CN108875032A - Area type determines method and device - Google Patents

Abstract

The embodiment of the present invention provides a kind of area type and determines method and device, belongs to computer application technology.This method includes：Obtain corresponding matrix and the arrival matrix of setting out of each city zoning；Specified parameter is input to preset model, the corresponding preset kind of each city zoning is determined based on the output result of preset model.Specified parameter is input to preset model, determines the corresponding preset kind of each city zoning based on the output result of preset model by the embodiment of the present invention by obtaining each city zoning corresponding set out matrix and arrival matrix.Since the corresponding area type of city zoning can be automatically determined based on mobile data of the user between the subregion of different cities, so that it is determined that result is more accurate.In addition, introducing time dimension due to when determining the corresponding area type of city zoning, being for same city zoning, it may be determined that the corresponding area type of the city zoning in different time sections, so as to the application scenarios that more gear to actual circumstances.

Description

Area type determines method and device

Technical field

The present embodiments relate to computer application technologies, determine method more particularly, to a kind of area type And device.

Background technique

In recent years, the scale in city is increasing, under application scenes it needs to be determined that in city different subregions class Type.For example, dividing according to function to city, area type may include residential area, shopping centre and industrial area etc..The relevant technologies In the area type for determining city zoning, the point of interest in each city zoning is usually first determined, such as shop, school and public affairs Department etc. then according to the point of interest in each city zoning, then determines by manual type the area type of each city zoning.By It then passes through manual type to determine, definitive result is not accurate enough.

Summary of the invention

To solve the above-mentioned problems, the embodiment of the present invention provides one kind and overcomes the above problem or at least be partially solved A kind of area type for stating problem determines method and device.

According to a first aspect of the embodiments of the present invention, it provides a kind of area type and determines that method, this method include：

By obtaining each city zoning corresponding set out matrix and arrival matrix, specified parameter is input to default mould Type determines the corresponding preset kind of each city zoning based on the output result of preset model.Since user can be based in difference Mobile data between city zoning automatically determines the corresponding area type of city zoning, so that it is determined that result is more accurate. In addition, introduce time dimension due to when determining the corresponding area type of city zoning, be for same city zoning, It can determine the corresponding area type of the city zoning in different time sections, so as to the application scenarios that more gear to actual circumstances.

Method provided in an embodiment of the present invention, by obtaining each city zoning corresponding set out matrix and arrival matrix, Specified parameter is input to preset model, the corresponding default class of each city zoning is determined based on the output result of preset model Type.Since the corresponding region class of city zoning can be automatically determined based on mobile data of the user between the subregion of different cities Type, so that it is determined that result is more accurate.In addition, introducing time dimension due to when determining the corresponding area type of city zoning Degree, is for same city zoning, it may be determined that the corresponding area type of the city zoning in different time sections, so as to more Gear to actual circumstances application scenarios.

According to a second aspect of the embodiments of the present invention, a kind of area type determining device is provided, which includes：

Module is obtained, for obtaining corresponding matrix and the arrival matrix of setting out of each city zoning；Wherein, city zoning is It is obtained after being divided to city；For any city zoning in city, the corresponding matrix packet that sets out of any city zoning The mobile frequency that all users are respectively moved to each other city zonings from any city zoning within a preset period of time is included, is appointed The corresponding arrival matrix of one city zoning includes that all users are mobile from each other city zonings respectively within a preset period of time To the mobile frequency of any city zoning, other city zonings are the city zoning in city in addition to any city zoning；

First determining module is determined for specified parameter to be input to preset model based on the output result of preset model The corresponding preset kind of each city zoning；Wherein, specifying parameter to include at least, preset kind is total, each city zoning is corresponding Set out matrix and reach matrix.

According to a third aspect of the embodiments of the present invention, a kind of electronic equipment is provided, including：

At least one processor；And

At least one processor being connect with processor communication, wherein：

Memory is stored with the program instruction that can be executed by processor, and the instruction of processor caller is able to carry out first party Area type provided by any possible implementation determines method in the various possible implementations in face.

According to the fourth aspect of the invention, a kind of non-transient computer readable storage medium, non-transient computer are provided Readable storage medium storing program for executing stores computer instruction, and computer instruction makes the various possible implementations of computer execution first aspect In area type provided by any possible implementation determine method.

It should be understood that above general description and following detailed description be it is exemplary and explanatory, can not Limit the embodiment of the present invention.

Detailed description of the invention

Fig. 1 is that a kind of area type of the embodiment of the present invention determines the flow diagram of method；

Fig. 2 is that a kind of area type of the embodiment of the present invention determines the flow diagram of method；

Fig. 3 is that a kind of city zoning of the embodiment of the present invention covers the schematic diagram of Thiessen polygon；

Fig. 4 is that a kind of area type of the embodiment of the present invention determines the flow diagram of method；

Fig. 5 is a kind of schematic diagram of building trip sequence of the embodiment of the present invention；

Fig. 6 is that a kind of area type of the embodiment of the present invention determines the flow diagram of method；

Fig. 7 is a kind of block diagram of area type determining device of the embodiment of the present invention；

Fig. 8 is the block diagram of a kind of electronic equipment of the embodiment of the present invention.

Specific embodiment

With reference to the accompanying drawings and examples, the specific embodiment of the embodiment of the present invention is described in further detail.With Lower embodiment is not limited to the range of the embodiment of the present invention for illustrating the embodiment of the present invention.

Increasing recently as city size, urban issues are more and more obvious, such as traffic congestion and air pollution Problem.Thus the angle excavated with analyzed from big data is needed, to construct the scheme to solve the above problems.For in city not Same city zoning, the scheme for determining that the area type of city zoning solves the above problems for building are of great significance.True When determining the area type of city zoning, the area type of each city zoning is usually determined according to the angle of functional areas, such as It is divided into residential area, shopping centre and industrial area etc..The relevant technologies are usually first determined in the area type for determining city zoning Point of interest in each city zoning, such as shop, school and company, then according to the point of interest in each city zoning, then lead to Cross the area type that manual type determines each city zoning.It is determined by then passing through manual type, definitive result is not accurate enough. In addition, the area type of each city zoning is substantially not due to after the area type for determining city zoning by manual type It can change, and for city zoning, the function of carrying may change with time change, such as with urban construction Propulsion or with the stream of people movement and change, thus by manual type determine area type cannot be in time dimension upper body Reveal the variation of area type.

For said circumstances, the embodiment of the invention provides a kind of area types to determine method.This method is suitable for any The division scene of urban function region, to determine area type of the different cities subregion in the angle of functional areas.The execution of this method Main body can be pre-set processing equipment, and such as server or terminal, the present invention is not especially limit this.Tool Body, this method includes：

101, corresponding matrix and the arrival matrix of setting out of each city zoning is obtained.

Before executing 101, city can be divided to obtain multiple city zonings.It specifically, can be according to road network number Multiple city zonings are divided into according to by city, the present invention is not especially limit this.In 101, in city Any city zoning, the corresponding matrix that sets out of the city zoning can have two dimensions of ranks.Wherein, the row of matrix of setting out can refer to Other city zonings that alternative family is moved to after the city zoning, when the matrix column that sets out can be referred to different default Between section.The matrix that sets out may include that all users are respectively moved to each other cities point from the city zoning within a preset period of time The mobile frequency in area, each element representation in matrix that sets out are moved in a certain preset time period from the city zoning a certain The mobile frequency of other city zonings, the present invention is not especially limit this.Wherein, other city zonings refer to city City zoning in area in addition to the city zoning.

For example, being respectively A, B, C and D with the city zoning in city, preset time period is respectively workaday 8 a.m. To for 9 points, 9 points to 10 points and 10 points to 11 points.For city zoning A, the line number of the corresponding matrix that sets out of A is 3 rows, Namely respectively correspond city zoning B, C and D.The corresponding matrix column number that sets out of A is 3 column, namely corresponding above three is different Period.For the element that the 1st row the 1st in the matrix that sets out arranges, the meaning of the element representation be all users 8 a.m. extremely 9 points of this periods from A, are moved to the mobile frequency of B.

It should be noted that preset time period can also be for the ease of subsequent statistical during actual implementation with hour Regular unit is integer time point.For example, for any user, if the user is in 8 a.m. 40 (namely in 8 a.m. to 9 points It is interior) from A be moved to B, then it can be unit hour by the user be 8 a.m. from the Time alignment of A or regular It is 9 points of the morning, the present invention is not especially limit this.At this point, corresponding its column reference of matrix of setting out of A is different Integer time point.In addition, the preset time period that is previously mentioned of the embodiment of the present invention and subsequent embodiment is in addition to for the workaday time Section is outer, can also can be afternoon or the period in evening, this hair with the period at weekend, other than the period in the morning Bright embodiment is not especially limited this.

Similarly, for any city zoning in city, the corresponding arrival matrix of the city zoning can equally have ranks Two dimensions.Wherein, the row for reaching matrix can be referred to user and which other city zoning be moved to the city zoning from, reach square The column of battle array can be referred to different preset time periods.Reaching matrix may include all users within a preset period of time from each other cities City's subregion is moved to the mobile frequency of the city zoning, reach each element representation in matrix in a certain preset time period from A certain other city zonings are moved to the mobile frequency of the city zoning, and the present invention is not especially limit this.

