database and the middle footprint database which are divided by the time interval dividing module with the invalid time set by the invalid setting module respectively, and eliminating the footprint data of which the recording time exceeds the invalid time in each time interval. The expression is as follows:
set dead time threshold:
invalid time start time: startTime;
end time of dead time: endTime;
early footprint database culls invalid data:
early footprint database = early footprint database [ record time ];
the middle footprint database eliminates invalid data:
middle footprint database = middle footprint database [ record time ];
late footprint database culls invalid data:
late footprint database = late footprint database [ record time.
The invalid data eliminating unit 30 further includes a user usage analysis module for performing frequent analysis according to the footprint data grabbed by the footprint grabbing unit 10, dividing the footprint data into a shift-in time and a shift-out time according to the analysis result, eliminating the footprint data used by the shift-in time, and transmitting the reserved data to the period dividing module for division. When predicting the behavior of the user, since the footprint data is changed along with the working content if the user is in duty, the footprint data has lower value of post-prediction due to the fact that the footprint data is not the behavior idea of the user, and meanwhile, the behavior prediction model is misled. The formula is as follows:
suppose the shift time is 9 a.m. to 6 a.m.:
and (3) data rejection during working hours: footprint data = footprint data [ time < '09:00:00' or time > '18:00:00' ];
the behavior data searching unit 40 is configured to screen according to the footprint data removed by the invalid data removing unit 30, obtain the feature data with the longest usage time in each period, and search the behavior data associated with the type in the footprint data according to the feature data;
the behavior data searching unit 40 includes a feature data acquiring module and an association matching module;
the characteristic data acquisition module is used for respectively carrying out data screening with the longest occupation time according to the early footprint database and the middle footprint database which are removed by the data screening module, and acquiring characteristic data with the longest occupation time corresponding to the late footprint database; the method comprises the following steps:
counting the sum of footprint occupation time in each time period: for an early footprint database, a middle footprint database and a late footprint database, calculating the sum of footprint occupation time in each time period respectively;
calculating the footprint ratio in the time period: dividing the total footprint occupation time in each time period by the total footprint occupation time to obtain the footprint occupation ratio in the time period;
screening the characteristic data with the highest duty ratio: and screening out the characteristic data with the highest duty ratio according to the footprint duty ratio in the time period.
The association matching module is used for performing behavior association analysis on the footprint data acquired by the footprint grabbing unit 10 according to the feature data, and acquiring behavior data of the footprint data and the feature data. The method comprises the following steps:
behavioral association analysis preparation: the feature data is prepared with the footprint data for behavioral association analysis. Ensuring that the data format and the table structure can be associated;
behavioral association analysis: performing association operation based on common fields of the feature data and the footprint data, such as user ID, date and the like, and finding out behavior association between the feature data and the footprint data;
extracting associated behavior data: and extracting behavior data associated with the feature data from the footprint data according to the result of the behavior association analysis.
The prediction data updating unit 50 is used for establishing a user behavior prediction model;
the prediction data updating unit 50 establishes a user behavior prediction model according to an early footprint database, a middle footprint database and a late footprint database by a neural network technology. The expression is as follows:
input layer: taking data of an early footprint database, a middle footprint database and a late footprint database as input;
hidden layer: the system comprises a plurality of neurons and is responsible for extracting characteristics of data and learning a mode of the data;
output layer: depending on the type of predicted target, there may be a two-classification, multi-classification, or regression output;
training process: through given training samples and labels, the weights and the biases of the neural network are adjusted by using optimization algorithms such as gradient descent and the like so as to minimize errors between the prediction and the true value;
the prediction process comprises the following steps: and calculating an output result by using a forward propagation algorithm according to the trained neural network model and the new input data to obtain a predicted result of the user behavior.
The characteristic data according to the time period is taken as behavior prediction data of the user, the latest footprint data is grabbed by combining the footprint grabbing unit 10 for feedback, and the behavior prediction data is updated according to the feedback result.
