CN109325177A - Information-pushing method, system, readable storage medium storing program for executing and computer equipment - Google Patents
Information-pushing method, system, readable storage medium storing program for executing and computer equipment Download PDFInfo
- Publication number
- CN109325177A CN109325177A CN201811036989.6A CN201811036989A CN109325177A CN 109325177 A CN109325177 A CN 109325177A CN 201811036989 A CN201811036989 A CN 201811036989A CN 109325177 A CN109325177 A CN 109325177A
- Authority
- CN
- China
- Prior art keywords
- terminal
- information
- selection region
- base station
- region
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/025—Services making use of location information using location based information parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/12—Messaging; Mailboxes; Announcements
Abstract
The present invention relates to a kind of information-pushing methods, system, readable storage medium storing program for executing and computer equipment, belong to mobile network's technical field, it obtains terminal and is in the stay time in base station signal coverage area, terminal, which is further obtained, in selection region by the incidence relation between base station and selection region stays duration, utilize the mutual corresponding end message of preset model analysis, selection region information and stay duration, to obtain recommending area information and push to terminal, since terminal can also be communicated in not used situation with base station, stay time of the available terminal in base station signal coverage area, pass through terminal, base station, the incidence relation of selection region obtains terminal and stays duration in selection region, according to terminal, area information is recommended in selection region and a length of terminal push of Sojourn times, network without obtaining terminal is visited Behavioral data is asked to determine that the network of terminal accesses Behavior preference, substantially increases the safety of end message data.
Description
Technical field
The present invention relates to mobile network's technical fields, more particularly to a kind of information-pushing method, system, readable storage medium
Matter and computer equipment.
Background technique
With the continuous development of mobile network's technology, more and more electronic commerce informations are answered extensively in a mobile network
With some of electronic commerce informations are used to be pushed to client terminal.
All without specific aim, the mode for generalling use regional broadcast is pushed general information push mode, interior at present
Hold complicated mixed and disorderly, it is difficult to provide practical and effective information for client terminal；There are also another push modes, based on client terminal
Network accesses Behavior preference purposefully to client terminal pushed information, and the information specific aim of this kind push is preferable.
Realize process of the present invention in, inventor find traditional technology in, at least there are the following problems: due to it should be understood that
The network of client terminal accesses Behavior preference, it is necessary to obtain the network access behavioral data of client terminal, the information of client terminal
Data disclosure risk is big, and safety substantially reduces.
Summary of the invention
Based on this, it is necessary to for the low problem of traditional pushed information mode safety, provide a kind of information push side
Method, system, readable storage medium storing program for executing and computer equipment.
A kind of information-pushing method, comprising the following steps:
It obtains terminal and is in the stay time in base station signal coverage area；
It is funny in selection region that terminal is obtained according to the incidence relation between stay time and base station and selection region
Stay duration；
By mutual corresponding end message, selection region information and the preset model of duration input is stayed, obtains preset model
The recommendation area information of output will recommend area information to push to terminal.
According to above- mentioned information method for pushing, obtains terminal and be in the stay time in base station signal coverage area, pass through base
The incidence relation between selection region of standing further obtains terminal and stays duration in selection region, utilizes preset model point
It analyses mutual corresponding end message, selection region information and stays duration, thus obtain recommending area information and push to terminal,
In this scheme, since terminal can also be communicated in not used situation with base station, available terminal is in base station signal
Stay time in coverage area obtains terminal staying in selection region by the incidence relation of terminal, base station, selection region
Duration recommends area information according to terminal, selection region and a length of terminal push of Sojourn times, and the network without obtaining terminal accesses
Behavioral data determines that the network of terminal accesses Behavior preference, substantially increases the safety of end message data, while can be with
It is consistent for terminal push with terminal behavior mode and effectively recommends area information.
In one of the embodiments, when stay time is more than or equal to preset duration, execute according to stay time and
Incidence relation acquisition terminal between base station and selection region is in selection region the step of staying duration.
In one of the embodiments, by mutual corresponding end message, selection region information and to stay duration input pre-
It is further comprising the steps of before the step of setting model:
Regional analysis model is established, multiple groups historical sample information is obtained and regional analysis model is trained, obtain preset
Model, wherein historical sample information includes the Sojourn times of end message, selection region information and terminal in selection region
It is long.
The step of multiple groups historical sample information is trained regional analysis model packet is obtained in one of the embodiments,
Include following steps:
The phase between the corresponding selection region of target terminal and other selection regions is calculated separately by regional analysis model
Like degree, target terminal is obtained to other according to the duration of staying of each similarity and target terminal in corresponding selection region respectively
The interest-degree of selection region is ranked up other selection regions according to each interest-degree, and will sort other preceding selection regions
Information as recommend area information output.
The corresponding selection region of target terminal and its are calculated separately by regional analysis model in one of the embodiments,
The step of similarity between his selection region the following steps are included:
According toCalculate similarity；
In formula, N (i) indicates that terminal quantity associated with selection region i, N (j) indicate associated with selection region j
Terminal quantity, wijIndicate the similarity between selection region i and selection region j.
Duration is stayed in corresponding selection region according to each similarity and target terminal in one of the embodiments,
Obtain the step of target terminal is to the interest-degrees of other selection regions respectively the following steps are included:
According to puj=∑i∈N(u)wijruiCalculate interest-degree；
In formula, wijIndicate the similarity between target terminal u corresponding selection region i and other selection regions j, ruiTable
Show that target terminal u stays duration in corresponding selection region i, N (u) indicates the collection of the corresponding selection region i of target terminal u
It closes, pujIndicate target terminal u to the interest-degree of other selection regions j.