For example, being respectively A, B, C and D with the city zoning in city, preset time period is respectively workaday 8 a.m. To for 9 points, 9 points to 10 points and 10 points to 11 points.For city zoning A, the corresponding line number for reaching matrix of A is 3 rows, Namely respectively correspond city zoning B, C and D.The corresponding arrival matrix column number of A arranges for 3, namely corresponding above three is different Period.For the element that the 1st row the 1st in the arrival matrix arranges, the meaning of the element representation is all users from B, and In the mobile frequency for being moved to A in 8 a.m. to 9 points of this periods.

Similarly, for the ease of subsequent statistical during actual implementation, preset time period can also be unit rule according to hour Whole is integer time point.For example, for any user, if the user from B, and in 8 a.m. 40 (namely in 8 a.m. To in 9 points) be moved to A, then can be unit hour by the Time alignment that the user reaches A be 8 a.m. or it is regular be upper 9 points of noon, the present invention is not especially limit this.At this point, its column reference of the corresponding arrival matrix of A is different integer Time point.

In addition, in 101, the corresponding matrix and when reaching matrix of setting out of each city zoning is being obtained, user can be based on The position data that uploads within a preset period of time obtains, and the present invention is not especially limit this.For example, user makes With positioning function would generally be opened when mobile terminal, so that each city can be obtained according to the position data that mobile terminal uploads Corresponding matrix and the arrival matrix of setting out of subregion.

102, specified parameter is input to preset model, each city zoning pair is determined based on the output result of preset model The preset kind answered；Wherein, specify that parameter includes at least preset kind sum, each city zoning corresponding set out and is arrived matrix Up to matrix.

It, will be from the city zoning corresponding matrix if the city zoning is considered as document for any city zoning And each element reached in matrix is considered as the word in the document, it is determined that the problem of area type of city zoning can turn The problem of turning to determining document subject matter.

For example, being respectively A, B, C and D with the city zoning in city, preset time period is respectively workaday 8 a.m. To for 9 points, 9 points to 10 points and 10 points to 11 points.If city zoning A is considered as document, A is corresponding set out matrix and The each element reached in matrix can be considered the word in the document.It, should for the element that the 1st row the 1st in the arrival matrix arranges The meaning of element representation is all users from B, and in the mobile frequency for being moved to A in 8 a.m. to 9 points of this periods It is secondary.That is, for the element of the 1st row the 1st column, a kind of element representation mobile relationship in a period of time, the movement Relationship can be considered as the word in document, and the corresponding frequency occurred in a document of the word is also the mobile frequency.

By above-mentioned principle it is found that the problem of determining the area type of city zoning can be converted into asking for determining document subject matter Topic.That is, probability distribution when the document is every kind of theme can be determined first for any document in a series of document, from And it can determine probability value when the document is every kind of theme, and using the maximum theme of probability value as the corresponding theme of the document. Correspondingly, in embodiments of the present invention, for any city zoning, by the exportable city zoning of preset model as every kind Probability distribution when preset kind, to can determine the maximum preset kind of probability value and corresponding default as the city zoning Type.

Based on above description, in 102, the type of preset model can generate model, such as LDA for document subject matter (Latent Dirichlet Allocation implies the distribution of Di Li Cray) model etc., the embodiment of the present invention is not made this specifically It limits.The specified parameter that preset model inputs when in use can be corresponding including at least preset kind sum, each city zoning Set out matrix and arrival matrix.Wherein, preset kind sum namely predetermined area type sum.For example, if according to function Can the angle in area determine the area type of each city zoning, then predetermined area type can be residential area, business Area, industrial area and school district.At this point, predetermined area type sum is 4 kinds.Specified parameter is input to preset model, it can Export probability distribution of each city zoning as every kind of area type when.For example, for any city zoning, if by default It is 0.25 that model, which exports the city zoning as the probability value of residential area, and the probability value as shopping centre is 0.35, as industry The probability value in area is 0.3, and the probability value as school district is 0.1.It therefore, can be by the maximum area type of probability value, namely business Area type of the area as the city zoning.After the area type for determining each city zoning, can there will be same area class The city zoning of type is clustered.

It should be noted that can be one the present embodiments relate to the quantity of the preset time period arrived, or Multiple, the present invention is not especially limit this.If the quantity the present embodiments relate to the preset time period arrived is more It is a, then it is subsequent setting out for multiple periods and to be reached Input matrix to preset model, also i.e. by multiple periods at matrix Set out matrix and reach matrix be fitted, to obtain each city zoning in corresponding overall time section of multiple period Area type.

Method provided in an embodiment of the present invention, by obtaining each city zoning corresponding set out matrix and arrival matrix, Specified parameter is input to preset model, the corresponding default class of each city zoning is determined based on the output result of preset model Type.Since the corresponding region class of city zoning can be automatically determined based on mobile data of the user between the subregion of different cities Type, so that it is determined that result is more accurate.In addition, introducing time dimension due to when determining the corresponding area type of city zoning Degree, is for same city zoning, it may be determined that the corresponding area type of the city zoning in different time sections, so as to more Gear to actual circumstances application scenarios.

Content based on the above embodiment, as a kind of alternative embodiment, the embodiment of the present invention is not to each city of acquisition The mode of the corresponding matrix that sets out of subregion specifically limits, including but not limited to：For any city zoning, each user is obtained Corresponding user sets out matrix；Wherein, for any user in all users, the corresponding user of any user sets out matrix packet Include the mobile frequency that any user is respectively moved to each other city zonings from any city zoning within a preset period of time；It will The corresponding user of each user matrix that sets out is overlapped, and obtains the corresponding matrix that sets out of any city zoning.

By the definition of the matrix corresponding of city zoning in above-described embodiment it is found that for any city zoning, the city The corresponding matrix that sets out of city's subregion include all users be respectively moved within a preset period of time from the city zoning it is each other The mobile frequency of city zoning.During actual implementation, it can first obtain the corresponding user of each user and set out matrix, by each use The corresponding user in the family matrix that sets out is overlapped, and the corresponding matrix that sets out of the city zoning can be obtained.It should be noted that will The corresponding user of each user matrix that sets out is overlapped, the element mainly to set out in matrix in same position to each user into Row superposition, so that the element in the corresponding matrix that sets out of the city zoning in same position can be obtained.

Wherein, for any user, the corresponding user of the user matrix that sets out can equally have two dimensions of ranks.Its In, the set out row of matrix of user can be referred to other city zonings that the user is moved to after the city zoning, user The matrix column that sets out can be referred to different preset time periods.User's matrix that sets out may include the user within a preset period of time from this City zoning is respectively moved to the mobile frequency of each other city zonings, and user sets out each element representation use in matrix Family is moved to the mobile frequency of a certain other city zonings, the embodiment of the present invention in a certain preset time period from the city zoning This is not especially limited.By the above process, the corresponding matrix that sets out of each city zoning can finally be obtained.

Content based on the above embodiment, as a kind of alternative embodiment, the embodiment of the present invention is not to each city of acquisition The corresponding mode for reaching matrix of subregion specifically limits, including but not limited to：For any city zoning, each user is obtained Corresponding user reaches matrix；Wherein, for any user in all users, the corresponding user of any user reaches matrix packet Include the mobile frequency that any user is moved to any city zoning from each other city zonings respectively within a preset period of time；It will The corresponding user of each user reaches matrix and is overlapped, and obtains the corresponding arrival matrix of any city zoning.

By the corresponding definition for reaching matrix of city zoning in above-described embodiment it is found that for any city zoning, the city Subregion corresponding arrival matrix in city's includes that all users are respectively moved to this from each other city zonings within a preset period of time The mobile frequency that city zoning is moved to.During actual implementation, it can first obtain the corresponding user of each user and reach matrix, it will The corresponding user of each user reaches matrix and is overlapped, and the corresponding arrival matrix of the city zoning can be obtained.It needs to illustrate , the corresponding user of each user is reached into matrix and is overlapped, mainly each user is reached in matrix in same position Element be overlapped, so that the corresponding element reached in matrix in same position of the city zoning can be obtained.

Wherein, for any user, the corresponding user of the user, which reaches matrix, can equally have two dimensions of ranks.Its In, the row that user reaches matrix can be referred to the user and which other city zoning be moved to the city zoning from, and user reaches square The column of battle array can be referred to different preset time periods.It may include the user within a preset period of time from each other that user, which reaches matrix, City zoning is respectively moved to the mobile frequency of the city zoning, and user reaches each element representation user in matrix at certain Be moved to the mobile frequency of the city zoning in one preset time period from a certain other city zonings, the embodiment of the present invention to this not Make specific limit.By the above process, the corresponding arrival matrix of each city zoning can finally be obtained.

By the content of above-described embodiment it is found that matrix and user reach the element representation in matrix at one user A kind of mobile relationship in period.For example, being respectively A, B, C and D with the city zoning in city, preset time period is respectively Workaday 8 a.m. is to for 9 points, 9 points to 10 points and 10 points to 11 points.If the corresponding user of user sets out matrix its City zoning as departure place is A, then sets out the elements of the 1st row the 1st column in matrix for the user, which contains Justice is the user from A, and in the mobile frequency namely the 1st row the 1st that are moved to B in 8 a.m. to 9 points of this periods A kind of mobile relationship that B is moved to from A of the element representation of column.