The prediction data update unit 50 includes a behavior prediction module and a feedback update module;
the behavior prediction module is used for sending the time period to the user behavior prediction model, and the user behavior prediction model is used for sending characteristic data of a footprint database as behavior prediction data according to the footprint database of the time corresponding to the time period; the method comprises the following steps:
acquiring the time period: determining a time period of the user according to the current time or other relevant time information, such as the morning, the afternoon, the evening and the like;
transmitting the time period to a user behavior prediction model: the acquired time period information is sent to a user behavior prediction model and is used as input of the prediction model;
period matching and footprint database selection: the user behavior prediction model matches corresponding footprint databases according to the received time period information, such as selecting an early footprint database according to the morning time period, selecting a middle footprint database according to the afternoon time period, and the like;
acquiring characteristic data of a footprint database: extracting feature data from the matched footprint database to be applied to the behavior prediction model;
transmitting the characteristic data to a user behavior prediction model: and taking the characteristic data of the footprint database as the input of the behavior prediction model, and sending the characteristic data to the user behavior prediction model for behavior prediction.
The feedback updating module is used for performing difference comparison according to the latest footprint data captured by the footprint capturing unit 10 and the behavior prediction data sent by the behavior prediction module, matching according to the latest footprint data in combination with the behavior data if the comparison result is wrong, sending the behavior prediction module to update by taking the behavior data as the behavior prediction data if the matching is correct, and continuing to predict if the comparison result is correct. The expression is as follows:
if prediction ! ＝ latestfootprint：
matchedbehaviordata ＝ match(latestfootprint,behaviordata)；
updatedprediction ＝ matchedbehaviordata send(updatedprediction)；
# sending the updated prediction data to the behavior prediction module;
else：
continuing prediction of continuous () # and;
wherein, the prediction: prediction data representing the behavior prediction results received from the behavior prediction module. latestfootprint: the latest footprint data represents the latest user footprint data acquired from the crawled data. matrichedbehaviordata: the matched behavior data represents the matched correct behavior data found after matching according to the latest footprint data and the behavior data. match (): and the matching function is used for matching the latest footprint data with the behavior data to find out the behavior data with correct matching. updatedpoint redirect: the updated prediction data represents the correct behavior data to be matched as new behavior prediction data. send (updatedprediction): and sending the updated prediction data to a behavior prediction module for updating the behavior prediction model. continuous prediction (): continuing the prediction means continuing the next prediction when the prediction coincides with the actual footprint.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the above-described embodiments, and that the above-described embodiments and descriptions are only preferred embodiments of the present invention, and are not intended to limit the invention, and that various changes and modifications may be made therein without departing from the spirit and scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (9)
1. The user behavior prediction system based on the network footprint analysis is characterized in that: comprises a footprint grabbing unit (10), an invalid time setting unit (20), an invalid data eliminating unit (30), a behavior data searching unit (40) and a predicted data updating unit (50);
the footprint grabbing unit (10) is used for grabbing footprint data of a user on a network;
the invalid time setting unit (20) is used for performing frequent analysis on the footprint data grabbed by the footprint grabbing unit (10) and setting the invalid time of the footprint data according to an analysis result;
the invalid data eliminating unit (30) is used for dividing the footprint data grabbed by the footprint grabbing unit (10) according to time intervals, comparing the footprint data of each time interval with the invalid time set by the invalid time setting unit (20), and eliminating the footprint data exceeding the invalid time;
the behavior data searching unit (40) is used for screening according to the footprint data which is removed by the invalid data removing unit (30), obtaining the characteristic data with the longest service time in each period, and searching the behavior data with the type association in the footprint data according to the characteristic data;
the predicted data updating unit (50) is used for establishing a user behavior predicted model, taking the characteristic data of the time period as behavior predicted data of the user, combining the latest footprint data grasped by the footprint grasping unit (10) for feedback, and updating the behavior predicted data according to the feedback result.
2. The network footprint based analysis user behavior prediction system of claim 1, wherein: the footprint grabbing unit (10) grabs the behavior trace data of the user on the network through a network footprint grabbing tool.