The step of multiple groups historical sample information is trained regional analysis model is obtained in one of the embodiments, also
The following steps are included:
The current selection region for obtaining multiple target terminals, according to each current selection region and each recommendation region zoning
The recall rate of analysis model is adjusted regional analysis model if recall rate is lower than preset value.
In one of the embodiments, by recommend area information push to terminal the step of the following steps are included:
Obtain terminal permanent residence coordinate and it is multiple recommend region position coordinates, calculate separately each position coordinate and often
The distance between guard station coordinate value；
Each recommendation region is ranked up according to each distance value, the information in each recommendation region after sequence is pushed into end
End.
In one of the embodiments, obtain terminal permanent residence coordinate the step of the following steps are included:
Determine the most target BS of the signal connection number of terminal in preset time period, the position for obtaining target BS is sat
It is denoted as the permanent residence coordinate for terminal.
A kind of information transmission system, comprising:
Information acquisition unit, the stay time being in base station signal coverage area for obtaining terminal；
Information process unit, for obtaining terminal according to the incidence relation between stay time and base station and selection region
Duration is stayed in selection region；
Information push unit, for by mutual corresponding end message, selection region information and stay duration input it is preset
Model obtains the recommendation area information of preset model output, area information will be recommended to push to terminal.
According to above- mentioned information supplying system, information acquisition unit obtains terminal and is in the stop in base station signal coverage area
Duration, information process unit further obtain terminal in selection region by the incidence relation between base station and selection region
Duration is stayed, information push unit utilizes the mutual corresponding end message of preset model analysis, selection region information and Sojourn times
It is long, to obtain recommending area information and push to terminal, in this scheme, since terminal also can be in not used situation
Base station is communicated, stay time of the available terminal in base station signal coverage area, passes through terminal, base station, selection area
The incidence relation in domain obtains terminal and stays duration in selection region, is pushed away according to terminal, selection region and a length of terminal of Sojourn times
Recommendation area information is sent, the network without obtaining terminal accesses behavioral data to determine that the network of terminal accesses Behavior preference, greatly
The safety of end message data is improved greatly, while can be consistent with terminal behavior mode and effectively be recommended for terminal push
Area information.
Information process unit is when stay time is more than or equal to preset duration in one of the embodiments, according to stop
Incidence relation between duration and base station and selection region obtains terminal and stays duration in selection region.
Information transmission system further includes model training unit in one of the embodiments, for establishing regional analysis mould
Type obtains multiple groups historical sample information and is trained to regional analysis model, obtains preset model, wherein historical sample information
Duration is stayed in selection region including end message, selection region information and terminal.
It is corresponding by regional analysis model to calculate separately target terminal for model training unit in one of the embodiments,
Similarity between selection region and other selection regions, according to each similarity and target terminal in corresponding selection region
It stays duration and obtains target terminal respectively to the interest-degree of other selection regions, other selection regions are carried out according to each interest-degree
Sequence, using the information for other the preceding selection regions that sort as recommendation area information output.
In one of the embodiments, model training unit according toCalculate similarity；
In formula, N (i) indicates that terminal quantity associated with selection region i, N (j) indicate associated with selection region j
Terminal quantity, wijIndicate the similarity between selection region i and selection region j.
Model training unit is according to p in one of the embodiments,uj=∑i∈N(u)wijruiCalculate interest-degree；
In formula, wijIndicate the similarity between target terminal u corresponding selection region i and other selection regions j, ruiTable
Show that target terminal u stays duration in corresponding selection region i, N (u) indicates the collection of the corresponding selection region i of target terminal u
It closes, pujIndicate target terminal u to the interest-degree of other selection regions j.
Model training unit obtains the current selection region of multiple target terminals in one of the embodiments, according to each
Current selection region and each recall rate for recommending region zoning analysis model, if recall rate is lower than preset value, to region
Analysis model is adjusted.
Information push unit obtains the permanent residence coordinate of terminal and multiple recommends region in one of the embodiments,
Position coordinates calculate separately the distance between each position coordinate and permanent residence coordinate value；According to each distance value to each recommendation region
It is ranked up, the information in each recommendation region after sequence is pushed into terminal.
Information push unit determines that the signal connection number of terminal in preset time period is most in one of the embodiments,
Target BS, obtain permanent residence coordinate of the position coordinates as terminal of target BS.
A kind of readable storage medium storing program for executing, is stored thereon with computer program, which realizes above-mentioned letter when being executed by processor
The step of ceasing method for pushing.
Above-mentioned readable storage medium storing program for executing realizes by the executable program that it is stored and obtains terminal in base station signal covering
Stay time in range, by terminal, base station, selection region incidence relation and stay duration, can for terminal push
Recommend area information, the network without obtaining terminal accesses behavioral data to determine that the network of terminal accesses Behavior preference, significantly
It improves the safety of end message data, while can be consistent for terminal push with terminal behavior mode and effective recommendation
Breath.
A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage
The step of computer program, processor realizes above- mentioned information method for pushing when executing program.