Content based on above description and above-described embodiment during actual implementation, can be obtained based on above-mentioned mobile relationship The corresponding user of each user is taken to set out matrix.Accordingly, as a kind of alternative embodiment, the embodiment of the present invention is not every to obtaining The set out mode of matrix of the corresponding user of one user specifically limits, including but not limited to：For any user, exist from the user The trip sequence for meeting the first preset condition is determined in all trip sequences in preset time period；Wherein, the first preset condition To be departure place and using other city zonings as destination using any city zoning, each trip sequence is corresponding with mobile frequency It is secondary；According to the trip sequence for meeting the first preset condition, determine that the corresponding user of any user sets out matrix.

Wherein, all trip sequences of the user within a preset period of time can be uploaded within a preset period of time based on user Position data obtains, and the present invention is not especially limit this.For example, user uploads positional number in 8 a.m. 40 According to, and position data is uploaded at 9 points of the morning.If based on two time points upload position data determine user during this period into It has gone movement, then the position data uploaded twice can be constructed to corresponding trip sequence.Wherein, following shape can be used in trip sequence Formula indicates：

((k, start_time), (k', end_time), w_i)

Wherein, k indicates the city zoning as departure place, and k' indicates city zoning as a purpose.start_time Indicate that, from the time of departure place, end_time indicates the time arrived at the destination, w_iIndicate the corresponding movement of trip sequence The frequency.

It is respectively A, B, C and D with the city zoning in city, preset time period is respectively workaday 8 a.m. to 9 For point, 9 points to 10 points and 10 points to 11 points.For any user, all trips of the user in 8 a.m. to 9 points Sequence refers to that (namely start_time is in the morning 8 in 8 a.m. to the trip sequence in 9 points from any city zoning O'clock in 9 points), and (namely end_time is in the morning 8 in 8 a.m. to the trip sequence for reaching any city zoning in 9 points O'clock in 9 points).If the city zoning as departure place is A, the sequence of setting out for meeting the first preset condition is above-mentioned trip sequence K value is the trip sequence of A in column expression formula, namely in 8 a.m. to the trip sequence in 9 points from A.

As shown in the above, trip sequence can determine one section of mobile relationship in a period between city zoning, And trip sequence is also corresponding with the mobile frequency.Therefore, it can determine that the corresponding user of user sets out matrix based on trip sequence.Example It such as, is respectively A, B, C and D with the city zoning in city, preset time period is respectively workaday 8 a.m. to 9 points, 9 points To 10 points and 10 points to 11 points.If meeting the trip sequence of the first preset condition using city zoning A as departure place Following several types can be divided into：

(1) in 8 a.m. to from A, being moved to the trip sequence of B in 9 points；

(2) the trip sequence of B is moved to from A in 9 points to 10 points of the morning；

(3) the trip sequence of B is moved to from A in 10 points to 11 points of the morning；

(4) in 8 a.m. to from A, being moved to the trip sequence of C in 9 points；

(5) the trip sequence of C is moved to from A in 9 points to 10 points of the morning；

(6) the trip sequence of C is moved to from A in 10 points to 11 points of the morning；

(7) in 8 a.m. to from A, being moved to the trip sequence of D in 9 points；

(8) the trip sequence of D is moved to from A in 9 points to 10 points of the morning；

(9) the trip sequence of D is moved to from A in 10 points to 11 points of the morning.

Wherein, above-mentioned every kind of trip sequence is corresponding with the mobile frequency.It should be noted that during actual implementation, on The quantity for stating every kind of trip sequence may more than one.For example, user in 8 a.m. to B is moved to from A in 9 points, connect In 8 a.m. to being moved to A from B in 9 points, it is last again in 8 a.m. to being moved to B from A in 9 points.At this point, meeting Occur two in 8 a.m. to from A, being moved to the trip sequence of B in 9 points.It, can be by this two trip sequences for the situation It arranges the corresponding mobile frequency to be added, and merges into a trip sequence.

By the content of above-described embodiment it is found that if departure place is city zoning A, the set out row of matrix of user can be referred to this Other city zonings that user is moved to after A, user's matrix column that sets out can be referred to different preset time periods.Knot Share family set out matrix definition and it is above-mentioned about trip sequence explanation, according to meet the first preset condition trip sequence Column can determine that the corresponding user of user sets out matrix.It should be noted that if departure place is other city zonings, equally may be used To determine the corresponding trip sequence for meeting the first preset condition of each user as procedure described above, and then determine each user's User sets out matrix, and details are not described herein again.By the above process, it may finally obtain in the feelings that different cities subregion is departure place Under condition, the corresponding user of each user sets out matrix.

Content based on above description and above-described embodiment during actual implementation, can be obtained based on above-mentioned mobile relationship The corresponding user of each user is taken to reach matrix.Accordingly, as a kind of alternative embodiment, the embodiment of the present invention is not every to obtaining The mode that the corresponding user of one user reaches matrix specifically limits, including but not limited to：For any user, from any user The trip sequence for meeting the second preset condition is determined in all trip sequences within a preset period of time；Wherein, the second default item It is departure place and using any city zoning as destination that part, which is using other city zonings, each trip sequence be corresponding with it is mobile frequently It is secondary；According to the trip sequence for meeting the second preset condition, determine that the corresponding user of any user reaches matrix.

Wherein, about the user within a preset period of time it is all trip sequences explanation, and trip sequence specific shape Formula can refer to the explanation in above-described embodiment,

It is respectively A, B, C and D with the city zoning in city, preset time period is respectively workaday 8 a.m. to 9 For point, 9 points to 10 points and 10 points to 11 points.For any user, all trips of the user in 8 a.m. to 9 points Sequence refers to that (namely start_time is in the morning 8 in 8 a.m. to the trip sequence in 9 points from any city zoning O'clock in 9 points), and (namely end_time is in the morning 8 in 8 a.m. to the trip sequence for reaching any city zoning in 9 points O'clock in 9 points).If city zoning as a purpose is A, the trip sequence for meeting the second preset condition is above-mentioned trip sequence K' value is the trip sequence of A in column expression formula, namely in 8 a.m. to the trip sequence of arrival A in 9 points.

As shown in the above, trip sequence can determine one section of mobile relationship in a period between city zoning, And trip sequence is also corresponding with the mobile frequency.Therefore, it can determine that the corresponding user of user reaches matrix based on trip sequence.Example It such as, is respectively A, B, C and D with the city zoning in city, preset time period is respectively workaday 8 a.m. to 9 points, 9 points To 10 points and 10 points to 11 points.If meeting the trip sequence of the second preset condition using city zoning A as destination Following several types can be divided into：

(1) from B, in 8 a.m. to the trip sequence for being moved to A in 9 points；

(2) from B, the trip sequence of A is moved in 9 points to 10 points of the morning；

(3) from B, the trip sequence of A is moved in 10 points to 11 points of the morning；

(4) from C, in 8 a.m. to the trip sequence for being moved to A in 9 points；

(5) from C, the trip sequence of A is moved in 9 points to 10 points of the morning；

(6) from C, the trip sequence of A is moved in 10 points to 11 points of the morning；

(7) from D, in 8 a.m. to the trip sequence for being moved to A in 9 points；

(8) from D, the trip sequence of A is moved in 9 points to 10 points of the morning；

(9) from D, the trip sequence of A is moved in 10 points to 11 points of the morning.

Wherein, above-mentioned every kind of trip sequence is corresponding with the mobile frequency.It should be noted that during actual implementation, on The quantity for stating every kind of trip sequence may more than one.For example, user from B in 8 a.m. to A is moved in 9 points, connect From A and in 8 a.m. to being moved to B in 9 points, it is last again from B and in 8 a.m. to being moved to A in 9 points.This When, it may appear that two from B, in 8 a.m. to the trip sequence for being moved to A in 9 points.It, can be two by this for the situation The corresponding mobile frequency of trip sequence is added, and merges into a trip sequence.

By the content of above-described embodiment it is found that the row that user reaches matrix can be referred to this if destination is city zoning A User is moved to A from which other city zoning, and user's matrix column that sets out can be referred to different preset time periods.In conjunction with User reach matrix definition and it is above-mentioned about trip sequence explanation, according to meet the second preset condition trip sequence, It can determine that the corresponding user of user reaches matrix.It should be noted that if destination is other city zonings, it equally can be by The corresponding trip sequence for meeting the second preset condition of each user is determined according to the above process, and then determines the user of each user Set out matrix, and details are not described herein again.By the above process, it may finally obtain the case where different cities subregion is destination Under, the corresponding user of each user reaches matrix.

