CN109460509A - User interest point appraisal procedure, device, computer equipment and storage medium - Google Patents
User interest point appraisal procedure, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109460509A CN109460509A CN201811187932.6A CN201811187932A CN109460509A CN 109460509 A CN109460509 A CN 109460509A CN 201811187932 A CN201811187932 A CN 201811187932A CN 109460509 A CN109460509 A CN 109460509A
- Authority
- CN
- China
- Prior art keywords
- point
- interest
- user
- position data
- data
- 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.)
- Pending
Links
Abstract
The present invention relates to data analysis field, disclosing a kind of user interest point appraisal procedure, device, computer equipment and storage medium, method includes: to obtain the position data of user；Target area range is determined according to the position data；Obtain preset point of interest discrimination model associated with the target area range；It obtains and the preset associated processing rule of point of interest discrimination model；According to position data described in the processing rule process, position input data is generated；The position input data is inputted into the preset point of interest discrimination model, the point of interest for obtaining the position data differentiates as a result, and differentiating that result determines the point of interest of user according to point of interest.The present invention is handled the position data of user by preset point of interest discrimination model, obtain the interest point data of user, the point of interest of user can be obtained based on sparse location data, judges the geographical location preference of user in the case where not increasing station acquisition equipment investment.
Description
Technical field
The present invention relates to data analysis field more particularly to a kind of user interest point appraisal procedures, device, computer equipment
And storage medium.
Background technique
Point of interest is derived from English POI (abbreviation of Point of Interest).In GIS-Geographic Information System, an interest
Point can be a house, a retail shop, a mailbox, a bus station etc..Existing POI is typically based on the base of high density
It inputs and obtains in location-based service (English is represented by LBS) data.
The point of interest for obtaining user can provide more high-quality, accurately service information pushing for user.However, for one
The application program being of little use a bit, what is obtained is less based on location-based service data, and the point of interest difficulty for obtaining user is larger.
Therefore, how based on application program obtain it is less amount of based on location-based service data calculate user point of interest at
The problem of for those skilled in the art's urgent need to resolve.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of user interest point appraisal procedure, device, computer
Equipment and storage medium are based on location-based service data to solve the existing small amount that can not be obtained based on application, calculate user's
The problem of point of interest.
A kind of user interest point appraisal procedure, comprising:
Obtain the position data of user；
Target area range is determined according to the position data；
Obtain preset point of interest discrimination model associated with the target area range；
It obtains and the preset associated processing rule of point of interest discrimination model；
According to position data described in the processing rule process, position input data is generated；
The position input data is inputted into the preset point of interest discrimination model, obtains the interest of the position data
Point differentiates as a result, and differentiating that result determines the point of interest of user according to point of interest.
A kind of user interest point assessment device, comprising:
Position data module is obtained, for obtaining the position data of user；
Regional scope module is determined, for determining target area range according to the position data；
Discrimination model module is obtained, differentiates mould for obtaining preset point of interest associated with the target area range
Type；
Processing rule module is obtained, for obtaining and the preset associated processing rule of point of interest discrimination model；
Input data module is generated, the position data according to the processing rule process is used for, position is generated and inputs number
According to；
It obtains and differentiates object module, for the position input data to be inputted the preset point of interest discrimination model,
The point of interest for obtaining the position data differentiates as a result, and differentiating that result determines the point of interest of user according to point of interest.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize such as above-mentioned user interest point assessment side when executing the computer program
The step of method.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
It realizes when calculation machine program is executed by processor such as the step of above-mentioned user interest point appraisal procedure.
Above-mentioned user interest point appraisal procedure, device, computer equipment and storage medium, are differentiated by preset point of interest
Model handles the position data of user, obtains the interest point data of user, can not increase station acquisition equipment investment
In the case where, the point of interest of user is obtained based on sparse location data, judges the geographical location preference of user.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is an application environment schematic diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 2 is a flow diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 3 is a flow diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 4 is a flow diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 5 is a flow diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 6 is a flow diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 7 is a flow diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 8 is a flow diagram of user interest point appraisal procedure in one embodiment of the invention；
Fig. 9 is a structural schematic diagram of user interest point assessment device in one embodiment of the invention；
Figure 10 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
User interest point appraisal procedure provided by the present application, can be applicable in the application environment such as Fig. 1, wherein client
It is communicated by network with server-side.Wherein, client can be, but not limited to various personal computers, laptop, intelligence
It can mobile phone, tablet computer and portable wearable device.Server-side can use independent server either multiple server groups
At server cluster realize.