Above-mentioned computer equipment realizes by the executable program run on processor and obtains terminal in base station signal
Stay time in coverage area obtains terminal staying in selection region by the incidence relation of terminal, base station, selection region
Duration recommends area information according to terminal, selection region and a length of terminal push of Sojourn times, and the network without obtaining terminal accesses
Behavioral data determines that the network of terminal accesses Behavior preference, substantially increases the safety of end message data, while can be with
It is consistent for terminal push with terminal behavior mode and effectively recommends area information.
Detailed description of the invention
Fig. 1 is the application scenarios schematic diagram of the information-pushing method in one embodiment；
Fig. 2 is the flow diagram of the information-pushing method in one embodiment；
Fig. 3 is the structural schematic diagram of the information transmission system in one embodiment；
Fig. 4 is the structural schematic diagram of the information transmission system in another embodiment；
Fig. 5 is the positional diagram of the base station and commercial circle in one embodiment；
Fig. 6 is the positional diagram of the base station and commercial circle in one embodiment；
Fig. 7 is the positional diagram of the base station and commercial circle in one embodiment.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention more comprehensible, with reference to the accompanying drawings and embodiments, to this
Invention is described in further detail.It should be appreciated that the specific embodiments described herein are only used to explain the present invention,
And the scope of protection of the present invention is not limited.
Fig. 1 is the application environment schematic diagram of information-pushing method in one embodiment, include in Fig. 1 terminal, base station and after
Platform server,
Terminal can carry out data communication in routine use with base station, and background server can be according to the logical of terminal and base station
Letter state determines stay time that terminal is in base station signal coverage area, and base station is provided at setting for a fixed geographical location
It is standby, selection region be typically also it is fixed, the two has correlation in position, utilizes this correlation available end
Duration is stayed in selection region in end, is analyzed in conjunction with end message and selection region, available recommendation area information
It is sent to terminal；Terminal can be the different types of mobile electronic device such as mobile phone, tablet computer, and information-pushing method can be
It is realized in background server.
It is shown in Figure 2, it is the flow diagram of the information-pushing method of one embodiment.Information in the embodiment pushes away
Delivery method the following steps are included:
Step S110: it obtains terminal and is in the stay time in base station signal coverage area；
Step S120: terminal is obtained in selection area according to the incidence relation between stay time and base station and selection region
Duration is stayed in domain；
In this step, the incidence relation between base station and selection region refers to the positional relationship of base station and chosen area,
When the position of base station and selection region is overlapped, residence time and residence time can be considered as equally；
Step S130: by mutual corresponding end message, selection region information and the preset model of duration input is stayed, is obtained
The recommendation area information of preset model output, will recommend area information to push to terminal.
In this step, the effect of preset model is to mutual corresponding end message, selection region information and Sojourn times
Length is analyzed, and obtains recommending area information, which is consistent with the behavior pattern of terminal.
According to above- mentioned information method for pushing, obtains terminal and be in the stay time in base station signal coverage area, pass through base
The incidence relation between selection region of standing further obtains terminal and stays duration in selection region, utilizes preset model point
It analyses mutual corresponding end message, selection region information and stays duration, thus obtain recommending area information and push to terminal,
In this scheme, since terminal can also be communicated in not used situation with base station, available terminal is in base station signal
Stay time in coverage area obtains terminal staying in selection region by the incidence relation of terminal, base station, selection region
Duration recommends area information according to terminal, selection region and a length of terminal push of Sojourn times, and the network without obtaining terminal accesses
Behavioral data determines that the network of terminal accesses Behavior preference, substantially increases the safety of end message data, while can be with
It is consistent for terminal push with terminal behavior mode and effectively recommends area information.
It should be noted that stay time of the terminal in base station signal coverage area is stop in a period of time
Duration, when reentering the range after terminal leaves base station signal coverage area, stay time restarts to calculate；If terminal
Enter another base station signal coverage area after leaving base station signal coverage area, another base station is not corresponding with selection region
It is identical as stay time of the terminal in previous base station signal coverage area to stay duration for relationship；Another base station and selection
Region also correlation, stay duration be stay time of the terminal in two base station signal coverage areas and value.
In one embodiment, it when stay time is more than or equal to preset duration, executes according to stay time and base station
Incidence relation acquisition terminal between selection region is in selection region the step of staying duration.
In the present embodiment, it can be obtained when stay time is more than or equal to preset duration and stay duration, stay time foot
When enough big, demand of the terminal to corresponding selection region could be indicated, be conducive to improve the validity and standard for recommending area information
True property, but also many long data of invalid Sojourn times can be reduced, improve the treatment effeciency of preset model.
Optionally, preset duration can be set to 30 minutes or more.
In one embodiment, it by mutual corresponding end message, selection region information and stays duration and inputs preset mould
It is further comprising the steps of before the step of type:
Regional analysis model is established, multiple groups historical sample information is obtained and regional analysis model is trained, obtain preset
Model, wherein historical sample information includes the Sojourn times of end message, selection region information and terminal in selection region
It is long.
In the present embodiment, available multiple groups end message, selection region information and terminal are in selection region
Duration is stayed, is trained using regional analysis model of the above-mentioned historical sample information to foundation, keeps regional analysis model continuous
Ground is learnt, and is converted into long to Sojourn times of the terminal in selection region to be analyzed and is obtained recommending the pre- of area information
Set model.
It should be noted that the terminal in historical sample information can have multiple, selection region can also have multiple, stay
Duration is different terminal and stays duration in different selection regions.
