CN109978627B - Modeling method for big data of broadband access network user surfing behavior - Google Patents
Modeling method for big data of broadband access network user surfing behavior Download PDFInfo
- Publication number
- CN109978627B CN109978627B CN201910250704.7A CN201910250704A CN109978627B CN 109978627 B CN109978627 B CN 109978627B CN 201910250704 A CN201910250704 A CN 201910250704A CN 109978627 B CN109978627 B CN 109978627B
- Authority
- CN
- China
- Prior art keywords
- data
- user
- internet surfing
- flow
- behavior
- 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
- 238000000034 method Methods 0.000 title claims abstract description 16
- 230000006399 behavior Effects 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 238000013441 quality evaluation Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 claims description 3
- 238000010219 correlation analysis Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 description 10
- 238000001303 quality assessment method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005206 flow analysis Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000010801 machine learning Methods 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
- 238000005457 optimization Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G06Q50/40—
-
- 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
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Abstract
The invention discloses a modeling method for big data of a broadband access network user surfing behavior, which provides a semi-supervised learning method based on a clustering algorithm and a regression algorithm for network traffic.
Description
Technical Field
The invention relates to the field of big data mining analysis and big data modeling, in particular to a modeling method for big data of surfing behavior of a broadband access network user.
Background
With the development of the internet, the number of network users is gradually increasing. In the field of telecommunications, there are hundreds of millions of broadband access network users, and the types of data provided by these broadband access network users are rich and diverse, including user basic data (such as user identity ID, user attribution, date of birth, occupation), internet surfing behavior data (such as internet surfing traffic, internet surfing time, browsing content, search keywords), and location data (regional climate, regional economic total amount, regional broadband access network user number); the higher requirements of broadband access network users on network quality and network services make network operators and service providers continually improve network service quality and add new services; in addition, the user data of the broadband access network are analyzed and mined timely and effectively, the distribution situation of different user groups in space is found, and guidance is provided for building an optimization server.
To improve the user satisfaction, the user is provided with interesting business and information, the behavior of the network user is required to be analyzed, the internet surfing characteristics, internet surfing interests and the like of the user are explored, so that the user requirements are deeply known, and meanwhile, an important information channel is provided for network marketing; to improve the network quality, the operation condition and the service condition of the network must be known in depth, the monitoring of the network traffic is maintained, the network structure and the bandwidth are continuously adjusted, the network problem is solved, the network service quality is improved, and a large amount of network traffic data is effectively processed; to optimize the construction of the server, the information such as the income, the regional difference, the service demand change, the distribution condition of different user groups and the proportion thereof of different services must be known, and in order to meet the demands of different user groups in the area on the aspects such as bandwidth, service and the like, the distribution proportion characteristic of the locations of different user groups on the server is obtained, so that the infrastructure cost of the server is reduced.
In order to maximize the utilization of customer resources, there are currently a number of methods for telecommunication broadband access network user behavior analysis and prediction. For the prediction of continuous telecommunication broadband access network user traffic data, various supervised learning methods have been proposed. Some of these studies consider network traffic as a linear model and predict using linear models such as auto-regressive moving average (ARMA) model, differential auto-regressive moving average (ARIMA) model, and differential auto-regressive sum-of-motion average (FARIMA) model, respectively. However, as the complexity of the network increases, the traffic characteristics of the network exceed the poisson distribution or markov distribution which is conventionally considered, so that the prediction by using the linear model has theoretical defects, and it is difficult to ensure the accuracy of the prediction. The prediction of the nonlinear model mainly comprises an artificial neural network, a support vector machine, a gray model and the like, and although the prediction precision of the nonlinear model is improved to a certain extent compared with that of the linear model, the prediction precision is still not ideal. The neural network has the defects of easy localization of local optimal values and difficult determination of network structures; although the number of samples required by the support vector machine is small, key parameters of the support vector machine are difficult to determine; while the gray model is only suitable for the case where the data change is not severe; therefore, there is a need to develop and design a new modeling method to optimize server construction and improve network quality and user satisfaction.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a modeling method for big data of the surfing behavior of a broadband access network user.
The technical scheme adopted for solving the technical problems is as follows:
a modeling method for big data of broadband access network user surfing behavior includes the following steps:
s1, acquiring internet surfing behavior data of a broadband access network user, performing quality evaluation on the data, and screening out high-quality data.
S2, preprocessing the screened high-quality data, carrying out user region division and marking on the screened high-quality data by using an unsupervised algorithm, and searching for the association relation between the Internet surfing behavior of the user and each data field by using an association algorithm in unsupervised learning in combination with the basic data and the position data of the user.
S3, predicting the time-flow of the marked user group by using a regression model in supervised learning to obtain the flow trend of each user group, and obtaining the total flow trend by statistical calculation.
S4, obtaining distribution conditions and quantity of each marked user group through statistical calculation, and obtaining the requirement characteristics of different user groups on the server.
Further, in the step S1, according to the basic information (such as the flow size and the data type) of the broadband access network user internet surfing behavior data, a chart is drawn, and the data is analyzed through the chart, so that unnecessary data is removed, and high-quality data is obtained.
In the step S2, the high-quality data is divided into two data packets, namely DataBill and Factorbill, wherein DataBill is two training sets formed by taking the internet surfing behavior (such as internet surfing time, browsing content, search keywords and the like) in the user internet surfing behavior data as a feature vector, taking the internet surfing flow (uplink flow and downlink flow) of the user internet surfing behavior data, and respectively performing training research on the internet surfing behavior data in time and space by using an unsupervised algorithm; factorBill consists of user profile data and location data.
The beneficial effects of the invention are as follows: according to the method, the regional user subdivision and marking are dynamically carried out on the user Internet surfing data by using a clustering algorithm, regional flow prediction is carried out on the subdivided user groups, and a regional server construction scheme is given through calculation statistics and is respectively provided for marketing departments, operation departments and infrastructure departments, so that fusion among all departments of telecommunication is innovatively promoted, network quality is effectively improved through construction of an optimized server, and better Internet surfing experience is brought to users.
Drawings
The invention will be further described with reference to the drawings and examples.
FIG. 1 is a diagram of modeling steps of the present invention;
FIG. 2 is a block diagram of the overall modeling of the present invention;
FIG. 3 is a block diagram of FIG. 2 at A;
FIG. 4 is a block diagram of FIG. 2 at B;
FIG. 5 is a block diagram of FIG. 2 at C;
FIG. 6 is a ratio diagram of different data types for data quality assessment of the present invention;
FIG. 7 is a ratio plot of each type of data in the analyzable data of the data quality assessment of the present invention;
FIG. 8 is a bar graph of the time coverage length of each type of data for data quality assessment of the present invention;
FIG. 9 is a time distribution plot of total data volume for data quality assessment of the present invention;
FIG. 10 is a block diagram of DataBill and FactorBill for data preprocessing of the present invention.
Detailed Description
Referring to fig. 1 to 10, a modeling method for big data of broadband access network user internet surfing behavior includes the following steps:
s1, acquiring internet surfing behavior data of a broadband access network user, performing quality evaluation on the data, and screening out high-quality data.
S2, preprocessing the screened high-quality data, carrying out user region division and marking on the screened high-quality data by using an unsupervised algorithm, and searching for the association relation between the Internet surfing behavior of the user and each data field by using an association algorithm in unsupervised learning in combination with the basic data and the position data of the user.
S3, predicting the time-flow of the marked user group by using a regression model in supervised learning to obtain the flow trend of each user group, and obtaining the total flow trend by statistical calculation.
S4, obtaining distribution conditions and quantity of each marked user group through statistical calculation, and obtaining the requirement characteristics of different user groups on the server.
Further, in the step S1, according to the basic information (such as the flow size and the data type) of the broadband access network user internet surfing behavior data, a chart is drawn, and the data is analyzed through the chart, so that unnecessary data is removed, and high-quality data is obtained.
In step S2, the high-quality data is divided into two data packets, namely DataBill and FactorBill, wherein DataBill is two training sets formed by taking the internet surfing behavior (such as internet surfing time, browsing content, search keywords, etc.) in the internet surfing behavior data of the user as a feature vector, taking the internet surfing flow (uplink flow and downlink flow) of the internet surfing behavior data of the user, and respectively performing training research on the internet surfing behavior data in time and space by using an unsupervised algorithm; the factor bill is composed of user basic data and location data, referring specifically to fig. 10, table 1, table 2 and table 3, in this embodiment, the training set obtained by training and researching the DataBill in space is denoted as B1 (which is a cell array of 1*7), the training set obtained by training and researching in time is denoted as B2 (which is a cell array of 1×22), the user basic data (denoted as UserFactor attribute) includes user ID, user home location, date of birth, occupation and income, the user location data (denoted as NaturalFactor attribute) includes week, weather, temperature, air quality, regional economic total amount and regional user number, and the correlation analysis is performed between the B1 and B2 and the user basic data and the location data, and for the nominal attributes such as week, weather and region, a difference method is adopted to take values, in this embodiment, the same distance is 1, and different values are taken as 0.
Table 1 data preprocessed space dimension training set B1 and time dimension training set B2
TABLE 2 UserFactor attribute for data preprocessing
| Attribute name | User identity ID | User home location | Birth date | Occupation of | Income (income) |
| Attribute type | Nominal scale | Nominal scale | Numerical value | Nominal scale | Numerical value |
TABLE 3 NaturalFactor Properties of data pretreatment
| Attribute name | Week of week | Weather of | Temperature (temperature) | Air quality | Region of | Total amount of regional economy | Number of regional users |
| Attribute type | Nominal scale | Nominal scale | Numerical value | Numerical value | Nominal scale | Numerical value | Numerical value |
Examples
In the embodiment, statistical information of large data of user behaviors of broadband access networks in Sichuan areas is selected, and the statistical information is divided into 6 different types according to 446.8MB of records and 1606995 records, wherein the time span is from 1 month, 23 days in 2015 to 4 months, 10 days in 2017; the basic information for different types of data can be seen in table 4, and the proportion of the different types of data to the total number can be seen clearly from fig. 6.
Table 4 basic information of six types of data for data quality assessment
Next, screening and removing the data, referring to table 5, it can be known that the data such as the number of users, the internet time, the browsing content, the search keyword, the number of entrance (uploading) bytes, the number of exit (downloading) bytes, the entrance (uploading) rate, the exit (downloading) rate, the total rate, the access number and the like are taken as the analyzable data, and the date, the user attribution, the birth date, the occupation, the value (sampling) time interval, the remote server IP and the bandwidth type are taken as the classification data; while the proportions of the analyzable data in the different data types are different, in Table 6 we can see this difference, for which we can let the data we use to analyze be what we really care, the amount of analyzable data can measure whether a data type is more important or has a higher data quality. The pie chart of fig. 7 is used for showing the proportion of each data type in all the analyzable data, so that the data of which type is more valuable to analyze is compared, and by combining fig. 6, all-time traffic statistics, 100M user analysis, online user analysis, bandwidth-divided online user analysis and peak traffic statistics types can be found to have considerable amounts of analyzable data, the data quantity of the accumulated users is small, and even the data of the type of the accumulated users can be considered to have no analytical value in the dimension of the data quantity, so that the accumulated user data can be removed.
Table 5 structural list of different types of data for data quality assessment
TABLE 6 proportion of analysis class data of different types of data for data quality assessment
In this embodiment, the data is widely distributed in time (from 23 in 2015 to 4 in 2017, 10 in 2015), and since the data is discontinuous in time and not all time points have complete data (data has missing values), the data in a certain period is selected and discarded, so that the data quality is improved, referring to table 7, fig. 8 and fig. 9, the time coverage length of various data types can be seen, and by observing, no data exists in the time from 30 in 2015 to 4 in 2016 up to 247 days, and the data amount is not too much after 1 in 2016, that is, the data richness after 7 in 2015 is insufficient, the data amount is obviously reduced by all types of data in the data set before 30 in 2015, and the total flow statistical data amount after 4 in 2016 is obviously reduced, and several values are obviously abnormal. Since the continuity of data is poor and the data quality is poor after the period of 30 months of 7 in 2015, we can improve the overall data quality by discarding the following data.
TABLE 7 time coverage length of various types of data for data quality assessment
| Number of records | Type weight (min) | Time coverage length (min) | |
| 100M user analysis | 73160 | 5 | 365800 |
| On-line user analysis | 353007 | 5 | 1765035 |
| Banded online user analysis | 762022 | 5 | 3810110 |
| Peak flow statistics | 30710 | 5 | 153550 |
| Full time period flow analysis | 994230 | 5 | 4971150 |
| Accumulating the number of users | 9552 | 35 | 334320 |
Through analysis and screening of the data, the study samples are finally determined to be 100M users in 22 areas of Sichuan province, wherein the peak period is 20:00 to 22:30 late in 2015, 1 month, 23 days, to 2015, 7 month, 30 days, and 20:00.
In the embodiment, the screened data are divided into two groups, namely X|Y (X or Y) and X+Y (X and Y), wherein X represents the internet surfing time, browsing content and search keywords in the user internet surfing behavior data, Y represents the internet surfing flow in the user internet surfing behavior data, a clustering algorithm is carried out on the X+Y to obtain marks, the marks are respectively added to the two groups of data sets to obtain X|Y marks and X+Y marks, the association analysis is carried out on the X+Y marks and the basic data and the position data of the user, and the association relation between the user internet surfing behavior and different data such as user business behavior, daily life and the like is deeply mined, so that a user group is divided, and enterprises can conveniently carry out sales means aiming at user group differentiation; the data marked by x|y is used for regression algorithm (supervised learning) to predict the flow of various users, specifically, if a region (denoted as P1) is subdivided and marked by regional users, 3 user groups U1 to U3 are obtained, namely U (P1) = { U1}, { U2}, { U3}, where U1= { U11, U12, U13, U14, U15}, U2= { U21, U22, U23, U24, U25}, U3= { U31, U32, U33, U34, U35, U36}. Taking 7 days a week as an example, for each user group, machine learning is performed according to training samples of monday to Saturday, and the regression algorithm is used for predicting the traffic trend of the monday, and for the same user group, the surfing traffic trend of the same user group is similar because the surfing behavior features of the same user group are similar, so that the prediction complexity is greatly reduced. And aiming at a certain sub-region P11-P15, adding the predicted flow statistics of the corresponding users to obtain the total flow trend of the sub-region. As shown in fig. 4, comparing and analyzing the flow trend of P11-P15, the method can provide a basis for resource allocation for network development in advance; the predicted flow of P11-P15 is statistically analyzed to obtain the prediction or trend of the flow in the P1 region, so that the prediction precision of the flow can be effectively improved by the total-to-total method; finally, according to the obtained data of the X+Y marks, the proportion of the user groups in each sub-area is calculated, the user group with the largest proportion is obtained, the server is built for meeting the requirement of the user group with the larger proportion, the distribution condition and the quantity of the regional servers can be obtained, and in fig. 5, the proportion of the users in the P1 area is calculated as follows:
in the present embodiment, the concept of "dictionary" is presented, and in fig. 5, the "dictionary" is generated by P11, P12, P14, and P13, P15 can be derived by "looking up the dictionary". The dictionary should have integrity (including various user groups in P1) and independence (including the user groups independent of each other). According to calculation, the proportion of the user group U1 in the P11 occupied by the area is the largest, so that the requirement of the U1 on the P11 server in the place is obtained to meet the requirement of the user group U1 on bandwidth, service and the like, and further the requirement of the P11 on the distribution and the quantity of the servers is obtained. Similarly, the requirements of the user groups U3 and U2 on the server distribution and number of the locations P12 and P14 are respectively obtained by P12 and P14, which creates a "dictionary" containing all types of user groups P1. Further, the overlapping area of the P1 server distribution is optimized, so that the distribution condition and the number of the P1 regional servers are obtained, and a server construction scheme is provided for an infrastructure department.
The above embodiments do not limit the protection scope of the invention, and those skilled in the art can make equivalent modifications and variations without departing from the whole inventive concept, and they still fall within the scope of the invention.
Claims (1)
1. A modeling method for big data of broadband access network user surfing behavior is characterized by comprising the following steps:
s1, acquiring internet behavior data of a broadband access network user, performing quality evaluation on the data, and screening out high-quality data;
s2, preprocessing the screened high-quality data, carrying out user region division and marking on the screened high-quality data by using an unsupervised algorithm, and searching for the association relation between the Internet surfing behavior of the user and each data field by using an association algorithm in unsupervised learning in combination with the basic data and the position data of the user;
dividing the screened data into two groups, namely X-Y, namely X or Y and X+Y, namely X and Y, wherein X represents the internet surfing time, browsing content and search keywords in the user internet surfing behavior data, Y represents the internet surfing flow in the user internet surfing behavior data, clustering algorithm is carried out on the X+Y to obtain marks, the marks are respectively added to the two groups of data sets to obtain X-Y marks and X+Y marks, correlation analysis is carried out on the X+Y marks and the basic data and the position data of the user, and the correlation relation between the user internet surfing behavior and different data in user business behaviors and daily life is deeply excavated, so that user groups are divided;
s3, predicting time-flow of the marked user group by using a regression model in supervised learning to obtain flow trend conditions of each user group, and obtaining total flow trend conditions through statistical calculation;
s4, obtaining the distribution condition and the quantity of each marked user group through statistical calculation, so as to obtain the requirement characteristics of different user groups on the server;
in the step S1, according to basic information of broadband access network user internet surfing behavior data, including flow size and data type, a chart is drawn, and unnecessary data is removed by analyzing the data through the chart, so as to obtain high-quality data;
in the step S2, the high-quality data is divided into two data packets, namely a DataBill and a FactorBill, wherein the DataBill is two training sets formed by taking the internet surfing behavior in the user internet surfing data as a feature vector, taking the internet surfing flow of the user internet surfing behavior data, including uplink flow and downlink flow, and respectively performing training research on the data in time and space by using an unsupervised algorithm; factorBill consists of user profile data and location data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910250704.7A CN109978627B (en) | 2019-03-29 | 2019-03-29 | Modeling method for big data of broadband access network user surfing behavior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910250704.7A CN109978627B (en) | 2019-03-29 | 2019-03-29 | Modeling method for big data of broadband access network user surfing behavior |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109978627A CN109978627A (en) | 2019-07-05 |
| CN109978627B true CN109978627B (en) | 2023-08-08 |
Family
ID=67081796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910250704.7A Active CN109978627B (en) | 2019-03-29 | 2019-03-29 | Modeling method for big data of broadband access network user surfing behavior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109978627B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111030893A (en) * | 2019-12-31 | 2020-04-17 | 上海途鸽数据科技有限公司 | Method and device for analyzing user behaviors in cloud communication application scene |
| CN111245650A (en) * | 2020-01-07 | 2020-06-05 | 广东九联科技股份有限公司 | Network bandwidth optimization management method based on machine learning |
| CN111815361A (en) * | 2020-07-10 | 2020-10-23 | 北京思特奇信息技术股份有限公司 | Region boundary calculation method and device, electronic equipment and storage medium |
| CN112910984B (en) * | 2021-01-26 | 2023-05-23 | 国网福建省电力有限公司泉州供电公司 | Electric power Internet of things flow prediction method based on FGn and Poisson processes |
| CN116723339B (en) * | 2023-08-11 | 2023-11-14 | 腾讯科技（深圳）有限公司 | Content data distribution method and device, storage medium and electronic equipment |
| CN116886571B (en) * | 2023-09-07 | 2023-11-21 | 武汉博易讯信息科技有限公司 | Analysis method, equipment and computer readable medium for home broadband user |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101990003A (en) * | 2010-10-22 | 2011-03-23 | 西安交通大学 | User action monitoring system and method based on IP address attribute |
| CN102238045A (en) * | 2010-04-27 | 2011-11-09 | 广州迈联计算机科技有限公司 | System and method for predicting user behavior in wireless Internet |
| CN102752123A (en) * | 2011-04-20 | 2012-10-24 | 中国移动通信集团设计院有限公司 | Method and device for forecasting flow and configuring capacity of network equipment interface |
| CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
| CN104331404A (en) * | 2013-07-22 | 2015-02-04 | 中国科学院深圳先进技术研究院 | A user behavior predicting method and device based on net surfing data of a user's cell phone |
| CN107682178A (en) * | 2017-08-30 | 2018-02-09 | 国信优易数据有限公司 | A kind of mobile subscriber's online operation behavior Forecasting Methodology and device |
| CN108462888A (en) * | 2018-03-14 | 2018-08-28 | 江苏有线数据网络有限责任公司 | The intelligent association analysis method and system of user's TV and internet behavior |
| CN109495317A (en) * | 2018-12-13 | 2019-03-19 | 中国南方电网有限责任公司 | Data network method for predicting and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160285704A1 (en) * | 2015-03-27 | 2016-09-29 | Iosif Gasparakis | Technologies for dynamic network analysis and provisioning |
| CN105808639B (en) * | 2016-02-24 | 2021-02-09 | 平安科技（深圳）有限公司 | Network access behavior identification method and device |
-
2019
- 2019-03-29 CN CN201910250704.7A patent/CN109978627B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102238045A (en) * | 2010-04-27 | 2011-11-09 | 广州迈联计算机科技有限公司 | System and method for predicting user behavior in wireless Internet |
| CN101990003A (en) * | 2010-10-22 | 2011-03-23 | 西安交通大学 | User action monitoring system and method based on IP address attribute |
| CN102752123A (en) * | 2011-04-20 | 2012-10-24 | 中国移动通信集团设计院有限公司 | Method and device for forecasting flow and configuring capacity of network equipment interface |
| CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
| CN104331404A (en) * | 2013-07-22 | 2015-02-04 | 中国科学院深圳先进技术研究院 | A user behavior predicting method and device based on net surfing data of a user's cell phone |
| CN107682178A (en) * | 2017-08-30 | 2018-02-09 | 国信优易数据有限公司 | A kind of mobile subscriber's online operation behavior Forecasting Methodology and device |
| CN108462888A (en) * | 2018-03-14 | 2018-08-28 | 江苏有线数据网络有限责任公司 | The intelligent association analysis method and system of user's TV and internet behavior |
| CN109495317A (en) * | 2018-12-13 | 2019-03-19 | 中国南方电网有限责任公司 | Data network method for predicting and device |
Non-Patent Citations (1)
| Title |
|---|
| "智能接入网用户行为建模及管控研究";倪明;《中国优秀硕士学位论文全文数据库 (信息科技辑)》(第12期);I138-299 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109978627A (en) | 2019-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109978627B (en) | Modeling method for big data of broadband access network user surfing behavior | |
| CN110245981B (en) | Crowd type identification method based on mobile phone signaling data | |
| CN100499512C (en) | Statistical method and device of internet data information clicking rates | |
| CN108417033B (en) | Expressway traffic accident analysis and prediction method based on multi-dimensional factors | |
| CN111614690B (en) | Abnormal behavior detection method and device | |
| CN110730099B (en) | Flow prediction method based on historical flow data time sequence | |
| CN101789887A (en) | Method and device for classifying network users and system for monitoring network services | |
| Fiadino et al. | Steps towards the extraction of vehicular mobility patterns from 3G signaling data | |
| CN111882367B (en) | Method for monitoring and tracking online advertisements through analysis of user surfing behavior | |
| DE102012223468A1 (en) | Determine a common starting point, destination, and route from a network record | |
| CN108462888A (en) | The intelligent association analysis method and system of user's TV and internet behavior | |
| CN106657007A (en) | Method for recognizing abnormal batch ticket booking behavior based on DBSCAN model | |
| CN106332052B (en) | Micro-area public security early warning method based on mobile communication terminal | |
| DE102012223496A1 (en) | ONLINE AND DISTRIBUTED OPTIMIZATION FRAMEWORK FOR MOBILE RADIO ANALYSIS | |
| CN107979602A (en) | A kind of method for detecting abnormality based on semi-supervised statistics in cellular network | |
| CN112149352B (en) | Prediction method for marketing activity clicking by combining GBDT automatic characteristic engineering | |
| CN112261645A (en) | Mobile application fingerprint automatic extraction method and system based on grouping and domain division | |
| CN113312531A (en) | User portrait identification method based on DPI analysis and decision tree model | |
| CN111475727B (en) | Floor page pushing method and device | |
| Mata et al. | Detection of traffic changes in large-scale backbone networks: The case of the Spanish academic network | |
| CN104750829A (en) | User position classifying method and system based on signing in features | |
| CN107832333A (en) | Method and system based on distributed treatment and DPI data structure user network data fingerprint | |
| Chu et al. | Exploiting spatial-temporal behavior patterns for fraud detection in telecom networks | |
| CN108597224B (en) | Method and system for identifying to-be-improved traffic facilities based on space-time trajectory data | |
| AT&T |
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 |