In one embodiment, it as shown in Fig. 2, providing a kind of user interest point appraisal procedure, applies in Fig. 1 in this way
Server-side for be illustrated, include the following steps:
S10, the position data for obtaining user.
In the present embodiment, server-side can obtain the position data of user by the application pre-installed in client.Prepackage
The position data is obtained using the authorization based on user.Position data includes but is not limited to the time obtained and user position
It sets.User location can be indicated with specific longitude and latitude.
S20, target area range is determined according to the position data.
The position data that server-side obtains will not be usually completely coincident with the practical site of user, and there are a certain distance
Deviation.For example, user's first is in store B at the A moment, but position data shows to be that user's first is at 10 meters of store B
Position, the position that position data the is shown positional distance actually located with user's first differ 10 meters.It can be obtained according to statistics
Position data and the mean longitudinal deviation of user practical site determine target area range.It can be with position data meaning
To position distance to a declared goal within the scope of region be target area range.Distance to a declared goal range can be mean longitudinal deviation
Specified multiple.Specified multiple can be 1-3 times.For example, mean longitudinal deviation is 15m, position pointed by position data is C meal
Shop, then target area range can be centered on the restaurant C, and radius is the regional scope of 30m.
S30, preset point of interest discrimination model associated with the target area range is obtained.
Server-side is stored with multiple preset point of interest discrimination models.Preset point of interest discrimination model is with geographical entity
Location is corresponding.Geographical entity address corresponding with preset point of interest discrimination model can also be indicated with target point of interest.Example
Such as, preset point of interest discrimination model can be car repair shop discrimination model, corresponding with entity car repair shop；It can also be with
It is chafing dish restaurant discrimination model, it is corresponding with entity chafing dish restaurant.
Preset point of interest discrimination model is for differentiating whether point of interest corresponding to position data is preset point of interest
Geographical entity address corresponding to discrimination model.Preset point of interest discrimination model is by moving to point of interest discrimination model
It moves learning training and constructs.Point of interest discrimination model is by the sample with point of interest label by training study building
Discrimination model.
If geographical entity address corresponding to preset point of interest discrimination model is within the scope of target area, determining should
Preset point of interest discrimination model is associated with target area range.Preset point of interest within the scope of target area differentiates
Model can be one or more.
S40, it obtains and the preset associated processing rule of point of interest discrimination model.
In the present embodiment, processing rule is acceptable defeated for position data to be processed into preset point of interest discrimination model
Enter form.Preset point of interest discrimination model is different, and associated processing rule can be identical, is also possible to difference.For example,
The chain store E of brand D corresponds to chain store's E discrimination model, is all that the chain store F of brand D corresponds to chain store's F discrimination model, chain
Processing rule associated by the E discrimination model of shop is identical as processing rule associated by chain store's F discrimination model；G pairs of seafood restaurant
Seafood restaurant G discrimination model is answered, associated by processing rule associated by seafood restaurant G discrimination model and chain store E discrimination model
Processing rule it is different.
S50, the position data according to the processing rule process generate position input data.
In the present embodiment, position input data is after the associated processing rule process of preset point of interest discrimination model
Position data.Obtain position input data after, preset point of interest discrimination model can be to the position input data at
Reason, obtains the differentiation result of position data.
S60, the position input data is inputted into the preset point of interest discrimination model, obtains the position data
Point of interest differentiates as a result, and differentiating that result determines the point of interest of user according to point of interest.
In the present embodiment, position input data can calculate corresponding after inputting preset point of interest discrimination model
Discriminant value.Compared with differentiating discrimination threshold corresponding with the preset point of interest discrimination model, if discriminant value is less than differentiation threshold
Value, then illustrate that the geographical entity address corresponding to the preset point of interest discrimination model user did not occurred, then positional number
According to point of interest differentiate result are as follows: geographical entity address corresponding to the preset point of interest discrimination model is not the interest of user
Point；If discriminant value is greater than or equal to discrimination threshold, illustrate user's geography corresponding to the preset point of interest discrimination model
Physical address occurred, then the point of interest of position data differentiates result are as follows: ground corresponding to the preset point of interest discrimination model
Manage the point of interest that physical address is user.Such as, discrimination threshold 0.5, calculated discriminant value are 0.36, which is less than
0.5, then illustrate that geographical entity address corresponding to the preset point of interest discrimination model is not the point of interest of user.
If preset point of interest discrimination model associated with target area range there are multiple, is likely to occur same
Position data corresponds to the case where multiple points of interest.At this point it is possible to by the way of ballot method to point of interest differentiate result carry out into
The processing of one step, determines the point of interest of user.
In step S10-S60, the position data of user is obtained, to obtain the initial data for judging user interest point.
