CN116108291B - Mobile internet traffic service recommendation method and system - Google Patents
Mobile internet traffic service recommendation method and system Download PDFInfo
- Publication number
- CN116108291B CN116108291B CN202310384249.6A CN202310384249A CN116108291B CN 116108291 B CN116108291 B CN 116108291B CN 202310384249 A CN202310384249 A CN 202310384249A CN 116108291 B CN116108291 B CN 116108291B
- Authority
- CN
- China
- Prior art keywords
- flow
- total
- traffic
- user
- time period
- 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 40
- 238000012360 testing method Methods 0.000 claims description 68
- 238000012549 training Methods 0.000 claims description 30
- 230000003203 everyday effect Effects 0.000 claims description 11
- 238000003066 decision tree Methods 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 2
- 238000004590 computer program Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 230000006399 behavior Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000013136 deep learning model Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000004075 alteration Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000002699 waste material Substances 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
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G06Q50/40—
Abstract
The invention provides a mobile internet flow service recommendation method and a system, which are used for counting the time sequence of the user using flow, acquiring the potential flow demand of the user through time sequence prediction, simultaneously matching the flow demand of the user with the content of the existing flow package, completing the flow package recommendation of the user, combining the flow demand prediction of the user and the service recommendation with high cost performance, carrying out more accurate personalized recommendation service for the user, simultaneously considering the cost performance problem of purchasing products by the user, improving the satisfaction degree of the user to the products and improving the viscosity of the user.
Description
Technical Field
The invention relates to the field of internet service recommendation, in particular to a mobile internet traffic service recommendation method and system.
Background
With the development of information industry, the internet, especially the mobile internet, has completely integrated into people's life, and the demands of users for internet resources are also increasing. It is also desirable for mobile operators to be able to provide rich, comprehensive, services that are tailored to the users themselves. However, due to the diversity of users, services such as various packages, traffic packages, etc. provided by mobile operators may not fully satisfy different user groups, and some users often need to purchase traffic packages outside packages to satisfy the requirements, while avoiding a large amount of additional costs caused by exceeding the traffic outside packages. Therefore, how to accurately recommend the flow service to be considered by the mobile operator reduces the extra cost of the user as much as possible on the premise of meeting the user demand, and increases the satisfaction degree of the user to the operator and the viscosity of the user on the basis of increasing the income of the operator.
Today in data science development, most of records of users can be recorded, and information such as contents of user subscription services, traffic usage rates, and traffic usage preferences are recorded by a server of a mobile operator.
Disclosure of Invention
The invention provides a mobile internet traffic service recommendation method and a mobile internet traffic service recommendation system aiming at the technical problems existing in the prior art.
According to a first aspect of the present invention, there is provided a mobile internet traffic service recommendation method, comprising:
determining a time period T based on total duration of surfing time of user surfing log 1 And time period T 2 ；
Acquiring the time period T of each object user 1 Total internet traffic per day and during time period T 2 Total flow of demand in, wherein all subject users are in time period T 1 Total internet traffic per day and during time period T 2 The total flow required in the system forms a total flow use time sequence;
training the prediction model by taking the total flow using time sequence as a training data set to obtain a trained prediction model;
predicting the total flow of the demand of the test user in a time period to be predicted based on the trained prediction model, and obtaining a total flow demand prediction result;
and matching the existing structured flow service products according to the ordered flow service products of the test user and the total flow demand prediction result, and recommending the matched flow service products to the test user.
According to a second aspect of the present invention, there is provided a mobile internet traffic service recommendation system comprising:
the acquisition module is used for determining a time period T based on the total duration of the Internet surfing time of the Internet surfing log of the user 1 And time period T 2 Acquiring the time period T of each object user 1 Total internet traffic per day and during time period T 2 Total flow of demand in, wherein all subject users are in time period T 1 Total internet traffic per day and during time period T 2 The total flow required in the system forms a total flow use time sequence;
the training module is used for training the prediction model by taking the total flow using time sequence as a training data set to obtain a trained prediction model;
the prediction module is used for predicting the total flow of the requirements of the test user in the period to be predicted based on the trained prediction model, and obtaining a total flow demand prediction result;
and the recommending module is used for matching the existing structured flow service products according to the flow service products ordered by the test user and the total flow demand predicting result, and recommending the matched flow service products to the test user.
According to a third aspect of the present invention, there is provided an electronic device comprising a memory, a processor for implementing the steps of a mobile internet traffic service recommendation method when executing a computer management class program stored in the memory.