In one embodiment, obtain multiple groups historical sample information the step of regional analysis model is trained include with
Lower step:
The phase between the corresponding selection region of target terminal and other selection regions is calculated separately by regional analysis model
Like degree, target terminal is obtained to other according to the duration of staying of each similarity and target terminal in corresponding selection region respectively
The interest-degree of selection region is ranked up other selection regions according to each interest-degree, and will sort other preceding selection regions
Information as recommend area information output.
In the present embodiment, regional analysis model is trained using multiple groups historical sample information as training sample
When, a terminal can be chosen from historical sample information as target terminal, determine the selection region that target terminal stayed,
That is the corresponding selection region of target terminal, the selection region that target terminal did not stayed are other selection regions, utilize target end
Hold corresponding selection region and other selection regions similarity can to target terminal to the interest-degrees of other selection regions into
Row judgement, other the preceding selection regions that can will sort after being ranked up according to interest-degree are exported as recommendation region,
The pushed information for target terminal so can be accurately obtained, which is to be consistent with the behavior pattern of target terminal
's.
In one embodiment, the corresponding selection region of target terminal and other choosings are calculated separately by regional analysis model
The step of selecting the similarity between region the following steps are included:
According toCalculate similarity；
In formula, N (i) indicates that terminal quantity associated with selection region i, N (j) indicate associated with selection region j
Terminal quantity, wijIndicate the similarity between selection region i and selection region j.
In the present embodiment, N (i) indicates that terminal quantity associated with selection region i, N (j) indicate and selection region j
Associated terminal quantity, N (i) ∩ N (j) indicate the terminal quantity stayed in selection region i and selection region j, this
One numerical value is bigger, represents higher, the Jin Erli of a possibility that terminal stayed in selection region i is stayed in selection region i
The similarity between selection region i and selection region j can be relatively accurately obtained with above-mentioned formula.
In one embodiment, duration difference is stayed in corresponding selection region according to each similarity and target terminal
Obtain the step of target terminal is to the interest-degrees of other selection regions the following steps are included:
According to puj=∑i∈N(u)wijruiCalculate interest-degree；
In formula, wijIndicate the similarity between target terminal u corresponding selection region i and other selection regions j, ruiTable
Show that target terminal u stays duration in corresponding selection region i, N (u) indicates the collection of the corresponding selection region i of target terminal u
It closes, pujIndicate target terminal u to the interest-degree of other selection regions j.
In the present embodiment, duration will be stayed using above-mentioned formula to combine with the similarity between selection region, it can be with
Terminal is fast and effeciently obtained to the interest-degree of other selection regions, chooses the selection region of recommendation on this basis.
Optionally, it when calculating interest-degree, can be carried out to the corresponding selection region i of used target terminal u is calculated
It chooses, can such as choose the selection region for being greater than similarity threshold with the similarity of other selection regions j, can be improved with this emerging
The accuracy of interesting degree.
In one embodiment, obtaining the step of multiple groups historical sample information is trained regional analysis model further includes
Following steps:
The current selection region for obtaining multiple target terminals, according to each current selection region and each recommendation region zoning
The recall rate of analysis model is adjusted regional analysis model if recall rate is lower than preset value.
In the present embodiment, during being trained to regional analysis model, obtained recommendation region can be with target
The current selection region of terminal is compared, the recall rate of zoning analysis model, and it is whole that multiple targets can be used in when calculating
It holds corresponding current selection region and recommends region, recall rate can characterize the accuracy of regional analysis model, in recall rate height
When preset value, indicate that regional analysis model can satisfy the accuracy requirement of information push, when recall rate is lower than preset value,
It indicates that the accuracy of regional analysis model is lower, then regional analysis model can be adjusted, recall rate is made to meet condition, protect
Demonstrate,prove the accuracy of the recommendation area information of regional analysis model output.
In one embodiment, by recommend area information push to terminal the step of the following steps are included:
Obtain terminal permanent residence coordinate and it is multiple recommend region position coordinates, calculate separately each position coordinate and often
The distance between guard station coordinate value；
Each recommendation region is ranked up according to each distance value, the information in each recommendation region after sequence is pushed into end
End.
In the present embodiment, moving range of the terminal in daily is limited, generally has permanent residence coordinate, it is contemplated that distance
Distance, the distance between the permanent residence coordinate value of available multiple position coordinates for recommending regions and terminal, according to distance
After value is ranked up each recommendation region, each recommendation region is pushed into terminal, convenient for terminal can rapidly to recommend region into
Row selection.
Optionally, permanent residence may include residence and place of working, and general user is in using terminal in residence and work
Make ground stay time longest, therefore it is available it is multiple recommend region position coordinates respectively with the residence coordinate of terminal or
The distance between place of working coordinate value.
In one embodiment, the step of obtaining the permanent residence coordinate of terminal the following steps are included:
Determine the most target BS of the signal connection number of terminal in preset time period, the position for obtaining target BS is sat
It is denoted as the permanent residence coordinate for terminal.
In the present embodiment, the position coordinates of the target BS of terminal signaling be can use to determine that the permanent residence of terminal is sat
Mark so may not need and carry out complicated accurate positioning to terminal, reduce positioning complexity, due to current base station coverage rate compared with
Greatly, too big error can't be generated using the position coordinates of base station as the permanent residence coordinate of terminal, to each row for recommending region
Sequence will not have an impact.
Optionally, permanent residence can be residence, the base for selecting the signal connection number of terminal between 0 point to 6 points most
The position coordinates stood are as residence coordinate；Permanent residence can be place of working, select the signal connection of terminal between 6 points to 18 points
The position coordinates of the most base station of number are as place of working coordinate.