Target area range is determined according to the position data, to reduce the determination range of user interest point.It obtains and the target area
The associated preset point of interest discrimination model of domain range, to obtain the preset point of interest differentiation mould for handling position data
Type.It obtains and the associated processing of preset point of interest discrimination model is regular, to obtain for adjusting position data format
Processing rule.According to position data described in the processing rule process, position input data is generated, position data is adjusted to
The preset accessible format of point of interest discrimination model.The position input data input preset point of interest is differentiated into mould
Type, the point of interest for obtaining the position data differentiate as a result, and differentiate that result determines the point of interest of user according to point of interest, to obtain
The point of interest for obtaining position data differentiates as a result, determining the point of interest of user.
Optionally, referring to figure 3., step S10 includes:
S101, the first position data for obtaining user in the first specified time；
S102, judge whether the size of the first position data is less than preset threshold；
If the size of S103, the first position data is not less than preset threshold, the first position data are determined
For the position data；
If the size of S104, the first position data is less than preset threshold, user in the second specified time is obtained
The second position data are determined as the position data by second position data, and second specified time is greater than described the
One specified time.
Specifically, server-side can also obtain the position data of user in specified time.First specified time can be three
A month, half a year, 1 year or other setting times.Second specified time was greater than at the first time.It in one embodiment, can be according to adopting
The size of data of the position data collected determines the specified period.Under normal conditions, time closer position data, value are got over
It is high；Time more long position data is worth lower.When server-side can get enough positional numbers within nearlyr a period of time
According to then the position data newer using the time uses in the longer time if the position data in nearlyr a period of time is insufficient
Position data.Specifically, first obtaining multiple position datas of the first specified time, preset if the number of the position data is less than
Point of interest number threshold value then obtains the position data of the second specified time, and it is specified that second specified time is greater than described first
Time.For example, default point of interest number threshold value is that 5Mb is obtained if position data number of user's first in 3 months is less than 5Mb
Take position data of the family first in 6 months.
In step S101-S104, the first position data of user in the first specified time are obtained, to obtain the newest of user
Position data.Judge whether the size of the first position data is less than preset threshold；If the size of the first position data
Not less than preset threshold, then the first position data are determined as the position data, if the big Grain Full of first position data
Foot requires, then can be using first position data as the initial data for calculating user interest point.If the first position data
Size is less than preset threshold, then obtains the second position data of user in the second specified time, and the second position data are true
It is set to the position data, second specified time is greater than first specified time；In first position, the size of data is not
When meeting the requirements, the position data in longer a period of time is obtained as the initial data for calculating user interest point.
Optionally, as shown in figure 4, step S20 includes:
S201, judge whether the position of the position data is within the scope of predeterminable area；
If S202, being within the scope of predeterminable area, the predeterminable area range is determined as target area range.
In the present embodiment, the distribution situation for the preset point of interest discrimination model that can be stored according to server-side marks off more
A predeterminable area range.In some commercial circle regions, it includes point of interest it is more, the distribution of preset point of interest discrimination model compared with
To be intensive, the boundary of predeterminable area range is closely packed together, then can also be the preset areas adjacent with the position of position data
Domain range is classified as target area range.That is, target area range may include more than one predeterminable area range.Example
Such as, the position H of position data is in predeterminable area range I, but H is in the near border of I, adjacent with the boundary of I
Predeterminable area range includes predeterminable area range J and predeterminable area range K, then target area range may include predeterminable area model
Enclose I, predeterminable area range J and predeterminable area range K.
In step S201-S202, judge whether the position of the position data is within the scope of predeterminable area, with choosing
Select suitable predeterminable area range.If being in predeterminable area range, the predeterminable area range is determined as target area range,
With the determination range of contracted position data, calculation amount is reduced.
Optionally, include: referring to Fig. 5, step S50
S501, the time point for first appearing target point of interest is determined, the preset point of interest discrimination model includes described
Target point of interest；
S502, the position input data, the setup time section packet are generated based on the position data in setup time section
Include start time and end time, the start time include the time for first appearing target point of interest before it is first specified when
Between point, the end time includes after the time for first appearing target point of interest in rear specified time point.