In practical application scene, the mobile terminal that user uses would generally generate signaling data.Wherein, signaling data master It include the generation moment of the Base Station Identification communicated with mobile terminal and signaling data.It follows that due to base station Position is the region that covers to another base station of zone migration determining, that user may cover from a base station, and base station The region of covering had with city zoning it is Chong Die, so that it is the motion track that can reflect user that signaling data, which is formed by sequence, 's.Content based on above description and above-described embodiment, as a kind of alternative embodiment, for any user, from the user Before determining the trip sequence for meeting the first preset condition in all trip sequences within a preset period of time, or from the use Before determining the trip sequence for meeting the second preset condition in all trip sequences of family within a preset period of time, this may further determine that All trip sequences of user within a preset period of time.

Correspondingly, the side of all trip sequences the embodiment of the invention provides a kind of determining user within a preset period of time Method, including but not limited to：For any user, based on each corresponding city zoning in base station in city and the user default Base station mobile sequence in period determines all trip sequences of any user within a preset period of time.

Due to above content it is found that the region of base station covering might have with city zoning and partly overlap, hence for times One base station, can be using the city zoning being overlapped with base station coverage area city zoning corresponding as the base station.In addition, base station is moved Dynamic sequence can be determined by the signaling data that mobile terminal generates, and signaling data arranged in sequence is such as generated the mobile sequence in base station Column.Base station mobile sequence can include at least user identifier, the Base Station Identification as departure place, the base as departure place of the user Letter corresponding to signaling data generation moment, Base Station Identification as a purpose and the base station as a purpose for standing corresponding Data are enabled to generate the moment.Wherein, the corresponding user for using mobile terminal of user identifier.For example, base station mobile sequence can be according to such as Under type indicates：

(user_id,(cell_id₀,time₀),(cell_id₁,time₁))

Wherein, user_id is user identifier, cell_id₀For the Base Station Identification as departure place, cell_id₁For as mesh Ground Base Station Identification, time₀Moment, time are generated for signaling data corresponding to the base station as departure place₁For as a purpose Signaling data corresponding to the base station on ground generates the moment.Wherein, (cell_id₀,time₀) for corresponding to the base station as departure place Signaling data, (cell_id₁,time₁) it is signaling data corresponding to base station as a purpose.

Since base station mobile sequence can reflect user's motion track between the base stations, and deposited between base station and city zoning In corresponding relationship, so as to which by according to the corresponding relationship between base station and city zoning, base station mobile sequence is mapped to trip Sequence.

Method provided in an embodiment of the present invention, for any user, by based on the corresponding city in base station each in city The base station mobile sequence of subregion and any user within a preset period of time determines any user within a preset period of time all Trip sequence.Since signaling data can reflect the motion track of user, thus in conjunction with the mobile sequence in base station determined by signaling data Column, to determine the trip sequence of user, can abundant digging utilization signaling data, and then subsequent determining city zoning type can be improved When accuracy.

Content based on the above embodiment, as a kind of alternative embodiment, based on the corresponding city in base station each in city The base station mobile sequence of city's subregion and any user within a preset period of time determines the institute of any user within a preset period of time Before having trip sequence, the corresponding city zoning in each base station may further determine that.Referring to fig. 2, the embodiment of the present invention is not every to determination The mode of the corresponding city zoning in one base station specifically limits, including but not limited to：

201, it determines the corresponding Thiessen polygon in each base station in city, includes a base in each Thiessen polygon It stands.

In 201, if city is considered as a space plane, each base station all can be considered on the space plane in city Point.Based on Thiessen polygon graphing method, which can be divided into multiple Thiessen polygons.Wherein, each Tyson is more Side shape includes a base station.Thiessen polygon is a kind of subdivision to space plane, its main feature is that any position in polygon It sets nearest with a distance from the sampling point (namely base sites) from the polygon, remote with a distance from sampling point in adjacent polygons and each more Side shape includes and only comprising a sampling point.

202, the city zoning for covering each Thiessen polygon, and the city zoning that each Thiessen polygon will be covered are determined As the corresponding city zoning in each base station.

It is considered as space plane due to city zoning and by city, is obtained after being divided according to road net data to city , so that Thiessen polygon can have Chong Die part with city zoning.As shown in figure 3, being located in the middle hexagon in Fig. 3 i.e. For a Thiessen polygon 1 in space plane, and the rectangle of dotted portion is city zoning in Fig. 3.From the figure 3, it may be seen that 3 dotted rectangles can just cover Thiessen polygon in Fig. 3, so that 3 city zonings are that covering Tyson is more The city zoning of side shape 1.Accordingly for the corresponding base station of Tyson shape changeable 1, above-mentioned 3 city zoning is the base station Corresponding city zoning.

Method provided in an embodiment of the present invention passes through the corresponding Thiessen polygon in base station each in determining city.Determination is covered The city zoning of each Thiessen polygon is covered, and the city zoning for covering each Thiessen polygon is corresponding as each base station City zoning.It, will since the city zoning of covering Thiessen polygon is the city zoning nearest from base station in Thiessen polygon The city zoning of Thiessen polygon is covered as the corresponding city zoning of user departure place and destination, prediction trip sequence can be improved Accuracy when column, and then accuracy when can be improved subsequent determining city zoning type.

By the content of above-described embodiment it is found that signaling data reflection is which base station user is in different time points Coverage area, base station mobile sequence can be generated by signaling data.Signaling data is ranked up according to the moment is generated, for Two signaling datas of arbitrary neighborhood after sequence may not be able to ensure in adjacent two signaling datas corresponding time In section, user is exactly in moving condition.The embodiment of the present invention is needed according to by signaling data base station mobile sequence generated Determine the trip sequence of user, and each trip sequence indicates one section of mobile relationship namely user between city zoning It is centainly in moving condition, hence for based on signaling data base station mobile sequence generated, therefore, to assure that the base station User is in moving condition in the mobile sequence corresponding period.For the demand, for any user, the embodiment of the present invention Additionally provide a kind of mode for determining the base station mobile sequence of the user within a preset period of time.Referring to fig. 4, including it is but unlimited In：

401, each signaling data of the user is successively sorted according to the generation moment of each signaling data, is obtained The base station trip sequence of the user within a preset period of time.

Specifically, obtaining base station trip sequence can indicate as follows：

(user_id,(cell_id₀,time₀),(cell_id₁,time₁)...)

Wherein, user_id is the user identifier of the user, (cell_id₀,time₀) and (cell_id₁,time₁) be Signaling data.cell_id₀And cell_id₁It is Base Station Identification.time₀For signaling data (cell_id₀,time₀) generation when It carves, time₁For signaling data (cell_id₁,time₁) the generation moment.Subsequent ellipsis indicates in base station trip sequence Other signaling datas.The sequence it should be noted that base station in 401 in preset time period is gone on a journey refers to base station trip sequence In each signaling data the generation moment within a preset period of time.

402, the subsequence for meeting third preset condition is chosen from base station trip sequence, and using subsequence as any use The base station mobile sequence of family within a preset period of time；Wherein, the generation of each signaling data is included at least in each signaling data Moment and own base station mark, third preset condition are that the difference at any two adjacent signaling data corresponding generation moment is small In the first preset threshold and user moving speed within a preset range.

In order to make it easy to understand, with the adjacent signaling data of any two for (cell_id₀,time₀) and (cell_id₁, time₁) for, 402 process is explained.Wherein, (cell_id₀,time₀) and (cell_id₁,time₁) corresponding The generation moment difference be (time₁-time₀).Due to can be determined which base station based on Base Station Identification, and base station location is It is predetermined, thus from base station cell_id₀To cell_id₁, user moving speed can be according to the distance between two base stations difference Value and the difference at above-mentioned generation moment are calculated.

After user moving speed is calculated, it can determine that whether user moving speed and the difference at above-mentioned generation moment meet Third preset condition.Wherein, the first preset threshold can be 1 hour in third preset condition, and preset range can be per hour 2 Kilometer is between 150 kilometers per hour, and the present invention is not especially limit this.It, can if meeting third preset condition Certain user is by base station cell_id₀It is moved to base station cell_id₁, namely it is believed that base station cell_id₀With base station cell_ id₁Between be continuous.

After having executed the above process, base station cell_id can be then judged in the same manner₁With base station cell_id₂ Between whether be continuous.If the two equally meets third preset condition, base station cell_id can determine₀, base station cell_id₁And Base station cell_id₂It is continuous between three.If judging base station cell_id₂With base station cell_id₃Between be not it is continuous, then First subsequence for meeting third preset condition is selected in the sequence that can go on a journey from base station, which can table in the following way Show：

(user_id,(cell_id₀,time₀),(cell_id₁,time₁),(cell_id₂,time₂))

Wherein, user_id is the user identifier of the user.For the subsequence, it can be understood as user is by base station cell_ id₀Position start to move, and be moved to base station cell_id₂Position stop.Therefore, cell_id₀For the base as departure place Station identifications, cell_id₂For Base Station Identification as a purpose.

From base station trip sequence in select the subsequence that first meets third preset condition after, due to having judged base station cell_id₂With base station cell_id₃Between be not it is continuous, then can be by base station cell_id₃It is considered as the base station as departure place Mark, and continue to judge base station cell_id in base station trip sequence₃With base station cell_id₄It whether is continuous, and according to above-mentioned Process traverses each signaling data in base station trip sequence one by one, until all signaling datas traversal finishes.On executing Process is stated, several subsequences for meeting third preset condition are selected in the sequence that can finally go on a journey from base station, if so as to should Dry meets base station mobile sequence of the subsequence of third preset condition as the user within a preset period of time.

Method provided in an embodiment of the present invention, for any user, by by each signaling data of the user according to every The generation moment of one signaling data is successively sorted, and the base station trip sequence of the user within a preset period of time is obtained.From base The subsequence for meeting third preset condition is chosen in trip sequence of standing, and within a preset period of time using subsequence as any user Base station mobile sequence.The subsequence for being in moving condition due to choosing user in the sequence that can go on a journey from base station, and as base station Mobile sequence, thus can ensure that in the base station mobile sequence corresponding period user in moving condition, and then it is subsequent When determining trip sequence based on base station mobile sequence, the accuracy of trip sequence can be improved.

Content based on the above embodiment, as a kind of alternative embodiment, for any user, the embodiment of the present invention is not right Base station mobile sequence based on the corresponding city zoning in base station each in city and any user within a preset period of time determines The mode of all trip sequences of the user within a preset period of time specifically limits, including but not limited to：By first kind city Subregion and the second class city zoning carry out any combination, obtain all trip sequences of any user within a preset period of time；Its In, first kind city zoning is city zoning corresponding to the base station in the mobile sequence of base station as departure place, the second class city Subregion is city zoning corresponding to the base station in any base station mobile sequence as a purpose.

By the content of above-described embodiment it is found that including Base Station Identification and the conduct as departure place in the mobile sequence of base station The Base Station Identification of destination.And each base station is corresponding with Thiessen polygon, Thiessen polygon corresponding for any base station, it should The corresponding city zoning in base station refers to covering the city zoning of the Tyson shape changeable.As shown in figure 3, Thiessen polygon 1 is corresponding City zoning have 3.Due to the corresponding city zoning of Thiessen polygon might have it is multiple, thus based on the mobile sequence in base station When column determine trip sequence, even if being uniquely determined as the base station of departure place and base station as a purpose, but still not The city zoning and city zoning as a purpose as departure place can be uniquely determined.

Above-mentioned any combination obtains the process of trip sequence in order to facilitate understanding, using in the mobile sequence of base station as departure place The corresponding Thiessen polygon in base station be TPi, the corresponding Thiessen polygon in base station in the mobile sequence of base station as a purpose is The corresponding city zoning of TPj, Thiessen polygon Tpi is R1 and R2, and the corresponding city zoning of Thiessen polygon Tpj is R3 and R4 is Example, the process of above-mentioned any combination can refer to Fig. 5.As shown in figure 5, trip sequence can be respectively (R1, R3), (R1, R4), (R2, R3) and (R2, R4).Wherein, R1 and R2 is the city zoning as departure place, and R3 and R4 are city point as a purpose Area.

By the content of above-described embodiment it is found that each trip sequence is corresponding with the mobile frequency.Based on each in city The base station mobile sequence of the corresponding city zoning in base station and any user within a preset period of time determines any user default After all trip sequences in period, the corresponding mobile frequency of each trip sequence can be also calculated.Correspondingly, based on above-mentioned The content of embodiment, as a kind of alternative embodiment, the embodiment of the invention also provides a kind of corresponding shiftings of calculating trip sequence The method of the dynamic frequency, including but not limited to：For any trip sequence, first part's Thiessen polygon and the first city point are calculated The first area ratio between area calculates the second area ratio between second part Thiessen polygon and the second city zoning, Using the mobile frequency corresponding as institute's trip sequence of the product between the first area ratio and second area ratio；Wherein, One city zoning is the city zoning in the trip sequence as departure place, and the second city zoning is that mesh is used as in the trip sequence Ground city zoning；The part Thiessen polygon that first part's Thiessen polygon includes by the first city zoning, second The part Thiessen polygon that point Thiessen polygon includes by the second city zoning.

In order to make it easy to understand, for (R1, R3), the mistake of the frequency is moved to above-mentioned calculating to go on a journey sequence now in conjunction with Fig. 5 Journey is illustrated.In the first city zoning (namely city zoning R1), first part's Thiessen polygon (namely Thiessen polygon The partial region of Tpi) it is located in R1, to calculate first between the partial region of Thiessen polygon Tpi and city zoning R1 Area ratio.Similarly, the second area ratio between the partial region and city zoning R3 that calculate Thiessen polygon Tpj can also be calculated Value, and be (R1, R3) corresponding mobile frequency using the product between the first area ratio and second area ratio as trip sequence It is secondary.

It should be noted that the mobile frequency its value is calculated by the above process is less than 1.In the embodiment of the present invention The corresponding mobile frequency of the trip sequence being calculated is not that the frequency of being gone on a journey by actual count user obtains, namely simultaneously The non-number for really indicating the corresponding mobile relationship of trip sequence and actually really occurring, but being more focused on indicates each trip Sequence is in determining multiple trip sequences, a possibility that corresponding mobile relationship really occurs.

In addition, it should also be noted that, by the content of above-described embodiment it is found that determining that each city zoning is corresponding pre- If when type, needing the problem of converting determining document subject matter for the problem of area type for determining city zoning.Wherein, city Subregion can be considered document, and the corresponding mobile relationship of each trip sequence can be considered the word in document, and sequence of going on a journey is corresponding The mobile frequency can be considered the frequency that word occurs in a document.Based on above description, due to the word appearance in the case where routinely understanding The frequency is integer, and for the ease of handling data, the embodiment of the present invention is being calculated through the above way less than 1 After the mobile frequency, integer, this hair can also be converted by the corresponding mobile frequency of each trip sequence according to unified mode Bright embodiment is not especially limited this.For example, transform mode can be directly multiplied by 100 and to be rounded the mobile frequency.

Content based on the above embodiment, as a kind of alternative embodiment, specified parameter further includes each city zoning pair The feature vector answered；Wherein, the corresponding feature vector of each city zoning is determined by the point of interest in each city zoning 's.The format of point of interest can be (interest point name, point of interest location, point of interest category), and point of interest category can be according to function It can divide, such as be divided into sight spot, shop, school, company and residential quarter, the present invention is not especially limit this.

It should be noted that by the content of above-described embodiment it is found that city zoning can be considered document.And implement in the present invention In example, the corresponding feature vector of city zoning can be considered as document meta information, such as document author or document Publication Year it is non- Word feature.In the above-described embodiments, if preset model is LDA model, without introducing text in document subject matter determination process Shelves metamessage, causes final definitive result not accurate enough.And in embodiments of the present invention, specified parameter further includes each city point The corresponding feature vector in area.Correspondingly, the preset model being related in above-described embodiment can also be DMR (Dirichlet- Multinomial Regression, Di Li Cray polynomial regression) model.By introducing each city point in specified parameter The corresponding feature vector in area, to determine the preset kind of each city zoning in conjunction with the point of interest in city zoning, to make It is more accurate to obtain definitive result.

Content based on the above embodiment, as a kind of alternative embodiment, if further including each city point in specified parameter The corresponding feature vector in area may further determine that each city zoning is corresponding then before specified parameter is input to preset model Feature vector.Correspondingly, the embodiment of the present invention does not limit the mode of the corresponding feature vector of each city zoning of determination specifically It is fixed, including but not limited to：The area of quantity and each city zoning based on each type point of interest in each city zoning, Determine the corresponding feature vector of each city zoning.

Wherein, for any city zoning, the corresponding feature vector of the city zoning can be directly by every in the city zoning The quantity of seed type point of interest and the area of the city zoning are constituted, and expression can be specifically carried out as follows：

f_i=(pf_i1,pf_i2,...,pf_iC,Area(i))

Wherein, f_iFor the corresponding feature vector of i-th of city zoning.pf_i1It is emerging for the first kind in i-th of the city zoning The quantity of interest point, pf_i2For the quantity ... ... of the second class point of interest in i-th of the city zoning, pf_icFor i-th of city point The quantity of C class point of interest in area, Area (i) are minimum in the corresponding area of i-th of city zoning and all city zonings Ratio between area.

Content based on the above embodiment, as a kind of alternative embodiment, the embodiment of the present invention is not to based on each city The quantity of each type point of interest and the area of each city zoning in subregion, determine the corresponding feature of each city zoning to The mode of amount specifically limits.Referring to Fig. 6, including but not limited to：

601, the area of quantity and each city zoning based on each type point of interest in each city zoning determines The corresponding initial characteristics vector of each city zoning.

Specifically, the area of quantity and each city zoning based on each type point of interest in each city zoning, It can first determine the corresponding feature vector to be selected of each city zoning.Feature vector to be selected can be carried out as follows expression：

f_i=(pf_i1,pf_i2,...,pf_iC,Area(i))

Wherein, f_iFor the corresponding feature vector to be selected of i-th of the city zoning.pf_i1It is in i-th of the city zoning first The quantity of class point of interest, pf_i2For the quantity ... ... of the second class point of interest in i-th of the city zoning, pf_icFor i-th of city The quantity of C class point of interest in city's subregion, Area (i) are in the corresponding area of i-th of city zoning and all city zonings Ratio between minimum area.

It, can be by the corresponding feature to be selected of each city zoning after obtaining the corresponding feature vector to be selected of each city zoning Vector be converted to the corresponding TF-IDF of each city zoning (TermFrequency-Inverse Document Frequency, Word frequency-inverse document frequency) vector namely initial characteristics vector.Wherein, the corresponding TF-IDF vector of each city zoning can It is carried out as follows expression：

TFIDF_i=(tfidf_i1,tfidf_i2,...,tfidf_ic,Area(i))

Wherein, TFIDF_iFor the corresponding TFIDF vector of i-th of the city zoning.tfidf_i1For in i-th of the city zoning The significance level of first kind point of interest, tfidf_i2For the significance level ... ... of the second class point of interest in i-th of the city zoning, tfidf_icFor the significance level of C class point of interest in i-th of the city zoning, Area (i) is that i-th of the city zoning is corresponding Ratio in area and all city zonings between minimum area.

For the significance level tfidf of jth class point of interest in i-th of the city zoning_ij, specifically refer to as under type into Row indicates：

Wherein, tfidf_ijFor the significance level of jth class point of interest in i-th of city zoning, pf_ijFor i-th of city point The quantity of jth class point of interest in area, M are the total quantity of city zoning, and Count (j) is emerging comprising jth class in all city zonings The city zoning quantity of interest point.

602, the corresponding initial characteristics vector of each city zoning is spliced, obtains all city zonings corresponding One matrix carries out singular value decomposition to the first matrix, obtains the second matrix；Wherein, the second matrix is corresponded to by each city zoning Feature vector formed.

It, can will be by the corresponding initial spy of each city zoning after obtaining the corresponding initial characteristics vector of each city zoning Sign vector is spliced, and also splices the corresponding TF-IDF vector of each city zoning, so that M* (C+1) dimension can be obtained The first matrix.By utilizing SVD (Singular Value Decomposition, singular value decomposition) matrix to the first matrix It is decomposed, the second matrix X=(x of M*F can be obtained₁,...x_M)^T.Wherein, x_iIndicate the corresponding feature of i-th of city zoning Vector.

By the content of above-described embodiment it is found that preset model output be each city zoning be every kind of preset kind when Probability distribution.Content based on the above embodiment, as a kind of alternative embodiment, it is corresponding that specified parameter may also include feature vector Weight.Correspondingly, the embodiment of the present invention is not input to preset model to by specified parameter, is made with obtaining the process of output result It is specific to limit, including but not limited to：Specified parameter is input to preset model, exports each city zoning and every kind of preset kind Between probability distribution, weight corresponding to the corresponding feature vector of each city zoning is adjusted based on gradient descent method, is laid equal stress on Each city zoning is newly corresponded into weight adjusted and specified parameter is input to preset model, and exports each city point again Probability distribution between area and every kind of preset kind repeats above-mentioned weight adjustment process and preset model input/output procedure, Until the probability distribution between each city zoning and every kind of preset kind of output tends towards stability.By the above process, can make The output result for obtaining preset model is more accurate.

Method provided in an embodiment of the present invention, by quantity based on each type point of interest in each city zoning and The area of each city zoning determines the corresponding initial characteristics vector of each city zoning.Each city zoning is corresponding just Beginning feature vector is spliced, and corresponding first matrix of all city zonings is obtained, and is carried out singular value decomposition to the first matrix, is obtained To the second matrix.Since the corresponding feature vector of each city zoning can be introduced in specified parameter, to combine in city zoning Point of interest determine the preset kind of each city zoning so that definitive result is more accurate.

If the quantity of city zoning can be relatively more in view of city zoning division is too small, lead to statistical result excessively Dispersion and it is not significant enough.For the situation, content based on the above embodiment is each obtaining as a kind of alternative embodiment Before city zoning corresponding set out matrix and arrival matrix, city zoning can also be merged.Correspondingly, the present invention is implemented Example does not limit specifically to the mode for merging city zoning, including but not limited to：If existing in city zoning set Area then merges the city zoning in city zoning set less than the city zoning of the second preset threshold, until city The area of each city zoning is all larger than the second preset threshold in partitioned set；It wherein, include being based on road in city zoning set Multiple city zonings that network data obtains after dividing city.

Wherein, the mode merged two-by-two can be used to close the multiple city zonings obtained after dividing based on road net data And the smallest city zoning of area and the small city zoning of area time is such as selected to merge every time from all city zonings, Until the area of each city zoning is all larger than the second preset threshold in city zoning set.

Method provided in an embodiment of the present invention, by city zoning set, there are areas less than the second preset threshold City zoning then merges the city zoning in city zoning set, until each city zoning in city zoning set Area be all larger than the second preset threshold.Due to that can be merged to the lesser city zoning of area in all city zonings, from And make subsequent statistical result more significantly and more targeted.

In view of connection can be even closer between the city zoning in all city zonings with common edge, and common edge is got over It is long then contact closer.Content based on the principle and above-described embodiment, as a kind of alternative embodiment, about to city zoning City zoning in set merges, until the area of each city zoning is all larger than the second default threshold in city zoning set The mode of value, the present invention is not especially limit this, including but not limited to：It is chosen from city zoning set and meets the The 4th city zoning for meeting the 4th preset condition is chosen in the third city zoning of three preset conditions from city zoning set, Third city zoning and the 4th city zoning are merged, city zoning set is retrieved, repeat above-mentioned selection and Merging process, until the area of each city zoning is all larger than the second preset threshold in city zoning set；Wherein, third is default Condition is that area is minimum and area is less than the second preset threshold, and the 4th preset condition is and and third adjacent with third city zoning Public side length longest between city zoning.

Specifically, the second preset threshold A can first be selected_th, and by city zonings all in city zoning set according to area It is ranked up, can specifically be denoted as from big to small：

FromThe middle the smallest city zoning b of selection area_iIf city zoning b_iArea be greater than the second preset threshold the Two preset threshold A_th, then terminate selection and merging process.If city zoning b_iArea it is pre- less than the second preset threshold second If threshold value A_th, then fromMiddle determination and city zoning b_iAdjacent city zoning b_ij, and calculate city zoning b_iWith city point Area b_ijBetween perimeter and public side length ratio, specifically refer to following formula：

Wherein, lc_ijFor city zoning b_iWith city zoning b_ijBetween public side length, lp_iFor city zoning b_iPerimeter, lp_ijFor city zoning b_ijPerimeter, pri_ijFor city subregion b_iWith city zoning b_ijBetween perimeter and public side length ratio.It needs It is noted that pri_ijValue it is bigger, then illustrate city zoning b_iWith city zoning b_ijBetween common edge it is longer.

Correspondingly, maximum pri can be chosen_ijCorresponding city zoning b_ij, and by city zoning b_iWith city zoning b_ijIt carries out Merge.Above-mentioned selection and merging process are repeated, until the area of each city zoning is all larger than the in city zoning set Two preset threshold A_th。

Method provided in an embodiment of the present invention meets the third of third preset condition by choosing from city zoning set City zoning, from city zoning set choose meet the 4th preset condition the 4th city zoning, by third city zoning with 4th city zoning merges, and retrieves city zoning set, repeats above-mentioned selection and merging process, until city The area of each city zoning is all larger than the second preset threshold in partitioned set.Due to can be smaller to area in all city zonings City zoning merge so that subsequent statistical result is more significant and more targetedly.

It should be noted that above-mentioned all alternative embodiments, can form optional implementation of the invention using any combination Example, this is no longer going to repeat them.

Content based on the above embodiment, the embodiment of the invention provides a kind of area type determining device, the region classes Type determining device determines method for executing the area type provided in above method embodiment.Referring to Fig. 7, which includes：

Module 701 is obtained, for obtaining corresponding matrix and the arrival matrix of setting out of each city zoning；Wherein, city point Area is obtained after dividing to city；For any city zoning in city, the corresponding square that sets out of any city zoning Battle array includes the mobile frequency that all users are respectively moved to each other city zonings from any city zoning within a preset period of time Secondary, the corresponding arrival matrix of any city zoning includes all users within a preset period of time respectively from each other city zonings It is moved to the mobile frequency of any city zoning, other city zonings are the city point in city in addition to any city zoning Area；

First determining module 702, for specified parameter to be input to preset model, the output result based on preset model is true Determine the corresponding preset kind of each city zoning；Wherein, parameter is specified to include at least preset kind sum, each city zoning pair Set out matrix and the arrival matrix answered.

As a kind of alternative embodiment, module 701 is obtained, including：

First acquisition unit is set out matrix for for any city zoning, obtaining the corresponding user of each user；Its In, for any user in all users, the corresponding user of any user matrix that sets out includes any user in preset time The mobile frequency of each other city zonings is respectively moved in section from any city zoning；

First superpositing unit obtains any city point for the corresponding user of each user matrix that sets out to be overlapped The corresponding matrix that sets out in area.

As a kind of alternative embodiment, module 701 is obtained, including：

Second acquisition unit, for obtaining the corresponding user of each user and reaching matrix for any city zoning；Its In, for any user in all users, it includes any user in preset time that the corresponding user of any user, which reaches matrix, It is moved to the mobile frequency of any city zoning in section from each other city zonings respectively；

Second superpositing unit is overlapped for the corresponding user of each user to be reached matrix, obtains any city point The corresponding arrival matrix in area.

As a kind of alternative embodiment, first acquisition unit is used for for any user, from any user in preset time The trip sequence for meeting the first preset condition is determined in all trip sequences in section；Wherein, the first preset condition is with any City zoning is departure place and using other city zonings as destination, and each trip sequence is corresponding with the mobile frequency；According to full The trip sequence of the first preset condition of foot determines that the corresponding user of any user sets out matrix.

As a kind of alternative embodiment, second acquisition unit is used for for any user, from any user in preset time The trip sequence for meeting the second preset condition is determined in all trip sequences in section；Wherein, the second preset condition is with other City zoning is departure place and using any city zoning as destination, and each trip sequence is corresponding with the mobile frequency；According to full The trip sequence of the second preset condition of foot determines that the corresponding user of any user reaches matrix.

As a kind of alternative embodiment, which further includes：

Second determining module, for based on the corresponding city zoning in base station each in city and appointing for any user The base station mobile sequence of one user within a preset period of time determines all trip sequences of any user within a preset period of time.

As a kind of alternative embodiment, which further includes：

Third determining module, for determining the corresponding Thiessen polygon in each base station in city, in each Thiessen polygon It include a base station；

4th determining module, for determining the city zoning for covering each Thiessen polygon, and it is more to cover each Tyson The city zoning of side shape is as the corresponding city zoning in each base station.

As a kind of alternative embodiment, which further includes：

Sorting module is used for for any user, by each signaling data of any user according to each signaling data The generation moment is successively sorted, and the base station trip sequence of any user within a preset period of time is obtained；

Module is chosen, meets the subsequence of third preset condition for choosing from base station trip sequence, and by subsequence As the base station mobile sequence of any user within a preset period of time；Wherein, each signaling is included at least in each signaling data The generation moment of data and own base station mark, third preset condition are the adjacent corresponding generation of signaling data of any two The difference at moment less than the first preset threshold and user moving speed within a preset range.

As a kind of alternative embodiment, the second determining module is used for first kind city zoning and the second class city zoning Any combination is carried out, all trip sequences of any user within a preset period of time are obtained；Wherein, first kind city zoning is base City zoning corresponding to the base station stood in mobile sequence as departure place, the second class city zoning are any base station mobile sequence In city zoning corresponding to base station as a purpose.

As a kind of alternative embodiment, which further includes：

Computing module, for for any trip sequence, calculate first part's Thiessen polygon and the first city zoning it Between the first area ratio, the second area ratio between second part Thiessen polygon and the second city zoning is calculated, by the Product between one area ratio and second area ratio is as the corresponding mobile frequency of any trip sequence；

Wherein, the first city zoning is the city zoning in any trip sequence as departure place, and the second city zoning is City zoning in any trip sequence as a purpose；The portion that first part's Thiessen polygon includes by the first city zoning Divide Thiessen polygon, the part Thiessen polygon that second part Thiessen polygon includes by the second city zoning.

As a kind of alternative embodiment, specified parameter further includes the corresponding feature vector of each city zoning；Wherein, each The corresponding feature vector of city zoning is determined by the point of interest in each city zoning.

As a kind of alternative embodiment, which further includes：

5th determining module, for based on each type point of interest in each city zoning quantity and each city point The area in area determines the corresponding feature vector of each city zoning.

As a kind of alternative embodiment, the 5th determining module, for based on each type point of interest in each city zoning Quantity and each city zoning area, determine the corresponding initial characteristics vector of each city zoning；By each city point The corresponding initial characteristics vector in area is spliced, and corresponding first matrix of all city zonings is obtained, and is carried out to the first matrix odd Different value is decomposed, and the second matrix is obtained；Wherein, the second matrix is made of the corresponding feature vector of each city zoning.

As a kind of alternative embodiment, which further includes：

Merging module, for when in city zoning set there are area less than the second preset threshold city zoning when, then City zoning in city zoning set is merged, until the area of each city zoning is all larger than in city zoning set Second preset threshold；It wherein, include the multiple cities point obtained after being divided city based on road net data in city zoning set Area.

As a kind of alternative embodiment, merging module meets third preset condition for choosing from city zoning set Third city zoning, from city zoning set choose meet the 4th preset condition the 4th city zoning, by third city Subregion is merged with the 4th city zoning, retrieves city zoning set, repeats above-mentioned selection and merging process, directly Into city zoning set, the area of each city zoning is all larger than the second preset threshold；Wherein, third preset condition is area Minimum and area less than the second preset threshold, the 4th preset condition be it is adjacent with third city zoning and with third city zoning it Between public side length longest.

Device provided in an embodiment of the present invention, by obtaining each city zoning corresponding set out matrix and arrival matrix, Specified parameter is input to preset model, the corresponding default class of each city zoning is determined based on the output result of preset model Type.Since the corresponding region class of city zoning can be automatically determined based on mobile data of the user between the subregion of different cities Type, so that it is determined that result is more accurate.In addition, introducing time dimension due to when determining the corresponding area type of city zoning Degree, is for same city zoning, it may be determined that the corresponding area type of the city zoning in different time sections, so as to more Gear to actual circumstances application scenarios.

Secondly, for any user, by being existed based on the corresponding city zoning in base station each in city and any user Base station mobile sequence in preset time period determines all trip sequences of any user within a preset period of time.Due to signaling Data can reflect the motion track of user, thus in conjunction with the base station mobile sequence determined by signaling data, to determine going out for user Row sequence, can abundant digging utilization signaling data, and then accuracy when can be improved subsequent determining city zoning type.

Again, pass through the corresponding Thiessen polygon in base station each in determining city.It determines and covers each Thiessen polygon City zoning, and the city zoning of each Thiessen polygon will be covered as the corresponding city zoning in each base station.Due to covering The city zoning of Thiessen polygon is the city zoning nearest from base station in Thiessen polygon, will cover the city of Thiessen polygon As the corresponding city zoning of user departure place and destination accuracy when prediction trip sequence can be improved, in turn in city's subregion Accuracy when subsequent determining city zoning type can be improved.

From secondary, for any user, by by each signaling data of the user according to each signaling data generation when Quarter is successively sorted, and the base station trip sequence of the user within a preset period of time is obtained.It is chosen from base station trip sequence full The subsequence of sufficient third preset condition, and the base station mobile sequence using subsequence as any user within a preset period of time.By The subsequence that user is in moving condition is chosen in the sequence that can go on a journey from base station, and as base station mobile sequence, so as to true It protects user in base station mobile sequence corresponding period and is in moving condition, and then is subsequent to be based on base station mobile sequence true When making row sequence, the accuracy of trip sequence can be improved.

In addition, passing through the face of quantity and each city zoning based on each type point of interest in each city zoning Product, determines the corresponding initial characteristics vector of each city zoning.The corresponding initial characteristics vector of each city zoning is spelled It connects, obtains corresponding first matrix of all city zonings, singular value decomposition is carried out to the first matrix, obtains the second matrix.Due to The corresponding feature vector of each city zoning can be introduced in specified parameter, it is every to be determined in conjunction with the point of interest in city zoning The preset kind of one city zoning, so that definitive result is more accurate.

Finally, by the city zoning in city zoning set there are area less than the second preset threshold, then to city City zoning in partitioned set merges, until the area of each city zoning is all larger than second in advance in city zoning set If threshold value.Due to that can be merged to the lesser city zoning of area in all city zonings, so that subsequent statistical result It is more significant and more targeted.

The embodiment of the invention provides a kind of electronic equipment.Referring to Fig. 8, which includes：Processor (processor) 801, memory (memory) 802 and bus 803；

Wherein, processor 801 and memory 802 complete mutual communication by bus 803 respectively；Processor 801 is used In calling the program instruction in memory 802, method is determined to execute area type provided by above-described embodiment, such as wrap It includes：Obtain corresponding matrix and the arrival matrix of setting out of each city zoning；Wherein, city zoning is obtained after dividing to city It arrives；For any city zoning in city, the corresponding matrix that sets out of any city zoning includes all users when default Between the mobile frequencys of each other city zonings is respectively moved in section from any city zoning, any city zoning is corresponding to arrive It include the shifting that all users are moved to any city zoning from each other city zonings respectively within a preset period of time up to matrix The dynamic frequency, other city zonings are the city zoning in city in addition to any city zoning；Specified parameter is input to default Model determines the corresponding preset kind of each city zoning based on the output result of preset model；Wherein, parameter is specified at least to wrap Include preset kind sum, corresponding matrix and the arrival matrix of setting out of each city zoning.

The embodiment of the present invention provides a kind of non-transient computer readable storage medium, the non-transient computer readable storage medium Matter stores computer instruction, which makes computer execute area type provided by above-described embodiment to determine method, For example including：Obtain corresponding matrix and the arrival matrix of setting out of each city zoning；Wherein, city zoning is drawn to city It is obtained after point；For any city zoning in city, the corresponding matrix that sets out of any city zoning includes that all users exist The mobile frequency of each other city zonings, any city zoning pair are respectively moved in preset time period from any city zoning The arrival matrix answered includes that all users are moved to any city point from each other city zonings respectively within a preset period of time The mobile frequency in area, other city zonings are the city zoning in city in addition to any city zoning；Specified parameter is inputted To preset model, the corresponding preset kind of each city zoning is determined based on the output result of preset model；Wherein, parameter is specified Including at least preset kind sum, corresponding matrix and the arrival matrix of setting out of each city zoning.