3. The network footprint based analysis user behavior prediction system of claim 1, wherein: the footprint grabbing unit (10) further comprises a user information collecting module, wherein the user information collecting module is used for sending an interest behavior hobby filling questionnaire to a user and acquiring footprint data of the user in a network according to the content of the user feedback questionnaire.
4. The network footprint based analysis user behavior prediction system of claim 1, wherein: the invalid time setting unit (20) comprises an invalid setting module, wherein the invalid setting module is used for setting invalid time thresholds corresponding to different occupied times in one day, and meanwhile, frequency analysis is carried out according to footprint data grabbed by the footprint grabbing unit (10), the average value of network footprint occupied times of a month user in each day is obtained, and the average value is compared with the invalid time thresholds, so that the corresponding invalid time is set.
5. The network footprint based analysis user behavior prediction system of claim 4, wherein: the invalid data eliminating unit (30) comprises a time interval dividing module and a data screening module;
the time interval division module is used for dividing the footprint data captured by the footprint capturing unit (10) according to the three time intervals of morning, evening and morning to obtain an early footprint database, a middle footprint database and a late footprint database;
the data screening module is used for comparing the early footprint database and the middle footprint database which are divided by the time interval dividing module with the invalid time set by the invalid setting module respectively, and eliminating the footprint data of which the record time exceeds the invalid time in each time interval.
6. The network footprint based analysis user behavior prediction system of claim 5, wherein: the invalid data eliminating unit (30) further comprises a user use analysis module, wherein the user use analysis module is used for carrying out frequent analysis according to the footprint data grabbed by the footprint grabbing unit (10), dividing the footprint data into the working time and the working time according to the analysis result, eliminating the footprint data used in the working time, and simultaneously sending the reserved data to the time interval dividing module for dividing.
7. The network footprint based analysis user behavior prediction system of claim 5, wherein: the behavior data searching unit (40) comprises a characteristic data acquisition module and an association matching module;
the characteristic data acquisition module is used for respectively carrying out data screening with the longest occupation time according to the early footprint database and the middle footprint database which are removed by the data screening module, and acquiring characteristic data with the longest occupation time corresponding to the late footprint database;
the association matching module is used for performing behavior association analysis on the footprint data acquired by the footprint grabbing unit (10) according to the characteristic data, and acquiring behavior data of the footprint data and the characteristic data in association.
8. The network footprint based analysis user behavior prediction system of claim 1, wherein: the prediction data updating unit (50) establishes a user behavior prediction model according to an early footprint database, a middle footprint database and a late footprint database through a neural network technology.
9. The network footprint based analysis user behavior prediction system of claim 1, wherein: the prediction data updating unit (50) comprises a behavior prediction module and a feedback updating module;
the behavior prediction module is used for sending the time period to the user behavior prediction model, and the user behavior prediction model is matched with the footprint database of the corresponding time according to the time period and sends the characteristic data of the footprint database as behavior prediction data;
the feedback updating module is used for carrying out difference comparison according to the latest footprint data acquired by the footprint acquiring unit (10) and the behavior prediction data sent by the behavior prediction module, carrying out matching according to the latest footprint data and the behavior data if the comparison result is wrong, sending the behavior prediction module to update by taking the behavior data as the behavior prediction data if the matching is correct, and continuing to predict if the comparison result is correct.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311206662.XA CN116955738B (en) | 2023-09-19 | 2023-09-19 | User behavior prediction system based on network footprint analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311206662.XA CN116955738B (en) | 2023-09-19 | 2023-09-19 | User behavior prediction system based on network footprint analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116955738A true CN116955738A (en) | 2023-10-27 |