In the present embodiment, the corresponding target point of interest of preset point of interest discrimination model.In order to make the positional number of user
It preferably matches, can be for further processing to the position data of user according to preset point of interest discrimination model.Specifically, first
It determines the time first appeared where target point of interest, is then matched according to the time point determination where first appearing target point of interest
Set the period.Wherein, first appearing target point of interest can be determined in the following manner: first calculate the position data of user
Real time position corresponding to middle various time points filters out real time position at a distance from target point of interest at a distance from target point of interest
Less than the time point of default screening distance, that time point of last access time earliest.For example, can be set default screening away from
From value be 5m, in the position data of a user, real time position corresponding to various time points is at a distance from target point of interest
Time point less than 5m includes 9:50,9:52,9:54,9:56,9:58,10:02, then the time for first appearing target point of interest is
9:50。
Corresponding setup time section can be determined based on the time point for first appearing target point of interest.Setup time section is used for
Interception and the stronger position data of target point of interest correlation.It is believed that before to first appear the time of target point of interest
First specified time be start time, after the time to first appear target point of interest is that the time is whole in rear specified time
The position data that the setup time section of point is included is for judging whether target point of interest is that the real point of interest of user is very heavy
It wants.Here, first specified time point and rear specified time point can according to specific preset point of interest discrimination model into
Row setting, can be 5min, 10min, 20min, 30min or other specified times.First specified time point can in rear finger
It fixes time a little equal, it can also be unequal.Such as, the time point where first appearing target point of interest is 10:50, when formerly specified
Between point be 30min, rear specified time point be 30min, then setup time section be 10:20-11:20.
In step S501-S502, the time point for first appearing target point of interest is determined, the preset point of interest differentiates mould
Type includes the target point of interest, to determine that user arrives at the time of target point of interest.Based on the positional number in setup time section
According to generating the position input data, the start time of the setup time section include first appear target point of interest time it
Preceding first specified time point, end time include after the time for first appearing target point of interest in rear specified time point,
To parse user's specifying after the first specified time point before arriving at target point of interest to arrival target point of interest rear
Run trace between time point.
Optionally, as shown in fig. 6, step S50 includes:
S5021, user's real time position is extracted from the position data based in setup time section by specified time interval；
S5022, the position input data is generated according to user's real time position, the position input data includes referring to
It fixes time a little and corresponding user's real time position, the specified time point is determined according to the specified time interval.
It should be noted that specified time interval mentioned here can be set according to actual needs, it can be 2 points
Clock, 1 minute or 30 seconds.The position input data of generation can be following form:
The position input data of user's first in 1 one embodiment of table
In upper table, if time interval is 2 minutes, the time at time point 1 is 12:00, then the time at time point 2 is 12:02,
The time of time point n is 2 (n-1) minutes after 12 points.Label is used to indicate the position at the place of target point of interest.Target is emerging
Interest point can refer to the position where preset point of interest discrimination model.
In step S5021 and S5022, extracted by specified time interval from the position data based in setup time section
User's real time position, to obtain user's real time position sequence with timing.According to user's real time position generation
Position input data, the position input data include specified time point and corresponding user's real time position, described specified
Time point determines according to the specified time interval, to generate the position input data that can characterize user's run trace.
Optionally, as shown in fig. 7, step S5022 further include:
S5023, user's real time position is calculated at a distance from the target point of interest and direction；
S5024, the position input number is generated at a distance from target point of interest with direction according to user's real time position
According to the position input data includes specified time point, user's real time position at a distance from target point of interest, user's real time position
With the direction of target point of interest, wherein the specified time point is determining according to the specified time interval, the real-time position of user
It sets corresponding with specified time point.
In the present embodiment, position input data may include user's real time position at a distance from target point of interest and direction.
Target point of interest can refer to the position where preset point of interest discrimination model.It can be according to user's real time position and target point of interest
Position calculate user's real time position at a distance from target point of interest.This distance can be user's real time position and target interest
The linear distance of point, is also possible to the path distance of user's real time position and target point of interest.At can be according to user's real time position
The direction of user's real time position and target point of interest is determined in the orientation of target point of interest.For example, user's real time position is in mesh
The positive north of point of interest is marked, then the direction of user's real time position and target point of interest is 0 degree；It is emerging that user's real time position is in target
The positive east of interest point, then the direction of user's real time position and target point of interest is 90 degree.And so on.Specified time interval can root
It is set, can be 2 minutes, 1 minute or 30 seconds according to actual needs.
The position input data of generation can be following form:
The position input data of user's second in 2 one embodiment of table
In table 2, label can be used for identifying the position where target point of interest.For example, label is for identifying in the n/2 time
Point whether there is target point of interest, and parameter value is yes/no.
In step S5023-S5024, user's real time position is calculated at a distance from the target point of interest and direction, with
Multiple behavioral datas relevant to target point of interest are obtained, can preferably reflect correlation of the user with point of interest.According to institute
It states user's real time position and generates the position input data, the position input data packet with direction at a distance from target point of interest
Specified time point and corresponding user's real time position are included at a distance from target point of interest and direction, the specified time point root
It is determined according to the specified time interval, to generate the position input data that can characterize user's run trace.