According to a fourth aspect of the present invention there is provided a computer readable storage medium having stored thereon a computer management class program which when executed by a processor implements the steps of a mobile internet traffic service recommendation method.
According to the mobile internet traffic service recommendation method and system, the time sequence of the traffic used by the user is counted, the potential traffic demand of the user is obtained through time sequence prediction, meanwhile, the traffic demand of the user is matched with the content of the existing traffic packet, the traffic packet recommendation of the user is completed, the traffic demand prediction and the high-cost performance service recommendation of the user are combined, more accurate personalized recommendation service can be performed for the user, meanwhile, the cost performance problem of purchasing products by the user is considered, the satisfaction degree of the user to the products is improved, and the viscosity of the user is improved.
Drawings
FIG. 1 is a flow chart of a mobile Internet traffic service recommendation method provided by the invention;
FIG. 2 is a flow chart of a flow difference calculation process;
FIG. 3 is a schematic diagram of a flow service product matching process;
fig. 4 is a schematic structural diagram of a mobile internet traffic service recommendation system according to the present invention;
fig. 5 is a schematic hardware structure of one possible electronic device according to the present invention;
fig. 6 is a schematic hardware structure of a possible computer readable storage medium according to the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. 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. In addition, the technical features of each embodiment or the single embodiment provided by the invention can be combined with each other at will to form a feasible technical scheme, and the combination is not limited by the sequence of steps and/or the structural composition mode, but is necessarily based on the fact that a person of ordinary skill in the art can realize the combination, and when the technical scheme is contradictory or can not realize, the combination of the technical scheme is not considered to exist and is not within the protection scope of the invention claimed.
There has been a great deal of related research on the recommendation method of mobile internet services, most of which are based on an assumption that: that is, the behavior patterns of the user during a period of time are the same, so that most of the existing recommendation methods find out the behavior patterns of the user according to the behavior of the user during the previous period of time, and consider that the behavior patterns of the user during the next period of time are the same as the found behavior patterns of the user. This recommendation method is affected by several factors including the user's satisfaction with the mobile network service, the user's preferences for various APP uses (including time and place of use), and the duration of the user's behavioral patterns.
Based on the method, the system and the device, the users buying the package outside the package for multiple times are taken as objects, the behavior mode of the users using the flow is found, the flow demand of the users in the later time period is predicted, the contents of the existing package service are combined, personalized recommendation is provided for the object users, the cost waste of the users caused by buying the package for multiple times is reduced, and the user satisfaction is improved.
Fig. 1 is a flowchart of a mobile internet traffic service recommendation method provided by the present invention, where, as shown in fig. 1, the method includes:
s1, determining a time period T based on the total duration of the Internet surfing time of a user Internet surfing log 1 And time period T 2 。
As an embodiment, the time period T is determined based on the total duration of the user log 1 And time period T 2 Comprising: extracting internet surfing logs of all users within a set time period, wherein the internet surfing logs at least comprise user IDs, internet surfing time, access addresses and internet surfing flow; extracting all user IDs in the Internet surfing log and the total Internet surfing time duration contained in the Internet surfing log, and equally dividing the total Internet surfing time duration into time periods T 1 And time period T 2 。
S2, each object user is in a time period T 1 Total internet traffic per day and during time period T 2 Total flow of demand in, wherein all subject users are in time period T 1 Total internet traffic per day and during time period T 2 The total flow required in the system constitutes a total flow usage time sequence.
As an embodiment, the acquiring each object user is performed during a time period T 1 Total internet traffic and time of dayInterval T 2 The total flow of demand in, include: extracting all flow service product information of a mobile operator to form structural product characteristics of each flow server product, wherein the flow service product information comprises packages and flow packets; selecting an object user who has ordered a flow service product with a structured product feature from all users; based on the internet log of each object user, counting each object user in a time period T 1 Total internet traffic F used every day a During a time period T 2 The total online traffic used every day in the system constitutes the total traffic use time sequence of each object user.
It will be appreciated that extracting all of the product information of the mobile operator, including packages and traffic packages, to form structured product features, removes unstructured products from the product information. Extracting all ordered products of all users, and if the ordered products of the users are in the extracted structured products, reserving the users to form object users; if any product ordered by the user cannot be structured, the user does not belong to the object user.
Extracting all internet surfing logs of each object user, and simultaneously screening out the internet surfing time at T 1 All log-on logs within a time period. Taking a day as a unit, counting the total flow F consumed by each object user in each day by surfing the Internet a . According to the counted daily internet surfing flow of each target user, at T 1 And forming a traffic use sequence of each object user in the time period. Likewise, each subject user is counted for a period of time T 2 The usage flow per day in the day, at T 2 And forming a traffic use sequence of each object user in the time period. At T 1 Traffic usage sequence for each subject user formed during a time period and at T 2 The traffic usage sequence of each target user formed in the time period constitutes a total traffic usage time sequence TS a 。
And S3, training the prediction model by taking the total flow using time sequence as a training data set, and obtaining the trained prediction model.
It can be appreciated that the time series TS is used based on the total traffic a Training the prediction model, wherein the specific training process is as follows:
the counted total flow is divided into training data sets and test data sets by using a time sequence, wherein the number of the training data sets accounts for 90% of the total data sets, and the number of the test data sets accounts for 10% of the total data sets.
Normalizing the time series of the training data set and the test data set to find the maximum value v at a single time point in the time series m The values at all times in the time series are normalized by the following equation.
；
From the above equation, the interval of time series values is normalized to be within the interval [ -1,1 ].
A time sequence prediction deep learning model is constructed, and the model is formed by combining two layers of LSTM networks and one layer of full-connection layer network. The LSTM networks of the two layers sequentially read each value of the time sequence, memorize the values, and simultaneously output the prediction of the LSTM of the two layers on the time sequence; after a layer of full-connection layer network reads the output value of the whole sequence, the single-step prediction of the time sequence is iterated.
Training the deep learning model based on the training data set and the test data set to obtain a trained prediction model. In the deep learning model, root mean square error is used as a loss function of the model, all data in a training data set are input into the deep learning model, a current training set is calculated as a loss value when the data is input, the loss value is fed back to a forward network in a fully connected network, and finally the model is converged into a model for predicting the flow demand of a user.
S4, predicting the total flow of the demand of the test user in the period to be predicted based on the trained prediction model, and obtaining a total flow demand prediction result.
As an embodiment, the training-based predictive model is applied to test usersPredicting the total flow of the demand in a period to be predicted, and obtaining a total flow demand prediction result, wherein the method comprises the following steps: acquiring a test user in a time period T 1 The total online flow used every day in' is input into a trained prediction model, and the time period T output by the prediction model is obtained 2 'within the test user's predicted total online traffic per day.
It can be understood that the prediction model is obtained by training based on the steps, and the use flow of the test user in a future period is predicted based on the prediction model. Specifically, the test user is obtained in a historical time period T 1 The total online flow used every day in' is input into a trained prediction model, and the future time period T output by the prediction model is obtained 2 'within the test user's predicted total online traffic per day.
And S5, matching in the existing structured flow service products according to the ordered flow service products of the test user and the total flow demand prediction result, and recommending the matched flow service products to the test user.
As an embodiment, according to the traffic packet subscribed by the test user and the total traffic demand prediction result, matching is performed in an existing structured traffic service product, to generate a traffic packet combination capable of meeting the traffic demand of the user, including: sequencing a plurality of flow service products according to the service time of the plurality of flow service products ordered by the test user; accessing the sequenced multiple flow service products one by one, and calculating the time period T of the multiple flow service products 2 ' total flow that can be used every day within; the predicted total online traffic and a plurality of traffic service products of each day are processed in a time period T 2 And comparing the total flow which can be used every day in' and determining the recommended flow service product for the test user according to the comparison result.
Wherein the predicted total internet traffic and a plurality of traffic service products of each day are processed in a time period T 2 Comparing the total flow which can be used each day in' and determining the recommended flow service product for the test user according to the comparison result, wherein the recommended flow service product comprisesThe method comprises the following steps: if the predicted total online traffic of any day is less than or equal to the total traffic which can be used by a plurality of traffic service products in a corresponding time period, not recommending the traffic service products to the test user in any day; if the predicted total online traffic of any day is greater than the total traffic which can be used by the traffic service products in the corresponding time period, calculating the traffic difference between the predicted total online traffic of any day and the total traffic which can be used by the traffic service products in the corresponding day, matching the predicted total online traffic with the total traffic which can be used by the traffic service products in the existing structured traffic service products based on the traffic difference, and recommending the matched traffic service products to the test user.