Those of ordinary skill in the art will appreciate that：Realize that all or part of the steps of above method embodiment can pass through The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer readable storage medium, the program When being executed, step including the steps of the foregoing method embodiments is executed；And storage medium above-mentioned includes：ROM, RAM, magnetic disk or light The various media that can store program code such as disk.

The embodiments such as electronic equipment described above are only schematical, wherein unit as illustrated by the separation member It may or may not be physically separated, component shown as a unit may or may not be physics list Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness Labour in the case where, it can understand and implement.

Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation Certain Part Methods of example or embodiment.

Finally, the present processes are only preferable embodiment, it is not intended to limit the protection model of the embodiment of the present invention It encloses.With within principle, any modification, equivalent replacement, improvement and so on should be included in all spirit in the embodiment of the present invention Within the protection scope of the embodiment of the present invention.

Claims (17)

1. a kind of area type determines method, which is characterized in that including：

Obtain corresponding matrix and the arrival matrix of setting out of each city zoning；Wherein, the city zoning is drawn to city It is obtained after point；For any city zoning in the city, the corresponding matrix that sets out of any city zoning includes institute There is user to be respectively moved to the mobile frequency of each other city zonings, institute from any city zoning within a preset period of time State the corresponding arrival matrix of any city zoning include all users in the preset time period respectively from each other cities Subregion is moved to the mobile frequency of any city zoning, and other city zonings are that any city is removed in the city City zoning except city's subregion；

Specified parameter is input to preset model, determines that each city zoning is corresponding based on the output result of the preset model Preset kind；Wherein, the specified parameter includes at least preset kind sum, each city zoning corresponding set out and is arrived matrix Up to matrix.

2. the method according to claim 1, wherein described obtain the corresponding matrix that sets out of each city zoning, Including：

For any city zoning, obtains the corresponding user of each user and set out matrix；Wherein, for all users In any user, the corresponding user of any user matrix that sets out includes any user in the preset time period The mobile frequency of each other city zonings is respectively moved to from any city zoning；

The corresponding user of each user matrix that sets out is overlapped, the corresponding matrix that sets out of any city zoning is obtained.

3. the method according to claim 1, wherein described obtain the corresponding arrival matrix of each city zoning, Including：

For any city zoning, obtains the corresponding user of each user and reach matrix；Wherein, for all users In any user, it includes any user in the preset time period that the corresponding user of any user, which reaches matrix, The mobile frequency of any city zoning is moved to from each other city zonings respectively；

The corresponding user of each user is reached matrix to be overlapped, obtains the corresponding arrival matrix of any city zoning.

4. according to the method described in claim 2, it is characterized in that, described obtain the corresponding user of each user and set out matrix, Including：

For any user, determines and meet in all trip sequences in the preset time period from any user The trip sequence of first preset condition；Wherein, first preset condition be by departure place of any city zoning and with Other city zonings are destination, and each trip sequence is corresponding with the mobile frequency；

According to the trip sequence for meeting first preset condition, determine that the corresponding user of any user sets out matrix.

5. according to the method described in claim 3, it is characterized in that, described obtain the corresponding user of each user and reach matrix, Including：

For any user, determines and meet in all trip sequences in the preset time period from any user The trip sequence of second preset condition；Wherein, second preset condition be by departure place of other city zonings and with Any city zoning is destination, and each trip sequence is corresponding with the mobile frequency；

According to the trip sequence for meeting second preset condition, determine that the corresponding user of any user reaches matrix.

6. method according to claim 4 or 5, which is characterized in that described to obtain the corresponding square that sets out of each city zoning Before battle array and arrival matrix, further include：

For any user, based on each corresponding city zoning in base station in the city and any user in institute The base station mobile sequence in preset time period is stated, determines all trip sequences of any user in the preset time period Column.

7. according to the method described in claim 6, it is characterized in that, described based on the corresponding city in base station each in the city The base station mobile sequence of subregion and any user in the preset time period, determines any user described pre- If further including before all trip sequences in the period：

It determines the corresponding Thiessen polygon in each base station in the city, includes a base station in each Thiessen polygon；

It determines the city zoning for covering each Thiessen polygon, and the city zoning of each Thiessen polygon will be covered as each The corresponding city zoning in base station.

8. according to the method described in claim 6, it is characterized in that, described based on the corresponding city in base station each in the city The base station mobile sequence of subregion and any user in the preset time period, determines any user described pre- If further including before all trip sequences in the period：

For any user, by each signaling data of any user according to each signaling data the generation moment into Row successively sequence obtains base station trip sequence of any user in the preset time period；

The subsequence for meeting third preset condition is chosen from base station trip sequence, and is appointed using the subsequence as described Base station mobile sequence of one user in the preset time period；

Wherein, pre- including at least the generation moment of each signaling data and own base station mark, third in each signaling data If condition is the difference at any two adjacent signaling data corresponding generation moment less than the first preset threshold and the mobile speed of user Degree is within a preset range.

9. according to the method described in claim 6, it is characterized in that, described based on the corresponding city in base station each in the city The base station mobile sequence of subregion and any user in the preset time period, determines any user described pre- If all trip sequences in the period, including：

First kind city zoning and the second class city zoning are subjected to any combination, obtain any user when described default Between all trip sequences in section；Wherein, the first kind city zoning is in the base station mobile sequence as departure place City zoning corresponding to base station, the second class city zoning are the base in any base station mobile sequence as a purpose City zoning corresponding to standing.

10. the method according to the description of claim 7 is characterized in that described based on the corresponding city in base station each in the city The base station mobile sequence of city's subregion and any user in the preset time period, determines any user described After all trip sequences in preset time period, further include：

For any trip sequence, the first area ratio between first part's Thiessen polygon and the first city zoning is calculated, The second area ratio between second part Thiessen polygon and the second city zoning is calculated, by first area ratio and institute Product between second area ratio is stated as the corresponding mobile frequency of any trip sequence；

Wherein, first city zoning is the city zoning in any trip sequence as departure place, second city City's subregion is the city zoning in any trip sequence as a purpose；First part's Thiessen polygon is described the The part Thiessen polygon that one city zoning is included, the second part Thiessen polygon are wrapped by second city zoning The part Thiessen polygon contained.

11. the method according to claim 1, wherein the specified parameter further includes that each city zoning is corresponding Feature vector；Wherein, the corresponding feature vector of each city zoning is determined by the point of interest in each city zoning.

12. according to the method for claim 11, which is characterized in that it is described specified parameter is input to preset model before, Further include：

The area of quantity and each city zoning based on each type point of interest in each city zoning, determines each city The corresponding feature vector of subregion.

13. the method according to claim 1, wherein described obtain the corresponding matrix that sets out of each city zoning And before reaching matrix, further include：

If the city zoning in city zoning set there are area less than the second preset threshold, in the city zoning set City zoning merge, until to be all larger than described second default for the area of each city zoning in the city zoning set Threshold value；

It wherein, include the multiple cities point obtained after being divided the city based on road net data in the city zoning set Area.

14. according to the method for claim 13, which is characterized in that the city zoning in the city zoning set It merges, until the area of each city zoning is all larger than second preset threshold in the city zoning set, including：

The third city zoning for meeting third preset condition is chosen from the city zoning set, from the city zoning set It is middle to choose the 4th city zoning for meeting the 4th preset condition, the third city zoning and the 4th city zoning are carried out Merge, retrieve the city zoning set, repeat above-mentioned selection and merging process, until the city zoning set In the area of each city zoning be all larger than second preset threshold；

Wherein, third preset condition is that area is minimum and area is less than second preset threshold, the 4th preset condition for institute State third city zoning adjoining and the public side length longest between the third city zoning.

15. a kind of area type determining device, which is characterized in that including：

Module is obtained, for obtaining corresponding matrix and the arrival matrix of setting out of each city zoning；Wherein, the city zoning is It is obtained after being divided to city；For any city zoning in the city, any city zoning is corresponding out Sending out matrix includes that all users are respectively moved to each other city zonings from any city zoning within a preset period of time The mobile frequency, the corresponding arrival matrix of any city zoning include all users in the preset time period respectively from Each other city zonings are moved to the mobile frequency of any city zoning, and other city zonings are in the city City zoning in addition to any city zoning；

First determining module is determined for specified parameter to be input to preset model based on the output result of the preset model The corresponding preset kind of each city zoning；Wherein, the specified parameter includes at least preset kind sum, each city zoning Corresponding matrix and the arrival matrix of setting out.

16. a kind of electronic equipment, which is characterized in that including：

At least one processor；And

At least one processor being connect with the processor communication, wherein：

The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to instruct energy Enough methods executed as described in claim 1 to 14 is any.

17. a kind of non-transient computer readable storage medium, which is characterized in that the non-transient computer readable storage medium is deposited Computer instruction is stored up, the computer instruction makes the computer execute the method as described in claim 1 to 14 is any.