Optionally, as shown in figure 8, before step S60, further includes:
S301, source sample set is obtained, the source sample set includes positive sample and negative sample, the source sample set and institute's rheme
The affiliated city for setting data is identical.
In the present embodiment, source sample set can be the business data from outside purchase.Source sample set includes multiple users
Point of interest sample, these point of interest samples can be used for constructing point of interest discrimination model.Point of interest sample can be divided into positive sample and
Negative sample.Positive sample refers to really appearing in the sample that point of interest occurs and registers, and negative sample is referred in point of interest
Activity nearby, but the sample for information of not registering.Point of interest sample in the sample set of source is that calculative position data belongs to
In a city.To prevent leakage of personal information, used point of interest sample is all handled through anonymization.It can in point of interest sample
To include the information of registering of user.
S302, the positive sample and negative sample are trained, construct source sample discrimination model.
Positive sample and negative sample can be trained using deep neural network algorithm or other machines learning algorithm, be passed through
After successive ignition, corresponding source sample discrimination model is constructed.
In different points of interest, user's residence time has certain otherness.Therefore, it is possible to be based on user interest point
Sample set can establish multiple source sample discrimination models, such as Automobile Service Factory's discrimination model, seafood restaurant discrimination model, building materials
Hardware market discrimination model, shuttlecock brand shop discrimination model, mail service Room discrimination model etc..The number of source sample discrimination model
Amount is related to the point of interest quantity that interest sample each in the sample set of source marks, and can be several hundred, it is also possible to more.
The building process of source sample discrimination model may include: the interest first chosen in any one source sample set
Then point obtains other samples comprising the point of interest, the sample comprising the point of interest is denoted as Positive training sample.Then it obtains
Adjacent point of interest in the point of interest specified range, then the sample comprising adjacent point of interest is obtained, and be denoted as negative sample.To positive instruction
Practice sample and negative training sample is trained, obtains the source sample discrimination model of the point of interest.Here, specified range can be
100m or other distance parameters.
S303, point of interest discrimination model is constructed based on the source sample discrimination model, and obtain comprising interest point information
Position data；
S304, the training in the point of interest discrimination model using the position data comprising interest point information, training
After the completion, the preset point of interest discrimination model is obtained.
The present embodiment is that preset point of interest discrimination model is constructed by way of transfer learning.Transfer learning
(Transfer Learning) is a kind of machine learning method, is in addition the knowledge of a field (i.e. source domain) is moved to
One field (i.e. target domain), enables target domain to obtain better learning effect.Position comprising interest point information
Data refer to the position data comprising clear interest point information that application program is collected into.
The building process of preset point of interest discrimination model may include: by the preceding n of trained source sample discrimination model
Layer copies to the preceding n-layer of point of interest discrimination model；Remaining other layer of random initializtion of point of interest discrimination model；Input is comprising emerging
The position data of interest point information, starts to train iteration.Wherein, it when iterative calculation, can choose there are two types of iterative manner:
(1) this preceding n-layer that migration is come is freezed, i.e., when training position input data, does not change the value of this n-layer；(2)
Do not freeze this preceding n-layer, but can constantly adjust their value, referred to as finely tunes.It can be according to including multiple position input datas
The number of parameters of the size of sample set and preceding n-layer selects one kind of above two iterative manner.If defeated comprising multiple positions
Enter the sample set very little of data, and there are many number of parameters, over-fitting, generallys use the mode of n-layer before freezing in order to prevent；Instead
It, then by the way of fine tuning.
In step S301-S304, source sample set is obtained, the source sample set includes positive sample and negative sample, the source sample
This collection is identical as the affiliated city of the position data, to obtain for source sample set needed for constructing source sample discrimination model.
The positive sample and negative sample are trained, source sample discrimination model is constructed, to obtain source sample discrimination model.Based on described
Source sample discrimination model constructs point of interest discrimination model, and obtains the position data comprising interest point information；Include using described
The training in the point of interest discrimination model of the position data of interest point information obtains preset point of interest and sentences after the completion of training
Other model, to complete the training to point of interest discrimination model, the preset point of interest discrimination model for obtaining training can be used for
Handle position input data.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of user interest point assessment device is provided, which assesses device and above-mentioned reality
User interest point appraisal procedure in example is applied to correspond.As shown in figure 9, user interest point assessment device includes obtaining positional number
According to module 10, determine regional scope module 20, acquisition discrimination model module 30, acquisition processing rule module 40, generation input number
According to module 50 and obtain differentiation object module 60.Detailed description are as follows for each functional module:
Position data module 10 is obtained, for obtaining the position data of user；
Regional scope module 20 is determined, for determining target area range according to the position data；
Discrimination model module 30 is obtained, is differentiated for obtaining preset point of interest associated with the target area range
Model；
Processing rule module 40 is obtained, for obtaining and the preset associated processing rule of point of interest discrimination model；
Input data module 50 is generated, the position data according to the processing rule process is used for, generates position input
Data；
It obtains and differentiates object module 60, for the position input data input preset point of interest to be differentiated mould
Type, the point of interest for obtaining the position data differentiate as a result, and differentiating that result determines the point of interest of user according to point of interest.