It will be appreciated that for a test user to be recommended, the test user is predicted to be in the future time period T based on the predictive model 2 ' demand flow per day within.
For a test user, the traffic service packages are ordered according to their subscribed traffic service package service times. And accessing the sequenced traffic packets one by one, and calculating the total traffic contained in each day in the traffic packet service time period. And calculating whether the total flow contained in all the flow packets meets the requirement of the test user according to the predicted required flow of the test user in each day and the total flow contained in all the flow packets ordered by the test user in each day.
Specifically, for any day, the predicted required flow rate is compared with the total flow rate contained in the flow rate packet, and the total flow rate required by the user is testedWhether the flow is smaller than the flow contained in the flow packet, if +.>If the flow content is larger than the flow content contained in the flow packet, calculating an excess total flow part as a flow difference value, and storing the flow difference value, wherein the calculation process of the flow difference value can be seen in fig. 2.
It can be understood that the flow difference is the basis of recommended flow service products, and the flow service package recommended to the test user not only contains the flow of the difference, but also requires higher summation ratio. In the embodiment of the invention, the recommended flow service products to the test user are determined based on a decision tree searching mode, wherein the process of matching and recommending the flow service products can be seen in fig. 3.
As an embodiment, based on the traffic difference, matching is performed in an existing structured traffic service product, and generating a traffic packet combination capable of meeting a user traffic demand includes: for each test user, extracting the subscribed flow service products and forming a decision tree of node flow service products, wherein the root node of the decision tree represents the flow service product recommended to the test user first, the cost flow rate of the upper node flow service product is highest, the flow service products represented by the same layer of nodes have the same cost performance, and the flow rate contained in the left node flow service product is greater than the flow rate contained in the right node flow service product; generating a breadth-first path on the decision tree, calculating whether the total flow contained in all the flow service products on the path is greater than the flow difference value or not through a node, if so, ending the search flow, and recommending all the flow service products contained in the path to the test user; if not, continuing searching until the total flow included in the flow service products on the path is greater than the flow difference value, recommending all the products included on the path to the user, and completing the flow service recommending task of the user.
Referring to fig. 4, there is provided a mobile internet traffic service recommendation system, which includes an acquisition module 401, a training module 402, a prediction module 403, and a recommendation module 404, wherein:
an obtaining module 401, configured to determine a time period T based on a total duration of a surfing time of a user surfing log 1 And time period T 2 Acquiring the time period T of each object user 1 Total internet traffic per day and during time period T 2 Total flow of demand in, wherein all subject users are in time period T 1 Total internet traffic per day and during time period T 2 The total flow required in the system forms a total flow use time sequence;
the training module 402 is configured to train the prediction model by using the total flow using time sequence as a training data set, and obtain a trained prediction model;
the prediction module 403 is configured to predict a total flow of a demand of the test user in a period to be predicted based on the trained prediction model, and obtain a total flow demand prediction result;
and the recommending module 404 is configured to match existing structured flow service products according to the flow service products subscribed by the test user and the total flow demand prediction result, and recommend the matched flow service products to the test user.
It can be understood that the mobile internet traffic service recommendation system provided by the present invention corresponds to the mobile internet traffic service recommendation method provided by the foregoing embodiments, and the relevant technical features of the mobile internet traffic service recommendation system may refer to the relevant technical features of the mobile internet traffic service recommendation method, which is not described herein again.
Referring to fig. 5, fig. 5 is a schematic diagram of an embodiment of an electronic device according to an embodiment of the invention. As shown in fig. 5, an embodiment of the present invention provides an electronic device 500, including a memory 510, a processor 520, and a computer program 511 stored in the memory 510 and executable on the processor 520, wherein the processor 520 implements steps of a mobile internet traffic service recommendation method when executing the computer program 511.
Referring to fig. 6, fig. 6 is a schematic diagram of an embodiment of a computer readable storage medium according to the present invention. As shown in fig. 6, the present embodiment provides a computer-readable storage medium 600 on which a computer program 611 is stored, which computer program 611 implements the steps of the mobile internet traffic service recommendation method when executed by a processor.
According to the mobile internet traffic service recommendation method and system provided by the embodiment of the invention, the time sequence of the traffic used by the user is counted, the potential traffic demand of the user is obtained through time sequence prediction, and meanwhile, the traffic demand of the user is matched with the content of the existing traffic package, so that the traffic package recommendation of the user is completed. The invention combines the flow demand prediction and the high cost performance service recommendation of the user, can perform more accurate personalized recommendation service for the user, considers the cost performance problem of purchasing products by the user, improves the satisfaction degree of the user on the products, and improves the viscosity of the user.
In the foregoing embodiments, the descriptions of the embodiments are focused on, and for those portions of one embodiment that are not described in detail, reference may be made to the related descriptions of other embodiments.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (8)
1. A mobile internet traffic service recommendation method, comprising:
determining a time period T based on total duration of surfing time of user surfing log 1 And time period T 2 ；
Acquiring the time period T of each object user 1 Total internet traffic per day and during time period T 2 Total flow of demand in, wherein all subject users are in time period T 1 Total internet traffic per day and during time period T 2 The total flow required in the system forms a total flow use time sequence;
training the prediction model by taking the total flow using time sequence as a training data set to obtain a trained prediction model;
predicting the total flow of the demand of the test user in a time period to be predicted based on the trained prediction model, and obtaining a total flow demand prediction result, wherein the total flow demand prediction result comprises the predicted internet surfing total flow of each day of the user in the time period to be predicted T2';
matching in the existing structured flow service products according to the ordered flow service products of the test user and the total flow demand prediction result, and recommending the matched flow service products to the test user;
according to the traffic packet ordered by the test user and the total traffic demand prediction result, matching is performed in the existing structured traffic service product, and a traffic packet combination capable of meeting the traffic demand of the user is generated, which comprises the following steps:
sequencing a plurality of flow service products according to the service time of the plurality of flow service products ordered by the test user;
accessing the sequenced multiple flow service products one by one, and calculating the time period T of the multiple flow service products 2 ' total flow that can be used every day within;
the predicted total online traffic and a plurality of traffic service products of each day are processed in a time period T 2 Comparing the total flow which can be used in each day in' and determining a recommended flow service product for the test user according to the comparison result;
said combining said predicted total network traffic and a plurality of traffic service products for each day over a period of time T 2 Comparing the total flow which can be used in each day in' and determining a recommended flow service product for the test user according to the comparison result, wherein the method comprises the following steps:
if the predicted total online flow of any day is greater than the total flow which can be used by a plurality of flow service products in a corresponding time period, calculating a flow difference between the predicted total online flow of any day and the total flow which can be used by the plurality of flow service products in the corresponding day, matching in the existing structured flow service products based on the flow difference, and recommending the matched flow service products to the test user;
based on the flow difference, matching is performed in the existing structured flow service product, and a flow packet combination capable of meeting the flow requirement of the user is generated, which comprises the following steps:
for each test user, extracting the subscribed flow service products and forming a decision tree of node flow service products, wherein the root node of the decision tree represents the flow service product recommended to the test user first, the cost flow rate of the upper node flow service product is highest, the flow service products represented by the same layer of nodes have the same cost performance, and the flow rate contained in the left node flow service product is greater than the flow rate contained in the right node flow service product;
generating a breadth-first path on the decision tree, calculating whether the total flow contained in all the flow service products on the path is greater than the flow difference value or not through a node, if so, ending the search flow, and recommending all the flow service products contained in the path to the test user; if not, the search is continued.
2. The mobile internet traffic service recommendation method according to claim 1, wherein the time period T is determined based on a total duration of a user log on the internet 1 And time period T 2 Comprising:
extracting internet surfing logs of all users within a set time period, wherein the internet surfing logs at least comprise user IDs, internet surfing time, access addresses and internet surfing flow;
extracting all user IDs in the Internet surfing log and the total Internet surfing time duration contained in the Internet surfing log, and equally dividing the total Internet surfing time duration into time periods T 1 And time period T 2 。
3. The mobile internet traffic service recommendation method according to claim 1, wherein said acquiring each of the object users is performed during a period T 1 Total internet traffic per day and during time period T 2 The total flow of demand in, include:
extracting all flow service product information of a mobile operator to form structural product characteristics of each flow server product, wherein the flow service product information comprises packages and flow packets;
screening out object users who have ordered flow service products with structural product characteristics from all users;
based on the internet log of each object user, counting each object user in a time period T 1 Total internet traffic F used every day a During a time period T 2 The total online traffic used every day in the system constitutes the total traffic use time sequence of each object user.
4. The mobile internet traffic service recommendation method according to claim 3, wherein training the predictive model using the total traffic usage time sequence as a training data set, and obtaining a trained predictive model, comprises:
dividing the total flow using time sequence into a training data set and a testing data set according to a preset proportion;
normalizing the time series of the training data set and the test data set to find the maximum value v at a single time point in the time series m Normalizing the values at all times in the time series by the following formula, so that the interval of the time series values is normalized to the interval [ -1,1]Inner:
。
5. the mobile internet traffic service recommendation method according to claim 4, wherein said predictive model comprises a two-layer LSTM network and a one-layer fully connected layer network;
the LSTM networks of the two layers sequentially read each data of the total flow use time sequence, memorize each data and output the prediction output of the total flow use time sequence;
the full-connection layer network reads the prediction output of the total flow using time sequence, and iterates to perform single-step prediction of the time sequence;
wherein the loss function of the prediction model is based on root mean square error.
6. The mobile internet traffic service recommendation method according to claim 1, wherein the predicting the total traffic demand of the test user in the period to be predicted based on the trained prediction model, to obtain the total traffic demand prediction result, includes:
acquiring a test user in a time period T 1 The total online flow used every day in' is input into a trained prediction model, and the time period T output by the prediction model is obtained 2 'within the test user's predicted total online traffic per day.
7. The mobile internet traffic service recommendation method according to claim 1, wherein said predicting total internet traffic and a plurality of traffic service products for each day is performed in a time period T 2 Comparing the total flow which can be used in each day in' and determining a recommended flow service product for the test user according to the comparison result, and further comprising:
if the predicted total online traffic of any day is less than or equal to the total traffic which can be used by the plurality of traffic service products in the corresponding time period, the traffic service products are not recommended to the test user in any day.
8. A mobile internet traffic service recommendation system, comprising:
the acquisition module is used for determining a time period T based on the total duration of the Internet surfing time of the Internet surfing log of the user 1 And time period T 2 Acquiring the time period T of each object user 1 Total internet traffic per day and during time period T 2 Total flow of demand in, wherein all subject users are in time period T 1 Total internet traffic per day and during time period T 2 The total flow required in the system forms a total flow use time sequence;
the training module is used for training the prediction model by taking the total flow using time sequence as a training data set to obtain a trained prediction model;
the prediction module is used for predicting the total flow of the requirements of the test user in the time period to be predicted based on the trained prediction model, and obtaining a total flow demand prediction result, wherein the total flow demand prediction result comprises the predicted internet surfing total flow of each day of the user in the time period to be predicted T2';
the recommendation module is used for matching the existing structured flow service products according to the ordered flow service products of the test user and the total flow demand prediction result, and recommending the matched flow service products to the test user;
according to the traffic packet ordered by the test user and the total traffic demand prediction result, matching is performed in the existing structured traffic service product, and a traffic packet combination capable of meeting the traffic demand of the user is generated, which comprises the following steps:
sequencing a plurality of flow service products according to the service time of the plurality of flow service products ordered by the test user;
accessing the sequenced multiple flow service products one by one, and calculating the time period T of the multiple flow service products 2 ' total flow that can be used every day within;
the predicted total online traffic and a plurality of traffic service products of each day are processed in a time period T 2 Comparing the total flow which can be used in each day in' and determining a recommended flow service product for the test user according to the comparison result;
said combining said predicted total network traffic and a plurality of traffic service products for each day over a period of time T 2 Comparing the total flow which can be used in each day in' and determining a recommended flow service product for the test user according to the comparison result, wherein the method comprises the following steps:
if the predicted total online flow of any day is greater than the total flow which can be used by a plurality of flow service products in a corresponding time period, calculating a flow difference between the predicted total online flow of any day and the total flow which can be used by the plurality of flow service products in the corresponding day, matching in the existing structured flow service products based on the flow difference, and recommending the matched flow service products to the test user;
based on the flow difference, matching is performed in the existing structured flow service product, and a flow packet combination capable of meeting the flow requirement of the user is generated, which comprises the following steps:
for each test user, extracting the subscribed flow service products and forming a decision tree of node flow service products, wherein the root node of the decision tree represents the flow service product recommended to the test user first, the cost flow rate of the upper node flow service product is highest, the flow service products represented by the same layer of nodes have the same cost performance, and the flow rate contained in the left node flow service product is greater than the flow rate contained in the right node flow service product;
generating a breadth-first path on the decision tree, calculating whether the total flow contained in all the flow service products on the path is greater than the flow difference value or not through a node, if so, ending the search flow, and recommending all the flow service products contained in the path to the test user; if not, the search is continued.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310384249.6A CN116108291B (en) | 2023-04-12 | 2023-04-12 | Mobile internet traffic service recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310384249.6A CN116108291B (en) | 2023-04-12 | 2023-04-12 | Mobile internet traffic service recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116108291A CN116108291A (en) | 2023-05-12 |
| CN116108291B true CN116108291B (en) | 2023-08-08 |
Family
ID=86254738
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310384249.6A Active CN116108291B (en) | 2023-04-12 | 2023-04-12 | Mobile internet traffic service recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116108291B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107230253A (en) * | 2016-03-24 | 2017-10-03 | 国立民用航空学院 | Object definition in virtual 3d environment |
| CN111478779A (en) * | 2019-09-09 | 2020-07-31 | 中国移动通信集团河北有限公司 | Method and device for determining unlimited set meal and computing equipment |
| CN113015116A (en) * | 2019-12-20 | 2021-06-22 | 中国移动通信集团陕西有限公司 | Dynamic quota method and device based on flow prediction |
| CN113240490A (en) * | 2021-05-18 | 2021-08-10 | 中国人民大学 | Intelligent service transaction recommendation method and system based on graph neural network |
| CN115049464A (en) * | 2022-07-04 | 2022-09-13 | 中国联合网络通信集团有限公司 | Operator package recommendation method and device and computer readable storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106708883B (en) * | 2015-11-17 | 2020-09-29 | 阿里巴巴集团控股有限公司 | Recommendation method and device |
| US11134016B2 (en) * | 2018-10-26 | 2021-09-28 | Hughes Network Systems, Llc | Monitoring a communication network |
-
2023
- 2023-04-12 CN CN202310384249.6A patent/CN116108291B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107230253A (en) * | 2016-03-24 | 2017-10-03 | 国立民用航空学院 | Object definition in virtual 3d environment |
| CN111478779A (en) * | 2019-09-09 | 2020-07-31 | 中国移动通信集团河北有限公司 | Method and device for determining unlimited set meal and computing equipment |
| CN113015116A (en) * | 2019-12-20 | 2021-06-22 | 中国移动通信集团陕西有限公司 | Dynamic quota method and device based on flow prediction |
| CN113240490A (en) * | 2021-05-18 | 2021-08-10 | 中国人民大学 | Intelligent service transaction recommendation method and system based on graph neural network |
| CN115049464A (en) * | 2022-07-04 | 2022-09-13 | 中国联合网络通信集团有限公司 | Operator package recommendation method and device and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 移动通信中数据套餐推荐模型的设计与实现;黎建文等;移动通信(22);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116108291A (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109034660B (en) | Method and related device for determining risk control strategy based on prediction model | |
| US11003564B2 (en) | Methods and systems for determining system capacity | |
| CN108121795B (en) | User behavior prediction method and device | |
| WO2017096877A1 (en) | Recommendation method and device | |
| JP5379135B2 (en) | Information recommendation method and apparatus using composite algorithm | |
| CN107451199A (en) | Method for recommending problem and device, equipment | |
| CN106469376B (en) | Risk control method and equipment | |
| CN104951465A (en) | Application recommendation method and device | |
| CN110417607B (en) | Flow prediction method, device and equipment | |
| US10073886B2 (en) | Search results based on a search history | |
| CN103530365A (en) | Method and system for acquiring downloading link of resources | |
| CN104239552B (en) | Generation association keyword, the method and system that association keyword is provided | |
| CN110096617B (en) | Video classification method and device, electronic equipment and computer-readable storage medium | |
| KR20160113685A (en) | Method, apparatus and system for content recommendation | |
| CN108595526B (en) | Resource recommendation method and device | |
| CN105654198B (en) | Brand advertisement effect optimization method capable of realizing optimal threshold value selection | |
| CN112966865B (en) | Number-carrying network-switching prediction method, device and equipment | |
| CN105022807A (en) | Information recommendation method and apparatus | |
| O'Doherty et al. | Towards trust inference from bipartite social networks | |
| Mehta et al. | Collaborative personalized web recommender system using entropy based similarity measure | |
| CN116108291B (en) | Mobile internet traffic service recommendation method and system | |
| CN107092650B (en) | Weblog analysis method and device | |
| CN117372742A (en) | Domain generalization method, server and client | |
| CN106610991A (en) | Data processing method and device | |
| CN109446432A (en) | A kind of information recommendation method and device |
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 |