According to above- mentioned information method for pushing, the embodiment of the present invention also provides a kind of information transmission system, pushes away below with regard to information
The embodiment of system is sent to be described in detail.
It is shown in Figure 3, it is the structural schematic diagram of the commercial circle information transmission system of one embodiment.Quotient in the embodiment
Enclosing information transmission system includes:
Information acquisition unit 210, the stay time being in base station signal coverage area for obtaining terminal；
Information process unit 220, for being obtained according to the incidence relation between stay time and base station and selection region
Terminal stays duration in selection region；
Information push unit 230, for by mutual corresponding end message, selection region information and to stay duration input pre-
Model is set, the recommendation area information of preset model output is obtained, area information will be recommended to push to terminal.
In the present embodiment, information acquisition unit 210 obtains terminal and is in the stay time in base station signal coverage area,
Information process unit 220 further obtains terminal teasing in selection region by the incidence relation between base station and selection region
Duration is stayed, information push unit 230 utilizes the mutual corresponding end message of preset model analysis, selection region information and Sojourn times
It is long, to obtain recommending area information and push to terminal.Since terminal can also be led in not used situation with base station
Letter, stay time of the available terminal in base station signal coverage area are closed by the association of terminal, base station, selection region
System obtains terminal and stays duration in selection region, recommends region according to terminal, selection region and a length of terminal push of Sojourn times
Information, the network without obtaining terminal access behavioral data to determine that the network of terminal accesses Behavior preference, substantially increase end
The safety of client information data, while can be consistent with terminal behavior mode for terminal push and effectively recommend area information.
In one embodiment, information process unit 220 is when stay time is more than or equal to preset duration, when according to stopping
Long and between base station and selection region incidence relation obtains terminal and stays duration in selection region.
In one embodiment, as shown in figure 4, information transmission system further includes model training unit 240, for establishing area
Domain analysis model obtains multiple groups historical sample information and is trained to regional analysis model, obtains preset model, wherein history
Sample information includes that end message, selection region information and terminal stay duration in selection region.
In one embodiment, it is corresponding by regional analysis model to calculate separately target terminal for model training unit 240
Similarity between selection region and other selection regions, according to each similarity and target terminal in corresponding selection region
It stays duration and obtains target terminal respectively to the interest-degree of other selection regions, other selection regions are carried out according to each interest-degree
Sequence, using the information for other the preceding selection regions that sort as recommendation area information output.
In one embodiment, 240 basis of model training unitCalculate similarity；
In formula, N (i) indicates that terminal quantity associated with selection region i, N (j) indicate associated with selection region j
Terminal quantity, wijIndicate the similarity between selection region i and selection region j.
In one embodiment, model training unit 240 is according to puj=∑i∈N(u)wijruiCalculate interest-degree；
In formula, wijIndicate the similarity between target terminal u corresponding selection region i and other selection regions j, ruiTable
Show that target terminal u stays duration in corresponding selection region i, N (u) indicates the collection of the corresponding selection region i of target terminal u
It closes, pujIndicate target terminal u to the interest-degree of other selection regions j.
In one embodiment, model training unit 240 obtains the current selection region of target terminal, according to current selection
Region and recommend region zoning analysis model recall rate, if recall rate be lower than preset value, to regional analysis model into
Row adjustment.
In one embodiment, information push unit 230 obtains the permanent residence coordinate of terminal and multiple recommends region
Position coordinates calculate separately the distance between each position coordinate and permanent residence coordinate value；According to each distance value to each recommendation region
It is ranked up, the information in each recommendation region after sequence is pushed into terminal.
In one embodiment, information push unit 230 determines that the signal connection number of terminal in preset time period is most
Target BS, obtain permanent residence coordinate of the position coordinates as terminal of target BS.
The information transmission system and above- mentioned information method for pushing of the embodiment of the present invention correspond, in above- mentioned information push side
The technical characteristic and its advantages that the embodiment of method illustrates are suitable for the embodiment of information transmission system.
According to above- mentioned information method for pushing, the embodiment of the present invention also provides a kind of readable storage medium storing program for executing and a kind of computer is set
It is standby.
A kind of readable storage medium storing program for executing, is stored thereon with computer program, which realizes above-mentioned letter when being executed by processor
The step of ceasing method for pushing.
Above-mentioned readable storage medium storing program for executing realizes by the executable program that it is stored and obtains terminal in base station signal covering
Stay time in range, by terminal, base station, selection region incidence relation and stay duration, can for terminal push
Recommend area information, the network without obtaining terminal accesses behavioral data to determine that the network of terminal accesses Behavior preference, significantly
It improves the safety of end message data, while can be consistent for terminal push with terminal behavior mode and effective recommendation
Breath.
A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage
The step of computer program, processor realizes above- mentioned information method for pushing when executing program.
Above-mentioned computer equipment realizes by the executable program run on processor and obtains terminal in base station signal
Stay time in coverage area, by terminal, base station, selection region incidence relation and stay duration, can be terminal
Area information is recommended in push, determines that the network of terminal accesses Behavior preference without obtaining the network access behavioral data of terminal,
The safety of end message data is substantially increased, while can be consistent with terminal behavior mode and effectively push away for terminal push
Recommend information.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, program can be stored in a non-volatile computer-readable storage
In medium, in embodiment, which be can be stored in the storage medium of computer system, and by the computer system extremely
A few processor executes, and includes the process such as the embodiment of above-mentioned information-pushing method with realization.Wherein, storage medium can be
Magnetic disk, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random Access
Memory, RAM) etc..