Specific restriction about user interest point assessment device may refer to above for user interest point appraisal procedure
Restriction, details are not described herein.Modules in above-mentioned user interest point assessment device can be fully or partially through software, hard
Part and combinations thereof is realized.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment,
It can also be stored in a software form in the memory in computer equipment, execute the above modules in order to which processor calls
Corresponding operation.
Optionally, obtaining position data module 10 includes:
First acquisition unit, for obtaining the first position data of user in the first specified time；
Judging unit, for judging whether the size of the first position data is less than preset threshold；
First determination unit, if the size for the first position data is not less than preset threshold, by described first
Position data is determined as the position data；
It is specified to obtain second if the size for the first position data is less than preset threshold for second determination unit
The second position data are determined as the position data by the second position data of user in the time, it is described second it is specified when
Between be greater than first specified time.
Optionally, determine that regional scope module 20 includes:
Regional scope unit is judged, for judging whether the position of the position data is in predeterminable area range
It is interior；
Target area unit is determined, if the predeterminable area range is determined as target for being in predeterminable area range
Regional scope.
Optionally, input data module 50 is generated further include:
Determine interest dot element, for determining the time point for first appearing target point of interest, the preset point of interest is sentenced
Other model includes the target point of interest；
Input data unit, it is described for generating the position input data based on the position data in setup time section
Setup time section start time include the time for first appearing target point of interest before first specified time point, end time
After time including first appearing target point of interest in rear specified time point.
Optionally, input data unit includes:
Position units are extracted, are used for being extracted by specified time interval from the position data based in setup time section
Family real time position；
Position generation unit, for generating the position input data according to user's real time position, the position is defeated
Entering data includes specified time point and corresponding user's real time position, and the specified time point is according between the specified time
Every determination.
Optionally, position generation unit includes:
Computer azimuth unit, for calculating user's real time position at a distance from target point of interest and direction；
Orientation generation unit, described in being generated at a distance from target point of interest with direction according to user's real time position
Position input data, the position input data include specified time point and corresponding user's real time position and target interest
The distance of point and direction, the specified time point are determined according to the specified time interval.
Optionally, user interest point assessment device further includes building model module, and the building model module includes:
Acquisition source sample set unit, for obtaining source sample set, the source sample set includes positive sample and negative sample, described
Source sample set is identical as the affiliated city of the position data；
Source model unit is constructed, for being trained to the positive sample and negative sample, constructs source sample discrimination model；
Object module unit is constructed, for constructing point of interest discrimination model based on the source sample discrimination model, uses institute
The training in the point of interest discrimination model of the position data comprising interest point information is stated, after the completion of training, is obtained described default
Point of interest discrimination model.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 10.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is used for storage location data and preset point of interest discrimination model.The network interface of the computer equipment is used
It is communicated in passing through network connection with external terminal.To realize a kind of user interest point when the computer program is executed by processor
Appraisal procedure.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor perform the steps of when executing computer program
Obtain the position data of user；
Target area range is determined according to the position data；
Obtain preset point of interest discrimination model associated with the target area range；
It obtains and the preset associated processing rule of point of interest discrimination model；
According to position data described in the processing rule process, position input data is generated；
The position input data is inputted into the preset point of interest discrimination model, obtains the interest of the position data
Point differentiates as a result, and differentiating that result determines the point of interest of user according to point of interest.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Obtain the position data of user；
Target area range is determined according to the position data；
Obtain preset point of interest discrimination model associated with the target area range；
It obtains and the preset associated processing rule of point of interest discrimination model；
According to position data described in the processing rule process, position input data is generated；
The position input data is inputted into the preset point of interest discrimination model, obtains the interest of the position data
Point differentiates as a result, and differentiating that result determines the point of interest of user according to point of interest.
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, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of user interest point appraisal procedure characterized by comprising
Obtain the position data of user；
Target area range is determined according to the position data；
Obtain preset point of interest discrimination model associated with the target area range；
It obtains and the preset associated processing rule of point of interest discrimination model；
According to position data described in the processing rule process, position input data is generated；
The position input data is inputted into the preset point of interest discrimination model, the point of interest for obtaining the position data is sentenced
Not as a result, and differentiating that result determines the point of interest of user according to point of interest.
2. user interest point appraisal procedure as described in claim 1, which is characterized in that the position data for obtaining user,
Include:
Obtain the first position data of user in the first specified time；
Judge whether the size of the first position data is less than preset threshold；
If the size of the first position data is not less than preset threshold, the first position data are determined as the position
Data；
If the size of the first position data is less than preset threshold, the second position number of user in the second specified time is obtained
According to the second position data being determined as the position data, second specified time is greater than first specified time.
3. user interest point appraisal procedure as described in claim 1, which is characterized in that described to be determined according to the position data
Target area range, comprising:
Judge whether the position of the position data is within the scope of predeterminable area；
If the predeterminable area range is determined as target area range within the scope of predeterminable area.
4. user interest point appraisal procedure as described in claim 1, which is characterized in that described according to the processing rule process
The position data generates position input data, comprising:
Determine the time point for first appearing target point of interest, the preset point of interest discrimination model includes the target interest
Point；
The position input data is generated based on the position data in setup time section, the setup time section includes start time
And end time, the start time include the time for first appearing target point of interest before first specified time point, it is described
End time includes after the time for first appearing target point of interest in rear specified time point.
5. user interest point appraisal procedure as claimed in claim 4, which is characterized in that the position based in setup time section
It sets data and generates the position input data, comprising:
User's real time position is extracted from the position data based in setup time section by specified time interval；
Generate the position input data according to user's real time position, the position input data include specified time point and
Corresponding user's real time position, the specified time point are determined according to the specified time interval.
6. user interest point appraisal procedure as claimed in claim 5, which is characterized in that described according to user's real time position
Generate the position input data, comprising:
User's real time position is calculated at a distance from the target point of interest and direction；
The position input data, the position are generated with direction at a distance from target point of interest according to user's real time position
Input data include specified time point, user's real time position at a distance from the target point of interest, user's real time position with it is described
The direction of target point of interest, wherein the specified time point is determining according to the specified time interval, user's real time position
It is corresponding with the specified time point.
7. the user interest point appraisal procedure as described in claim 1-6, which is characterized in that the acquisition and the target area
Before the associated preset point of interest discrimination model of range, further includes:
Acquisition source sample set, the source sample set include positive sample and negative sample, the source sample set and the position data
Affiliated city is identical；
The positive sample and negative sample are trained, source sample discrimination model is constructed；
Point of interest discrimination model is constructed based on the source sample discrimination model, and obtains the position data comprising interest point information；
The training in the point of interest discrimination model using the position data comprising interest point information, after the completion of training, obtains institute
State preset point of interest discrimination model.
8. a kind of user interest point assesses device characterized by comprising
Position data module is obtained, for obtaining the position data of user；
Regional scope module is determined, for determining target area range according to the position data；
Discrimination model module is obtained, for obtaining preset point of interest discrimination model associated with the target area range；
Processing rule module is obtained, for obtaining and the preset associated processing rule of point of interest discrimination model；
Input data module is generated, the position data according to the processing rule process is used for, generates position input data；
It obtains and differentiates object module, for the position input data to be inputted the preset point of interest discrimination model, obtain
The point of interest of the position data differentiates as a result, and differentiating that result determines the point of interest of user according to point of interest.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
The step of any one of 7 user interest point appraisal procedure.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In realization user interest point appraisal procedure as described in any one of claim 1 to 7 when the computer program is executed by processor