| CN116955738B CN116955738B (en) | 2023-12-08 |
Family
ID=88458677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311206662.XA Active CN116955738B (en) | 2023-09-19 | 2023-09-19 | User behavior prediction system based on network footprint analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116955738B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2016212597A (en) * | 2015-05-07 | 2016-12-15 | 富士通株式会社 | Information collection program, information collection system, and information collection method |
| CN106611345A (en) * | 2015-10-23 | 2017-05-03 | 北京国双科技有限公司 | A method and apparatus for acquiring user behavior data |
| CN108108438A (en) * | 2017-12-20 | 2018-06-01 | 深圳市恒腾网络有限公司 | The recognition methods of behavioral data and device |
| CN112148770A (en) * | 2020-09-16 | 2020-12-29 | 翼集分电子商务(上海)有限公司 | User behavior prediction method and system based on multi-source data and semantic fusion |
| CN114840570A (en) * | 2022-03-11 | 2022-08-02 | 神策网络科技（北京）有限公司 | Data processing method and device, electronic equipment and storage medium |
-
2023
- 2023-09-19 CN CN202311206662.XA patent/CN116955738B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2016212597A (en) * | 2015-05-07 | 2016-12-15 | 富士通株式会社 | Information collection program, information collection system, and information collection method |
| CN106611345A (en) * | 2015-10-23 | 2017-05-03 | 北京国双科技有限公司 | A method and apparatus for acquiring user behavior data |
| CN108108438A (en) * | 2017-12-20 | 2018-06-01 | 深圳市恒腾网络有限公司 | The recognition methods of behavioral data and device |
| CN112148770A (en) * | 2020-09-16 | 2020-12-29 | 翼集分电子商务(上海)有限公司 | User behavior prediction method and system based on multi-source data and semantic fusion |
| CN114840570A (en) * | 2022-03-11 | 2022-08-02 | 神策网络科技（北京）有限公司 | Data processing method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 张甲 等: "基于事件序列的蠕虫网络行为分析算法", 山东大学学报（理学版）, vol. 42, no. 9 * |
| 王海雷: "搜索引擎广告用户行为预测与特征分析", 计算机应用研究, vol. 30, no. 5 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116955738B (en) | 2023-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107526807B (en) | Information recommendation method and device | |
| CN107958317A (en) | A kind of method and apparatus that crowdsourcing participant is chosen in crowdsourcing project | |
| CN108874959B (en) | User dynamic interest model building method based on big data technology | |
| CN101894351A (en) | Multi-agent based tour multimedia information personalized service system | |
| CN103218431A (en) | System and method for identifying and automatically acquiring webpage information | |
| CN109544197A (en) | A kind of customer churn prediction technique and device | |
| CN112149352B (en) | Prediction method for marketing activity clicking by combining GBDT automatic characteristic engineering | |
| CN109815631A (en) | A kind for the treatment of method and apparatus of game data | |
| CN104992318B (en) | Method for actively recommending events by calendar | |
| CN112632405A (en) | Recommendation method, device, equipment and storage medium | |
| CN111652735A (en) | Insurance product recommendation method based on user behavior label characteristics and commodity characteristics | |
| CN112468853A (en) | Television resource recommendation method and device, computer equipment and storage medium | |
| CN111105849A (en) | Channel collaborative satisfaction investigation method and system based on big data | |
| CN116955738B (en) | User behavior prediction system based on network footprint analysis | |
| CN116304351B (en) | Intelligent data information statistical management system and method based on big data | |
| CN116320626B (en) | Method and system for calculating live broadcast heat of electronic commerce | |
| CN116452212A (en) | Intelligent customer service commodity knowledge base information management method and system | |
| CN112734142B (en) | Resource learning path planning method and device based on deep learning | |
| CN110335626A (en) | Age recognition methods and device, storage medium based on audio | |
| CN111143688B (en) | Evaluation method and system based on mobile news client | |
| CN114417884A (en) | Intelligent display device and method for information technology consultation service | |
| CN106202473A (en) | A kind of brands based on big data analyze method and system | |
| CN101655850B (en) | Generating equipment for knowledge extraction process, regulating equipment for knowledge extraction process and methods thereof | |
| CN117271905B (en) | Crowd image-based lateral demand analysis method and system | |
| Sudhakar et al. | Web User Profile Generation and Discovery Analysis using LSTM Architecture |
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 |