In one embodiment, information-pushing method can be applied in the scene of commercial circle information push.In concrete application
When, the overlay area of commercial circle is selection region, and recommending area information is to recommend commercial circle information, and acquisition terminal stays the history of commercial circle
Footprint, based on place context article Collaborative Filtering Recommendation Algorithm (Item Collaboration Filter, is called in the following text:
ItemCF), the commercial circle for recommending preference for terminal meets the consumption demand of terminal user, while latent for commercial circle introducing most consumption
The user of power.
Detailed process is as follows:
One: acquisition terminal stays the history footprint of commercial circle
1) corresponding relationship of base station and commercial circle is established
The covering radius of commercial circle is usually 2-3 kilometers, and the covering radius of base station is usually 5-10 kilometers in city, base station and
The case where there are overlay area intersections for commercial circle, then there are corresponding relationship in base station and commercial circle.
Assuming that city shares m commercial circle, the coordinate of each commercial circle is Bi(x,y)(1≤i≤m)；N base station, each base station
Coordinate be Cj(x, y) (1≤j≤n), then there are three kinds of relationships between each commercial circle and each base station:
As shown in figure 5, base station covers commercial circle；As shown in fig. 6, base station and commercial circle intersection；As shown in fig. 7, base station and commercial circle
There is no intersection；
If base station and commercial circle have covering and converging relation, the two has corresponding relationship.It can be between commercial circle and base station
It is the relationship of multi-to-multi, i.e., a commercial circle might have multiple base stations and cover, and a base station may also have friendship with multiple commercial circles
Collection.
2) judge the residence and place of working of terminal
This is important information of the ItemCF based on place context.Judge that the residence of terminal is very direct, selects 0. -6 point
Between terminal registration base station in there is the most base station of signal connection number.Judge that the place of working principle of terminal is the same, selects
Occurs the most base station of signal connection number in the base station of terminal registration between 6. -18 points.
During this judgement is user's sleep or work based on terminal, base station and terminal meeting periodic communication, if one
User stays in a place for a long time, then the base station in this place be register user terminal it is most.
Assuming that the coordinate of base station belonging to the residence of mobile phone user is Ch(x, y), the coordinate of base station belonging to place of working
It is Cw(x,y)。
3) the commercial circle record that acquisition terminal history is stayed
Assuming that terminal (non-residence and place of working) near certain base station is stopped, more than 30 points, (this threshold value can be according to not
With the definition of theme commercial circle, the time in general staying commercial circle all can be more than 30 points substantially), and base station and commercial circle just have pair
It should be related to, i.e.,It then may determine that terminal is stayed and consumed in the commercial circle.
If terminal leaves the commercial circle, i.e., the base station that base station is shifted, and registered be not covered with commercial circle or
Converging relation then records terminal and leaves the commercial circle time, and can calculate user and stay duration in the commercial circle.
Acquisition, the processing for completing data, eventually generate the data such as 1 sample set of table, are used for ItemCF model training.
1 sample set of table
Two: ItemCF model training and assessment
Terminal is recommended using commercial circle as a kind of article in this programme.ItemCF algorithm to terminal user recommend and it is clear before
The similar article of the article look at, the behavior record that it mainly passes through analysing terminal calculate the similarity between article, that is to say, that
Article A and article B there is very big similarity to be because the terminal user for liking article A mostly also likes article B；Meanwhile
The history footprint that ItemCF also can use terminal is provided to recommendation results to be recommended to explain, the article for explanation is all that terminal is used
The article browsed before family.
1) similarity between commercial circle is calculated
Formula definition:Wherein N (i) indicates to like terminal user's number of commercial circle i, and N (j) is indicated
The terminal user's number for liking commercial circle j, the record quantity of the terminal number exactly stayed in commercial circle in sample set.
By the formula, the similarity between its all commercial circle, and building one commercial circle-commercial circle similarity moment can establish
Battle array, value is exactly wij。
2) interest of the computing terminal user to a commercial circle j:
Formula definition: puj=∑i∈N(u)∩S(j,k)wijrui, wherein N (u) is the set for the commercial circle that terminal u was stayed, S (j,
It k) is set with the K commercial circle of commercial circle j most like (i.e. similarity be greater than similarity threshold), wijExactly between commercial circle i and j
Similarity, ruiIt is that (be reflected in is exactly to stay duration in sample set to interest of the user of terminal u to commercial circle i, stays longer interest more
It is high).
Formula explanation and terminal user more similar commercial circle, interested commercial circle in history, more it is possible that pushing away in terminal
It recommends and obtains higher ranking in list, the preceding commercial circle of ranking can recommend terminal, and terminal user is recommended to go to the commercial circle recommended
Consumption.
2) model evaluation
ItemCF model is assessed with recall rate, test set comes from real data, stays going through for commercial circle according to acquisition terminal
The mode of history footprint acquires real data.
Recall rate definition:Wherein R (u) indicates the N that ItemCF model is recommended to terminal u
A commercial circle, T (u) indicate that user stayed the set of commercial circle on test set, and recall rate is higher, illustrates that model is more accurate, is recalling
When rate is lower than preset value, the commercial circle for indicating that ItemCF model is recommended is not accurate enough, needs to be adjusted ItemCF model.