The step of.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811187932.6A CN109460509A (en) | 2018-10-12 | 2018-10-12 | User interest point appraisal procedure, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811187932.6A CN109460509A (en) | 2018-10-12 | 2018-10-12 | User interest point appraisal procedure, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109460509A true CN109460509A (en) | 2019-03-12 |
Family
ID=65607612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811187932.6A Pending CN109460509A (en) | 2018-10-12 | 2018-10-12 | User interest point appraisal procedure, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109460509A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852354A (en) * | 2019-10-22 | 2020-02-28 | 上海中旖能源科技有限公司 | Vehicle track point identification method and device |
| CN111858787A (en) * | 2019-09-24 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | POI information acquisition method and device |
| CN112069281A (en) * | 2020-09-08 | 2020-12-11 | 智慧足迹数据科技有限公司 | Method and device for determining position of interest point, computer equipment and storage medium |
| CN112232845A (en) * | 2019-07-15 | 2021-01-15 | 中国移动通信集团重庆有限公司 | Method and device for predicting user behavior preference based on user position |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750292A (en) * | 2011-04-20 | 2012-10-24 | 北京千橡网景科技发展有限公司 | Method and device for providing interest points |
| CN104239453A (en) * | 2014-09-02 | 2014-12-24 | 百度在线网络技术（北京）有限公司 | Data processing method and device |
| CN104615707A (en) * | 2015-02-02 | 2015-05-13 | 百度在线网络技术（北京）有限公司 | Information point determining method and device |
| CN106709606A (en) * | 2016-12-29 | 2017-05-24 | 平安科技（深圳）有限公司 | Personalized scene prediction method and apparatus |
| US20180137204A1 (en) * | 2016-05-10 | 2018-05-17 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for recommending personalized content |
-
2018
- 2018-10-12 CN CN201811187932.6A patent/CN109460509A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750292A (en) * | 2011-04-20 | 2012-10-24 | 北京千橡网景科技发展有限公司 | Method and device for providing interest points |
| CN104239453A (en) * | 2014-09-02 | 2014-12-24 | 百度在线网络技术（北京）有限公司 | Data processing method and device |
| CN104615707A (en) * | 2015-02-02 | 2015-05-13 | 百度在线网络技术（北京）有限公司 | Information point determining method and device |
| US20180137204A1 (en) * | 2016-05-10 | 2018-05-17 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for recommending personalized content |
| CN106709606A (en) * | 2016-12-29 | 2017-05-24 | 平安科技（深圳）有限公司 | Personalized scene prediction method and apparatus |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112232845A (en) * | 2019-07-15 | 2021-01-15 | 中国移动通信集团重庆有限公司 | Method and device for predicting user behavior preference based on user position |
| CN112232845B (en) * | 2019-07-15 | 2023-12-19 | 中国移动通信集团重庆有限公司 | Method and device for predicting user behavior preference based on user position |
| CN111858787A (en) * | 2019-09-24 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | POI information acquisition method and device |
| CN110852354A (en) * | 2019-10-22 | 2020-02-28 | 上海中旖能源科技有限公司 | Vehicle track point identification method and device |
| CN112069281A (en) * | 2020-09-08 | 2020-12-11 | 智慧足迹数据科技有限公司 | Method and device for determining position of interest point, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460509A (en) | User interest point appraisal procedure, device, computer equipment and storage medium | |
| Buelens et al. | Comparing inference methods for non‐probability samples | |
| Konowalik et al. | Evaluation metrics and validation of presence-only species distribution models based on distributional maps with varying coverage | |
| Fordham et al. | How complex should models be? Comparing correlative and mechanistic range dynamics models | |
| US9709408B2 (en) | Segment validation | |
| Arsanjani et al. | Integration of logistic regression, Markov chain and cellular automata models to simulate urban expansion | |
| Wrenn et al. | Geographically and temporally weighted likelihood regression: Exploring the spatiotemporal determinants of land use change | |
| CN110490608B (en) | Risk assessment method and device, computer equipment and storage medium | |
| US9582819B2 (en) | Automated-valuation-model training-data optimization systems and methods | |
| Mateu et al. | Spatio-temporal design: Advances in efficient data acquisition | |
| CN110457420A (en) | Point of interest location recognition methods, device, equipment and storage medium | |
| CN104764868A (en) | Soil organic carbon predication method based on geographically weighted regression | |
| EP3731159A1 (en) | Adaptive multiyear economic planning method for energy systems, microgrid and distributed energy resources | |
| CN112184089B (en) | Training method, device and equipment of test question difficulty prediction model and storage medium | |
| Roumboutsos | Public private partnerships in transport: trends and theory | |
| CN104008208A (en) | Situation recognition system and method based on opportunity perception | |
| CN116188676A (en) | Method and device for establishing digital twin model of digital center | |
| CN111177565B (en) | Interest point recommendation method based on correlation matrix and word vector model | |
| CN113326877A (en) | Model training method, data processing method, device, apparatus, storage medium, and program | |
| Patel et al. | Identification of Best CMIP6 Global Climate Model for Rainfall by Ensemble Implementation of MCDM Methods and Statistical Inference | |
| CN107609205A (en) | Business simulating system and method | |
| Osmond et al. | Design and Implementation of Smart Trip Planner | |
| CN117473327B (en) | Regional population model training method and regional population prediction method | |
| Abraham et al. | Selection of suitable general circulation model outputs of precipitation for a humid tropical basin | |
| CN110321495A (en) | Method for pushing, device, computer equipment and the storage medium of alive messages |
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 |