Three: recommending ranking and filtering in commercial circle
ItemCF based on sample set training does not account for place context when providing and recommending commercial circle list.It is pushing away
It recommends in filtering, needs that the commercial circle list recommended is reconstructed, resequence according to distance.
During nonworkdays, with the distance between the commercial circle coordinate of recommendation and residence coordinate for sort by, i.e., | Bi
(x,y)-Ch(x,y)|；
Between date, with the distance between the commercial circle coordinate of recommendation and place of working coordinate for sort by, i.e., | Bi(x,
y)-Cw(x,y)|。
After considering place context, recommendation is provided according to commercial circle list ranking.
When recommending, the label or theme of commercial circle can also be defined, as major class can define amusement, shopping, diet, automobile
Deng, group can also be defined under major class, then apply the proposed algorithm based on label, can be associated with terminal user content and
Interest preference.
According to above-mentioned commercial circle information recommendation mode, example case is as follows: Xiao Ming lives the continent Pa, and work is hunted moral in the Milky Way, gone through
There are Baima garments city, Huambo center, the square Zheng Jia etc. in the commercial circle stayed in history, by ItemCF and combines place or more
Text recommends the commercial circles such as international Textile City, golden hippocampus household, racecourse to it.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
Those of ordinary skill in the art will appreciate that implement the method for the above embodiments be can be with
Relevant hardware is instructed to complete by program.The program can store in read/write memory medium.The program exists
When execution, include the steps that described in the above method.The storage medium, comprising: ROM/RAM, magnetic disk, CD etc..
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (12)
1. a kind of information-pushing method, which comprises the following steps:
It obtains terminal and is in the stay time in base station signal coverage area；
The terminal is obtained in the choosing according to the incidence relation between the stay time and the base station and selection region
It selects in region and stays duration；
Mutual corresponding end message, selection region information and the duration of staying are inputted into preset model, obtained described preset
The recommendation area information of model output, pushes to the terminal for the recommendation area information.
2. information-pushing method according to claim 1, which is characterized in that when the stay time is more than or equal to default
When long, execute the incidence relation according between the stay time and the base station and selection region and obtain the terminal
In selection region the step of staying duration.
3. information-pushing method according to claim 1, which is characterized in that described by mutual corresponding end message, choosing
Select area information and it is described stay duration input preset model the step of before, it is further comprising the steps of:
Regional analysis model is established, multiple groups historical sample information is obtained and the regional analysis model is trained, described in acquisition
Preset model, wherein the historical sample information includes end message, selection region information and terminal in selection region
Stay duration.
4. information-pushing method according to claim 3, which is characterized in that the acquisition multiple groups historical sample information is to institute
State the step of regional analysis model is trained the following steps are included:
The phase between the corresponding selection region of target terminal and other selection regions is calculated separately by the regional analysis model
Like degree, the mesh is obtained according to the duration of staying of each similarity and the target terminal in corresponding selection region respectively
Terminal is marked to the interest-degree of other selection regions, other selection regions are ranked up according to each interest-degree, sequence is existed
The information of other preceding selection regions is as recommendation area information output.
5. information-pushing method according to claim 4, which is characterized in that described to be distinguished by the regional analysis model
The step of calculating the similarity between the corresponding selection region of target terminal and other selection regions the following steps are included:
According toCalculate the similarity；
In formula, N (i) indicates that terminal quantity associated with selection region i, N (j) indicate terminal associated with selection region j
Quantity, wijIndicate the similarity between selection region i and selection region j.
6. information-pushing method according to claim 4, which is characterized in that described according to each similarity and the mesh
It marks stay duration of the terminal in corresponding selection region and obtains the target terminal respectively to the interest-degree of other selection regions
The step of the following steps are included:
According to puj=∑i∈N(u)wijruiCalculate the interest-degree；
In formula, wijIndicate the similarity between target terminal u corresponding selection region i and other selection regions j, ruiIndicate target
Terminal u stays duration in corresponding selection region i, and N (u) indicates the set of the corresponding selection region i of target terminal u, puj
Indicate target terminal u to the interest-degree of other selection regions j.
7. information-pushing method according to claim 4, which is characterized in that the acquisition multiple groups historical sample information is to institute
It is further comprising the steps of to state the step of regional analysis model is trained:
The current selection region for obtaining multiple target terminals, according to each current selection region and each recommendation region
The recall rate of the regional analysis model is calculated, if the recall rate is lower than preset value, the regional analysis model is carried out
Adjustment.
8. information-pushing method as claimed in any of claims 1 to 7, which is characterized in that described by the recommendation
Area information push to the step of terminal the following steps are included:
The permanent residence coordinate and multiple position coordinates for recommending region for obtaining the terminal, calculate separately each position
The distance between coordinate and the permanent residence coordinate value；
Each recommendation region is ranked up according to each distance value, the information in each recommendation region after sequence is pushed to
The terminal.
9. information-pushing method according to claim 8, which is characterized in that the permanent residence coordinate for obtaining the terminal
The step of the following steps are included:
It determines the most target BS of the signal connection number of the terminal in preset time period, obtains the position of the target BS
Set permanent residence coordinate of the coordinate as the terminal.
10. a kind of information transmission system characterized by comprising
Information acquisition unit, the stay time being in base station signal coverage area for obtaining terminal；
Information process unit, for being obtained according to the incidence relation between the stay time and the base station and selection region
The terminal stays duration in the selection region；
Information push unit, for by mutual corresponding end message, selection region information and it is described stay duration input it is preset
Model obtains the recommendation area information of the preset model output, the recommendation area information is pushed to the terminal.
11. a kind of readable storage medium storing program for executing, is stored thereon with executable program, which is characterized in that when the program is executed by processor
The step of realizing information-pushing method described in any one of claim 1 to 9.
12. a kind of computer equipment, which is characterized in that including memory, processor and store on a memory and can handle
The executable program run on device, processor realize information push described in any one of claim 1 to 9 when executing program
The step of method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811036989.6A CN109325177B (en) | 2018-09-06 | 2018-09-06 | Information pushing method, system, readable storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811036989.6A CN109325177B (en) | 2018-09-06 | 2018-09-06 | Information pushing method, system, readable storage medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109325177A true CN109325177A (en) | 2019-02-12 |
| CN109325177B CN109325177B (en) | 2021-05-14 |
Family
ID=65263737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811036989.6A Active CN109325177B (en) | 2018-09-06 | 2018-09-06 | Information pushing method, system, readable storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109325177B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111757265A (en) * | 2020-06-22 | 2020-10-09 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for pushing playing content |
| CN112468964A (en) * | 2020-11-30 | 2021-03-09 | 上海欣方智能系统有限公司 | Information release system, method and device |
| CN113254811A (en) * | 2021-06-23 | 2021-08-13 | 佛山市墨纳森智能科技有限公司 | Interest place recommendation method, system and storage medium based on big data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1661604A (en) * | 2004-02-25 | 2005-08-31 | 松下电器产业株式会社 | Active recording analysis of mobile terminal and auto information recommendation system and method thereof |
| CN104778231A (en) * | 2015-03-31 | 2015-07-15 | 北京奇艺世纪科技有限公司 | Feature identification method and device for geographic areas |
| CN107341693A (en) * | 2017-07-06 | 2017-11-10 | 中国联合网络通信集团有限公司 | The method of user's screening of travelling and the system of tourism user's screening |
-
2018
- 2018-09-06 CN CN201811036989.6A patent/CN109325177B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1661604A (en) * | 2004-02-25 | 2005-08-31 | 松下电器产业株式会社 | Active recording analysis of mobile terminal and auto information recommendation system and method thereof |
| CN104778231A (en) * | 2015-03-31 | 2015-07-15 | 北京奇艺世纪科技有限公司 | Feature identification method and device for geographic areas |
| CN107341693A (en) * | 2017-07-06 | 2017-11-10 | 中国联合网络通信集团有限公司 | The method of user's screening of travelling and the system of tourism user's screening |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111757265A (en) * | 2020-06-22 | 2020-10-09 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for pushing playing content |
| CN111757265B (en) * | 2020-06-22 | 2023-03-10 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for pushing playing content |
| CN112468964A (en) * | 2020-11-30 | 2021-03-09 | 上海欣方智能系统有限公司 | Information release system, method and device |
| CN112468964B (en) * | 2020-11-30 | 2023-04-07 | 上海欣方智能系统有限公司 | Information release system, method and device |
| CN113254811A (en) * | 2021-06-23 | 2021-08-13 | 佛山市墨纳森智能科技有限公司 | Interest place recommendation method, system and storage medium based on big data |
| CN113254811B (en) * | 2021-06-23 | 2021-09-24 | 佛山市墨纳森智能科技有限公司 | Interest place recommendation method, system and storage medium based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109325177B (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106303955B (en) | For carrying out matched method and apparatus to hotspot and POI | |
| CN104573042B (en) | The determination method and device of POI | |
| EP2592885B1 (en) | System and method for acquiring statistics of navigation information | |
| CN107291888B (en) | Machine learning statistical model-based living recommendation system method near living hotel | |
| Elhamshary et al. | SemSense: Automatic construction of semantic indoor floorplans | |
| CN107436893A (en) | A kind of webpage recommending method and device | |
| CN108734502B (en) | Data statistics method and system based on user position | |
| US20130205196A1 (en) | Location-based mobile application marketplace system | |
| CN105184612B (en) | A kind of route recommendation method and user terminal | |
| CN109325177A (en) | Information-pushing method, system, readable storage medium storing program for executing and computer equipment | |
| CN109672715A (en) | User's permanent residence judgment method, device, equipment and computer readable storage medium | |
| CN105808637A (en) | Personalized recommendation method and device | |
| CN104636382A (en) | Social relation reasoning method and device | |
| CN108810806B (en) | Positioning information correction method, client, storage medium and equipment | |
| CN108471455A (en) | Method for pushing, server and the storage medium of house property information | |
| CN105335515A (en) | Information recommendation method and information recommendation device | |
| CN107025251A (en) | A kind of data push method and device | |
| CN111954175B (en) | Method for judging visiting of interest point and related device | |
| CN105844502A (en) | Information release method, device and system based on cloud and positioning technology | |
| CN110545317A (en) | Grid-perception-based power-assisted region division small service method and device | |
| CN109068272A (en) | Similar users recognition methods, device, equipment and readable storage medium storing program for executing | |
| CN107038649A (en) | A kind of friend recommendation method and device of terminal user | |
| CN108734501B (en) | Mobile positioning platform | |
| CN112241489A (en) | Information pushing method and device, readable storage medium and computer equipment | |
| CN107729528A (en) | A kind of building information acquisition methods and device, computer installation and computer-readable recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |