CN110008414A

CN110008414A - The determination method and apparatus of geography information point

Info

Publication number: CN110008414A
Application number: CN201910263239.0A
Authority: CN
Inventors: 程允胜; 吴海山; 汪天一; 许梦雯
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Baidu Online Network Technology Beijing Co Ltd; Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2016-03-31
Filing date: 2016-03-31
Publication date: 2019-07-12
Anticipated expiration: 2036-03-31
Also published as: CN105893537A; CN105893537B; CN110008414B

Abstract

This application discloses the determination method and apparatus of geography information point.One specific embodiment of the method includes: to obtain the location information of user, wherein the location information includes that user positions coordinate；The user is positioned into coordinate as the input value of bayes predictive model trained in advance, and the probability value that the user is in each geography information point at least one geography information point is obtained according to the bayes predictive model, wherein, the bayes predictive model is obtained using the essential information of geography information point as sample data training, wherein, the essential information includes the history location information of the positioning coordinate of geography information point, history visiting user；The corresponding geography information point of most probable value is determined as the point of geography information locating for the user.The embodiment realizes the point of geography information locating for determining user.

Description

The determination method and apparatus of geography information point

It is March 31, entitled " in 2016 that the application, which is application No. is CN201610196304.9, the applying date, The divisional application of the Chinese patent application of the determination method and apparatus of reason information point ".

Technical field

This application involves field of computer technology, and in particular to Internet technical field more particularly to geography information point Determine method and apparatus.

Background technique

Geography information point (POI, Point Of Interest), also known as " information point " or " point of interest ", refer to having The place of definite meaning, such as restaurant, school, parking lot.The positioning of the prior art especially positions user, both for User at absolute position study.

The data of the point of geography information at user are excavated and calculated however, the prior art lacks, it cannot be true Determine the point of geography information locating for user.

Summary of the invention

The purpose of the application is to propose a kind of determination method and apparatus of improved geography information point, to solve above carry on the back The technical issues of scape technology segment is mentioned.

In a first aspect, this application provides a kind of determination methods of geography information point, which comprises obtain user's Location information, wherein the location information includes that user positions coordinate；The user is positioned into coordinate as shellfish trained in advance The input value of this prediction model of leaf, and the user is obtained according to the bayes predictive model and is at least one geography information The probability value of each geography information point in point, wherein the bayes predictive model is made using the essential information of geography information point It is obtained for sample data training, wherein the essential information includes that the positioning coordinate of geography information point, the history user that visits go through History location information；The corresponding geography information point of most probable value is determined as the point of geography information locating for the user.

In some embodiments, the bayes predictive model parameter includes each geography at least one geography information point The history visiting probability of information point, wherein positioning coordinate and the history of the history visiting probability according to geographical information point Location information obtains, in which: obtains history visiting probability according to the positioning coordinate of geographical information point and the history location information It include: to choose at least one geography information point according to preset rules, and establish geography information point set；According to the geography information Each history visiting user positions letter in the positioning coordinate of each geography information point and the geography information point set in point set Breath obtains each history visiting number in the geography information point set；Calculate each geography in the geography information point set The summation of the history visiting number of information, using the summation as the history of geography information point set visiting total degree；According to institute The history visiting number for stating history visiting total degree and the geography information point in the geography information point set, obtains the geography The history visiting probability of each geography information point in information point set.

In some embodiments, described to be visited according to the positioning coordinate of geographical information point and the history of geography information point User's location information obtains the history visiting number of the geography information point, comprising: chooses in geography information point preset range Historical user；Obtain the history location information and historical search record of the corresponding historical user of the history positioning coordinate, wherein History location information includes history positioning time when history positioning coordinate positions coordinate with the acquisition history；If described go through It include the identification information of the geography information point in history search record；It then calculates the history positioning time and searches for geography letter Cease the time interval between the time point of point；It is less than predetermined threshold in response to the time interval, the historical user is determined For the history visiting user of the geography information point.

In some embodiments, the history location information of history visiting user includes history positioning coordinate and described goes through History visiting user is located at history positioning time when history positioning coordinate；And the bayes predictive model parameter packet It includes: the time probability distribution of geography information point, wherein the time probability distribution is visited according to the history of the geography information point The history positioning time of user obtains.

In some embodiments, when the location information of the user further includes that the user is located at user positioning coordinate User's positioning time；And it is described using user positioning coordinate as the input of bayes predictive model trained in advance Value, and the user is obtained according to the bayes predictive model and is in each geography information point at least one geography information point Probability value, comprising: user's positioning time and the user are positioned into coordinate as the Bayesian forecasting mould trained in advance The input value of type, and the user is obtained according to the bayes predictive model and is at least one geography information point eachly Manage the probability value of information point.

In some embodiments, the bayes predictive model parameter includes the positioning probability of geography information point, wherein institute It states positioning probability to be obtained according to the distance between the geography information point and cluster centre, wherein the cluster centre is by least One geography information point clusters to obtain.

In some embodiments, the cluster centre is clustered to obtain by least one geography information point, comprising: passes through K- Means algorithm clusters to obtain the cluster centre of at least one geography information point, in which: at least one geography information point is chosen, and Establish cluster geography information point set；Clusters number is determined according to the total degree of the cluster geography information point set visiting；Choosing Take the clusters number positioning coordinate as initial cluster center；The clusters number, the initial cluster center is corresponding Coordinate and cluster geography information point set in the positioning coordinate of geography information point be set as the input value of K-means algorithm, obtain To the clusters number cluster centre.

In some embodiments, when the location information of the user further includes that the user is located at user positioning coordinate User's positioning time；And the location information for obtaining user, comprising: filter out user's positioning time in preset time User in section positions coordinate, and establishes original user positioning coordinate set；It rejects in the original user positioning coordinate set Abnormal point, obtain user position coordinate set, wherein the abnormal point refers to the distance moved in the second preset time period Greater than the coordinate points of pre-determined distance threshold value；The user is positioned at least one user positioning coordinate in coordinate set and passes through rail Mark clustering algorithm aggregates into a track centers coordinate；When the user to be positioned at least one user positioning in coordinate set Between corresponding time point average time point as the track centers time；And it is described by the track centers coordinate and institute Input value of the track centers time as bayes predictive model trained in advance is stated, and is obtained according to the bayes predictive model The probability value of each geography information point at least one geography information point is in the user, comprising: by the track centers Input value of the coordinate as bayes predictive model trained in advance, and the user is obtained according to the bayes predictive model The probability value of each geography information point at least one geography information point.

In some embodiments, the corresponding geography information point of most probable value is being determined as the user institute by the method After the geography information point at place, further includes: add the corresponding geography information of the most probable value to the location information of the user The history visiting user's mark of point；The location information of the user with history visiting user's mark is added to the pattra leaves In the sample data sets of this prediction model；New Bayes is generated using the sample data training in the sample data sets Prediction model.

Second aspect, this application provides a kind of determining device of geography information point, described device includes: acquisition module, It is configured to obtain the location information of user, wherein the location information includes that user positions coordinate；Computing module, configuration are used In the user is positioned coordinate as the input value for the bayes predictive model trained in advance, and according to the Bayesian forecasting Model obtains the probability value that the user is in each geography information point at least one geography information point, wherein the pattra leaves This prediction model is obtained using the essential information of geography information point as sample data training, wherein the essential information includes The history location information for positioning coordinate, history visiting user of geography information point；Determining module is configured to most probable value Corresponding geography information point is determined as the point of geography information locating for the user.

In some embodiments, when the location information of the user further includes that the user is located at user positioning coordinate User's positioning time；And the computing module, it is further used for: user's positioning time and the user is positioned Input value of the coordinate as bayes predictive model trained in advance, and the user is obtained according to the bayes predictive model The probability value of each geography information point at least one geography information point.

In some embodiments, when the location information of the user further includes that the user is located at user positioning coordinate User's positioning time；And the acquisition module, it is further used for: filters out user's positioning time within a preset period of time User position coordinate, and establish original user positioning coordinate set；It rejects different in the original user positioning coordinate set Chang Dian obtains user and positions coordinate set, wherein the abnormal point refers to that the distance moved in the second preset time period is greater than The coordinate points of pre-determined distance threshold value；The user is positioned at least one user positioning coordinate in coordinate set to gather by track Class algorithm aggregates into a track centers coordinate；The user is positioned at least one user's positioning time in coordinate set The average time point at corresponding time point is as the track centers time；And it is described by the track centers coordinate and the rail Input value of the mark centre time as bayes predictive model trained in advance, and institute is obtained according to the bayes predictive model State the probability value that user is in each geography information point at least one geography information point, comprising: by the track centers coordinate As the input value of bayes predictive model trained in advance, and the user is obtained according to the bayes predictive model and is in The probability value of each geography information point at least one geography information point.

In some embodiments, described device further includes update module, is configured to: being added to the location information of the user Add the history visiting user's mark of the corresponding geography information point of the most probable value；By the institute with history visiting user's mark The location information for stating user is added in the sample data sets of the bayes predictive model；Utilize the sample data sets In sample data training generate new bayes predictive model.

The determination method and apparatus of geography information point provided by the present application, by the location information for obtaining user, wherein institute Stating location information includes that user positions coordinate；The user is positioned into coordinate as the defeated of bayes predictive model trained in advance Enter value, and the user is obtained according to the bayes predictive model and is in each geography information at least one geography information point The probability value of point, wherein the bayes predictive model is using the essential information of geography information point trained as sample data It arrives, wherein the essential information includes the history location information of the positioning coordinate of geography information point, history visiting user；It will most The corresponding geography information point of greatest is determined as the point of geography information locating for the user, realizes ground locating for determining user Manage information point.

Detailed description of the invention

By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other Feature, objects and advantages will become more apparent upon:

Fig. 1 is that this application can be applied to exemplary system architecture figures therein；

Fig. 2 is the flow chart according to one embodiment of the determination method of the geography information point of the application；

Fig. 3 is the flow chart according to another embodiment of the determination method of the geography information point of the application；

Fig. 4 is distributed according to the time probability of the geography information point of the determination method of the geography information point of the application；

Fig. 5 is the structural schematic diagram according to one embodiment of the determining device of the geography information point of the application；

Fig. 6 is adapted for the structural representation of the computer system for the terminal device or server of realizing the embodiment of the present application Figure.

Specific embodiment

The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to Convenient for description, part relevant to related invention is illustrated only in attached drawing.

It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

Fig. 1 is shown can be using the determining device of the determination method or geography information point of the geography information point of the application The exemplary system architecture 100 of embodiment.

As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105. Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with Including various connection types, such as wired, wireless communication link or fiber optic cables etc..

User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out Send message etc..Various telecommunication customer end applications, such as map class application, purchase can be installed on terminal device 101,102,103 Species application, searching class application, instant messaging tools, mailbox client, social platform software etc..

Terminal device 101,102,103 can be the various electronic equipments with display screen and supported web page browsing, packet Include but be not limited to smart phone, tablet computer, E-book reader, MP3 player (Moving Picture Experts Group Audio Layer III, dynamic image expert's compression standard audio level 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert's compression standard audio level 4) it is player, on knee portable Computer and desktop computer etc..

Server 105 can be to provide the server of various services, such as take to the positioning of terminal device 101,102,103 Business provides the location service server supported.Location service server can analyze the data such as the location data received Deng processing.

It should be noted that the determination method of the point of geography information provided by the embodiment of the present application is generally by server 105 It executes, correspondingly, the determining device of geography information point is generally positioned in server 105.

It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need It wants, can have any number of terminal device, network and server.

With continued reference to Fig. 2, the process of one embodiment of the determination method of the geography information point according to the application is shown 200.The determination method of above-mentioned geography information point, comprising the following steps:

Step 201, the location information of user is obtained.

In the present embodiment, electronic equipment (such as the clothes shown in FIG. 1 of the determination method operation of geography information point thereon Business device) it can be based on the location information of acquisition for mobile terminal user used by a user.It should be pointed out that above-mentioned be based on user The location information of used acquisition for mobile terminal user can realize there are many mode, and herein, implementation includes but not It is limited to the positioning based on GPS (Global Positioning System, global positioning system), the base based on mobile operation network The positioning stood, the positioning based on AGPS (AssistedGPS, auxiliary global satellite positioning system), the positioning based on WiFi and Other currently known or exploitation in the future mobile terminal location modes.

In the present embodiment, the location information of above-mentioned user includes that user positions coordinate.Herein, user's positioning coordinate can To be latitude and longitude coordinates.

Step 202, user is positioned into coordinate as the input value of bayes predictive model trained in advance, and according to pattra leaves This prediction model obtains the probability value that user is in each geography information point at least one geography information point.

In the present embodiment, coordinate is positioned based on the user for obtaining user in step 201, above-mentioned electronic equipment (such as Fig. 1 Shown in server) can first using above-mentioned positioning coordinate as it is preparatory training bayes predictive model input value；Later again The probability that user is in some geography information point is obtained using bayes predictive model；It is one or many to utilize Bayesian forecasting mould Type obtains the probability that user is in each geography information point at least one geography information point, as an example, can be by user's User positions coordinate and obtains user to be in the probability on first ground being a as the input value of bayes predictive model, recycles Bayes Prediction model predicts that above-mentioned user obtains user to be in the probability on second ground being b.

In the present embodiment, bayes predictive model is using the essential information of geography information point trained as sample data It arrives, wherein above-mentioned essential information includes the positioning coordinate of geography information point, the history location information of history visiting user.At this In, above-mentioned bayes predictive model is the prediction model established with Bayesian formula for basic principle, as an example, being applied to The Bayesian formula of the present embodiment can be indicated with following formula:

P (U | poi)=A*B

Wherein, a certain geography information point of poi expression, the location information of U expression user, P (U | poi) indicate that user is in certain The probability of 1 geography information point, A, B are bayes predictive model parameter, * expression parameter bayes predictive model A and Bayes There is operation relation, above-mentioned operation relation includes but is not limited to multiplication relationship, adduction relationship between prediction model parameters B.

In some optional implementations of the present embodiment, above-mentioned bayes predictive model parameter includes at least one ground Manage the history visiting probability of each geography information point in information point, wherein above-mentioned history visiting probability is according to geographical information point Positioning coordinate and above-mentioned history location information obtain, wherein can obtain being sat according to the positioning of geography information by following steps Mark and above-mentioned history location information on earth history visit probability: choose at least one geography information point according to preset rules, and build On the spot manage information point set；According to the positioning coordinate of geography information point each in above-mentioned geography information point set and above-mentioned geographical letter Each history visiting user's location information obtains each history visiting number in above-mentioned geography information point set in breath point set；Meter The summation for counting stating the history visiting number of each geography information in geography information point set in, believes above-mentioned summation as geography Cease the history visiting total degree of point set；According to the geographical letter in above-mentioned history visiting total degree and above-mentioned geography information point set The history visiting number of breath point obtains the history visiting probability of each geography information point in above-mentioned geography information point set.

As an example, the closer three geography information point of selected distance establishes geography information point set, three geography information Point be respectively designated as geography information point a, geography information point b, geography information point c, these three geography information points visiting at one day The total degree of middle visiting is 100 times, wherein the number of visiting geography information point a is 20 times, the number of visiting geography information point b is stifled Vehicle 30 times, the number of visiting geography information point c is 50 times, then it is hundred that the history visiting probability of geography information point a, which is 20/100, / bis- ten, then it is 30 percent that the history visiting probability of geography information point a, which is 30/100, then geography information point a It is 50 percent that history visiting probability, which is 50/100,.

Optionally, according to history visiting user's location information of the positioning coordinate of geographical information point and above-mentioned geography information point The history visiting number of above-mentioned geography information point is obtained, can be obtained by following steps: choose geography information point preset range Interior historical user；The history location information and historical search record of the corresponding historical user of above-mentioned history positioning coordinate are obtained, Wherein, history location information includes history positioning time when history positioning coordinate positions coordinate with the above-mentioned history of acquisition；If It include the identification information of above-mentioned geography information point in above-mentioned historical search record；It then calculates above-mentioned history positioning time and search should Time interval between the time point of geography information point；It is less than predetermined threshold in response to above-mentioned time interval, above-mentioned history is used Family is determined as the history visiting user of above-mentioned geography information point.

As an example, choosing 10 points to 11 points of historical user of the morning within the scope of 100 meters of radius of geography information point first, example As user Zhang Yi in 10: 30 quartile of the morning within the scope of this, then obtain an one search record, if search record, if Include the identification information of geography information point first in search record, for example, geography information point first title, with geography information point first The similar title of title, then after, obtain searching for geo information point first identification information time point, such as be 10 points 15 minutes, search 10 points of rope time point are less than predetermined threshold with 10 points of 30 minutes time intervals of history positioning time in 15 minutes, then above-mentioned historical user is true It is set to the history visiting user of above-mentioned geography information point, it is to be understood that above-mentioned predetermined threshold can be half an hour, can be One day, it is also possible to one month.

Optionally, according to history visiting user's location information of the positioning coordinate of geographical information point and above-mentioned geography information point The history visiting number of above-mentioned geography information point is obtained, can be obtained in the following manner: be obtained historical user and connect WiFi's Data can determine that this historical user is the geography information point if this WiFi, which has determined that, belongs to a certain geography information point History visit user.

Optionally, at least one geography information point is chosen according to preset rules, and establishes geography information point set, wherein Choosing at least one geographical preset rules can be selection customer-centric, several geography information in preset range Point.Be also possible to establish previously according to the concentration of geography information point some region of geography information point is divided into it is several A geography information point set, for example, a certain campus is established three centered on main teaching building, gymnasium, library respectively A geography information point set, wherein may include equipment room, playground in the geography information point set centered on gymnasium.

In some optional implementations of the present embodiment, above-mentioned bayes predictive model parameter includes geography information point Position probability, wherein above-mentioned positioning probability is obtained according to the distance between above-mentioned geography information point and cluster centre, wherein on Cluster centre is stated to cluster to obtain by least one geography information point.For example, by above-mentioned geography information point and above-mentioned cluster centre it Between distance positioning probability of the inverse ratio as above-mentioned geography information point.It is, of course, also possible to by geography information point and cluster centre it Between distance positioning probability of the inverse ratio multiplied by coefficient as above-mentioned geography information point.

It is alternatively possible to which the bridge using above-mentioned cluster centre between above-mentioned user and geography information point, calculates first User-cluster centre probability, such as can be general as user-cluster centre using above-mentioned user and the inverse ratio of cluster centre distance Rate, then using the product of user-cluster centre probability and above-mentioned positioning probability as user geography information point probability.As showing Example, user's first is near cluster centre A and cluster centre B, and the distance between user's first and cluster centre A are 10 meters, Yong Hujia The distance between cluster centre A is 20 meters；There are geography information point c and ground in the cluster geography information point set of cluster centre A Manage information point d, wherein 1 meter of the distance between geography information point c and cluster centre A, between geography information point c and cluster centre A Distance be 2 meters；There is geography information point e in the cluster geography information point set of cluster centre B, wherein geography information point e and poly- The distance between class center B is 4 meters；So, if it is general using the inverse ratio of distance between geographic center and cluster centre as positioning Rate, then the positioning probability of geography information point a is 1/1, then the positioning probability of geography information point b is 1/2, then geography information The positioning probability of point c is 1/4, and the user between user and cluster centre A-cluster centre probability is 1/10, in user and cluster User-cluster centre probability between heart B is 1/20；Finally, obtaining, probability of the above-mentioned user in geography information point a is (1/ 1) (1/10) *, above-mentioned user are (1/2) * (1/10) in the probability of geography information point b, and above-mentioned user is general geography information point c's Rate is (1/4) * (1/20).Herein, "/" indicates that the division sign, " * " indicate operation, it is preferable that " * " indicates multiplication sign.

Optionally, at least one geography information point is clustered to obtain cluster centre and can be can use poly- with random division region Class algorithm cluster, wherein clustering algorithm includes but is not limited to: k-means clustering algorithm, hierarchical clustering algorithm, SOM cluster are calculated Method, FCM clustering algorithm.It should be understood that the calculating process of above-mentioned clustering algorithm itself be it is as well known to those skilled in the art, Therefore not to repeat here.

It is alternatively possible to cluster to obtain the cluster centre of at least one geography information point by K-means algorithm, it is optional Ground, detailed process is as follows: choosing at least one geography information point, and establishes cluster geography information point set；According to above-mentioned cluster The total degree of geography information point set visiting determines clusters number；Above-mentioned clusters number positioning coordinate is chosen as initial clustering Center；By geography information in above-mentioned clusters number, the corresponding coordinate of above-mentioned initial cluster center and cluster geography information point set The positioning coordinate of point is set as the input value of K-means algorithm, obtains above-mentioned clusters number cluster centre.It should be understood that K- Means algorithm is clustering algorithm well known to those skilled in the art, and therefore not to repeat here.

It, can also be with it is alternatively possible to determine clusters number according to the number of geography information point in cluster geography information set According to geography information point distribution in cluster geography information point set whether with the determining geographical letter of the intensive phenomenon of apparent subregion Cease the number of point.

It is alternatively possible to randomly select above-mentioned clusters number positioning coordinate as initial cluster center, can also will go through History visits positioning coordinate of the probability greater than the geography information point of predetermined threshold as initial cluster center.

Step 203, the corresponding geography information point of most probable value is determined as the point of geography information locating for above-mentioned user.

In the present embodiment, the user obtained based on step 202 is in each geographical letter at least one geography information point The probability value for ceasing point, selects maximum probability value in above-mentioned probability value as most probable value, and most probable value is corresponding Geography information point is determined as the point of geography information locating for above-mentioned user.

The method provided by the above embodiment of the application is by the history visiting probability using geography information point and positions general Rate realizes the point of geography information locating for determining user.

With further reference to Fig. 3, it illustrates the processes 300 of another embodiment of the determination method of geography information point.It should The process 300 of the determination method of geography information point, comprising the following steps:

Step 301, the user for obtaining user positions coordinate and user's positioning time.

In the present embodiment, electronic equipment (such as the clothes shown in FIG. 1 of the determination method operation of geography information point thereon Business device) it can be based on the location information of acquisition for mobile terminal user used by a user.

In the present embodiment, the location information of above-mentioned user includes that user's positioning coordinate and above-mentioned user are located at above-mentioned user Position user's positioning time when coordinate.The location information of user can be obtained by following steps: when filtering out user's positioning Between user within a preset period of time position coordinate, and establish original user positioning coordinate set；It is fixed to reject above-mentioned original user Abnormal point in the coordinate set of position obtains user and positions coordinate set, wherein above-mentioned abnormal point refers in the second preset time period The distance of interior movement is greater than the coordinate points of pre-determined distance threshold value；Above-mentioned user is positioned at least one user in coordinate set to determine Position coordinate aggregates into a track centers coordinate by trajectory clustering algorithm；Above-mentioned user is positioned at least one in coordinate set The average time point at the corresponding time point of a user's positioning time is as the track centers time；By above-mentioned track centers coordinate and Input value of the above-mentioned track centers time as bayes predictive model trained in advance, and according to above-mentioned bayes predictive model Obtain the probability value that above-mentioned user is in each geography information point at least one geography information point.

Step 302, user is positioned into coordinate and user's positioning time as the defeated of bayes predictive model trained in advance Enter value, and the probability that user is in each geography information point at least one geography information point is obtained according to bayes predictive model Value.

In the present embodiment, coordinate is positioned based on the user for obtaining user in step 301, above-mentioned electronic equipment (such as Fig. 1 Shown in server) can first using above-mentioned positioning coordinate as it is preparatory training bayes predictive model input value；Later again The probability that user is in some geography information point is obtained using bayes predictive model；It is one or many to utilize Bayesian forecasting mould Type obtains the probability that user is at least one geography information point.

P (U | poi)=A*B

In the present embodiment, above-mentioned bayes predictive model parameter includes each geographical letter at least one geography information point The history visiting probability of breath point, wherein above-mentioned history visiting probability is fixed according to the positioning coordinate of geographical information point and above-mentioned history Position information obtains.

In the present embodiment, above-mentioned bayes predictive model parameter includes the positioning probability of geography information point, wherein above-mentioned Positioning probability is obtained according to the distance between above-mentioned geography information point and cluster centre, wherein above-mentioned cluster centre is by least one A geography information point clusters to obtain.

In the present embodiment, above-mentioned bayes predictive model parameter includes the time probability distribution of geography information point, wherein Above-mentioned time probability distribution is obtained according to the history positioning time of the history of geographical information point visiting user, wherein above-mentioned history Positioning time is time point when history visiting user is in history positioning coordinate, and history positions coordinate and history positioning time belongs to In the history location information of history visiting user, the history location information by acquiring historical user obtains above-mentioned history positioning and sits Mark and history positioning time.As an example, Fig. 4 can be referred to, it illustrates geography information o'clock according to one week time probability minute Cloth.Geography information point first a total of 100 history visiting user, the week in one week seven days have 10 history to arrive daily User is visited, Saturday there are 20 history visiting users, and Sunday there are 30 history visiting users, here, it is assumed that each history is visited User's visiting geography information point first is primary.It can establish and be distributed by the time probability in period of week.

Optionally, the period of time probability distribution can be one day twenty four hours, can be seven days of one week, can be with It is one month number of days, can be 12 months of 1 year, is also possible to 1 year four season.It is of course also possible to above-mentioned week The combination of phase form.As an example, can establish one day twenty four hours and combination one week seven days, certain user in Friday 19 The probability of point visiting geography information point first, can corresponding time probability and 19 points exist in geography information point first one week by Friday Corresponding time probability in geography information point first one day, as the parameter value calculation user Friday 19 points of visiting geography information The probability of point first.

In some optional implementations of the present embodiment, the location information of above-mentioned user further includes above-mentioned user positioned at upper User's positioning time when user positions coordinate is stated, above-mentioned user's positioning time and above-mentioned user can be positioned into coordinate as pre- The first input value of trained bayes predictive model, and above-mentioned user is obtained according to above-mentioned bayes predictive model and is at least one The probability value of each geography information point in a geography information point.

Step 303, the corresponding geography information point of most probable value is determined as the point of geography information locating for user.

In the present embodiment, the user obtained based on step 302 is in each geographical letter at least one geography information point The probability value for ceasing point, selects maximum probability value in above-mentioned probability value as most probable value, and most probable value is corresponding Geography information point is determined as the point of geography information locating for above-mentioned user.

Step 304, the location information based on user generates new bayes predictive model.

In the present embodiment, it is based on step 303, it is corresponding to add above-mentioned most probable value to the location information of above-mentioned user The history visiting user's mark of geography information point, the location information of the above-mentioned user with history visiting user's mark is added to In the sample data sets of above-mentioned bayes predictive model, generated using the sample data training in above-mentioned sample data sets new Bayes predictive model.

From figure 3, it can be seen that compared with the corresponding embodiment of Fig. 2, the determination side of the geography information point in the present embodiment The process 300 of method highlights the step of user's positioning time for introducing user, and the time probability of geography information point is utilized Distribution more accurately determines geography information point locating for user to realize.

With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, this application provides a kind of geography information points Determining device one embodiment, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which specifically can be with Applied in various electronic equipments.

As shown in figure 5, the determining device 500 of the above-mentioned geography information point of the present embodiment includes: to obtain module 501, calculate Module 502, determining module 503.Wherein, module is obtained, is configured to obtain the location information of user, wherein above-mentioned positioning letter Breath includes that user positions coordinate；Computing module is configured to above-mentioned user positioning coordinate is pre- as Bayes trained in advance The input value of model is surveyed, and above-mentioned user is obtained according to above-mentioned bayes predictive model and is at least one geography information point every The probability value of a geography information point, wherein above-mentioned bayes predictive model is using the essential information of geography information point as sample Data training obtains, wherein above-mentioned essential information includes the history positioning of the positioning coordinate of geography information point, history visiting user Information；Determining module is configured to for the corresponding geography information point of most probable value to be determined as the letter of geography locating for above-mentioned user Breath point.

In the present embodiment, the acquisition module 501 of the determining device 500 of geography information point can be based on used by a user The location information of acquisition for mobile terminal user.It should be pointed out that above-mentioned be based on acquisition for mobile terminal user used by a user Location information, can realize there are many mode.

In the present embodiment, based on the location information of user for obtaining module 501 and obtaining, above-mentioned computing module 502 can be with User's positioning coordinate that user is obtained in module 501 is obtained, above-mentioned electronic equipment (such as server shown in FIG. 1) can be first Using above-mentioned positioning coordinate as the input value of preparatory training bayes predictive model；Bayes predictive model is recycled to obtain later User is in the probability of some geography information point；One or many utilization bayes predictive models obtain user and are at least one The probability of each geography information point in geography information point.

In the present embodiment, the user obtained based on computing module 502 is at least one geography information point eachly The probability value of information point is managed, determining module 503 selects maximum probability value as most probable value from above-mentioned probability value, incites somebody to action The corresponding geography information point of most probable value is determined as the point of geography information locating for above-mentioned user.

In some optional implementations of the present embodiment, the determining device 500 of geography information point further includes update module 504, it is configured to add the location information of above-mentioned user the history visiting of the corresponding geography information point of above-mentioned most probable value The location information of above-mentioned user with history visiting user's mark is added to above-mentioned bayes predictive model by user's mark In sample data sets, new bayes predictive model is generated using the sample data training in above-mentioned sample data sets.

It will be understood by those skilled in the art that the determining device 500 of above-mentioned geography information point further includes some other known Structure, such as processor, memory etc., in order to unnecessarily obscure embodiment of the disclosure, these well known structures are in Fig. 5 It is not shown.

Below with reference to Fig. 6, it illustrates the computer systems 600 for the server for being suitable for being used to realize the embodiment of the present application Structural schematic diagram.

As shown in fig. 6, computer system 600 includes central processing unit (CPU) 601, it can be read-only according to being stored in Program in memory (ROM) 602 or be loaded into the program in random access storage device (RAM) 603 from storage section 608 and Execute various movements appropriate and processing.In RAM 603, also it is stored with system 600 and operates required various programs and data. CPU 601, ROM 602 and RAM 603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to always Line 604.

I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.；It is penetrated including such as cathode The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 608 including hard disk etc.； And the communications portion 609 of the network interface card including LAN card, modem etc..Communications portion 609 via such as because The network of spy's net executes communication process.Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to read from thereon Computer program be mounted into storage section 608 as needed.

Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be tangibly embodied in machine readable Computer program on medium, above-mentioned computer program include the program code for method shown in execution flow chart.At this In the embodiment of sample, which can be downloaded and installed from network by communications portion 609, and/or from removable Medium 611 is unloaded to be mounted.

Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants It is noted that the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, Ke Yiyong The dedicated hardware based system of defined functions or operations is executed to realize, or can be referred to specialized hardware and computer The combination of order is realized.

Being described in module involved in the embodiment of the present application can be realized by way of software, can also be by hard The mode of part is realized.Described module also can be set in the processor, for example, can be described as: a kind of processor packet It includes and obtains module, computing module, determining module.Wherein, the title of these modules is not constituted under certain conditions to the module The restriction of itself, for example, obtaining module is also described as " obtaining the module of the location information of user ".

As on the other hand, present invention also provides a kind of nonvolatile computer storage media, the non-volatile calculating Machine storage medium can be nonvolatile computer storage media included in above-mentioned apparatus in above-described embodiment；It is also possible to Individualism, without the nonvolatile computer storage media in supplying terminal.Above-mentioned nonvolatile computer storage media is deposited One or more program is contained, when said one or multiple programs are executed by an equipment, so that above equipment: obtaining The location information of user, wherein above-mentioned location information includes that user positions coordinate；Above-mentioned user is positioned into coordinate as preparatory instruction The input value of experienced bayes predictive model, and above-mentioned user is obtained according to above-mentioned bayes predictive model and is at least one ground Manage the probability value of each geography information point in information point, wherein above-mentioned bayes predictive model is basic using geography information point Information is obtained as sample data training, wherein above-mentioned essential information includes the positioning coordinate of geography information point, history visiting use The history location information at family；The corresponding geography information point of most probable value is determined as the point of geography information locating for above-mentioned user.

Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein Can technical characteristic replaced mutually and the technical solution that is formed.

Claims

1. a kind of determination method of geography information point, which is characterized in that the described method includes:

Obtain the location information of user, wherein the location information includes that user's positioning coordinate and the user are located at the use Family positions user's positioning time when coordinate；

The user is positioned into coordinate as the input value for the bayes predictive model trained in advance, and it is pre- according to the Bayes It surveys model and obtains the probability value that the user is in each geography information point at least one geography information point, wherein the shellfish This prediction model of leaf is obtained using the essential information of geography information point as sample data training, wherein the essential information packet Include the positioning coordinate of geography information point, the history location information of history visiting user；

The corresponding geography information point of most probable value is determined as the point of geography information locating for the user；Wherein

The location information for obtaining user, comprising:

It filters out the user of user's positioning time within a preset period of time and positions coordinate, and establish original user positioning coordinate set It closes；

The abnormal point in the original user positioning coordinate set is rejected, user is obtained and positions coordinate set, wherein the exception Point refers to that the distance moved in the second preset time period is greater than the coordinate points of pre-determined distance threshold value；

The user is positioned at least one user positioning coordinate in coordinate set and aggregates into one by trajectory clustering algorithm Track centers coordinate；

The user is positioned to the average time point at the corresponding time point of at least one user's positioning time in coordinate set As the track centers time；And

It is described that the user is positioned into coordinate as the input value of bayes predictive model trained in advance, and according to the pattra leaves This prediction model obtains the probability value that the user is in each geography information point at least one geography information point, comprising:

It is described using the track centers coordinate and the track centers time as the defeated of bayes predictive model trained in advance Enter value, and the user is obtained according to the bayes predictive model and is in each geography information at least one geography information point The probability value of point.

2. the method according to claim 1, wherein the bayes predictive model parameter includes at least one ground Manage the history visiting probability of each geography information point in information point, wherein the history visiting probability is according to geographical information point Positioning coordinate and the history location information obtain, in which:

Obtaining history visiting probability according to the positioning coordinate of geographical information point and the history location information includes:

At least one geography information point is chosen according to preset rules, and establishes geography information point set；

According to every in the positioning coordinate of geography information point each in the geography information point set and the geography information point set A history visiting user's location information obtains each history visiting number in the geography information point set；

Calculate each geography information in the geography information point set history visiting number summation, using the summation as The history visiting total degree of geography information point set；

According to the history visiting number of the geography information point in history visiting total degree and the geography information point set, obtain The history visiting probability of each geography information point into the geography information point set.

3. according to the method described in claim 2, it is characterized in that, the positioning coordinate according to geographical information point and described History visiting user's location information of reason information point obtains the history visiting number of the geography information point, comprising:

Choose the historical user in geography information point preset range；

Obtain the history location information and historical search record of the corresponding historical user of the history positioning coordinate, wherein history Location information includes history positioning time when history positioning coordinate positions coordinate with the acquisition history；

If in the historical search record including the identification information of the geography information point；

It then calculates the history positioning time and searches for the time interval between time point of the geography information point；

It is less than predetermined threshold in response to the time interval, the history that the historical user is determined as the geography information point is arrived Visit user.

4. the method according to claim 1, wherein the history location information of history visiting user includes going through History positioning coordinate and history visiting user are located at history positioning time when history positioning coordinate；And

The bayes predictive model parameter includes:

The time probability of geography information point is distributed, wherein the time probability distribution is arrived according to the history of the geography information point The history positioning time for visiting user obtains.

5. according to the method described in claim 4, it is characterized in that, the location information of the user further includes that the user is located at The user positions user's positioning time when coordinate；And

User's positioning time and the user are positioned into coordinate as the input value for the bayes predictive model trained in advance, And the user is obtained according to the bayes predictive model and is in each geography information point at least one geography information point Probability value.

6. the method according to claim 1, wherein the bayes predictive model parameter includes geography information point Positioning probability, wherein the positioning probability is obtained according to the distance between the geography information point and cluster centre, wherein The cluster centre is clustered to obtain by least one geography information point.

7. according to the method described in claim 6, it is characterized in that, the cluster centre is clustered by least one geography information point It obtains, comprising:

It clusters to obtain the cluster centre of at least one geography information point by K-means algorithm, in which:

At least one geography information point is chosen, and establishes cluster geography information point set；

Clusters number is determined according to the total degree of the cluster geography information point set visiting；

The clusters number positioning coordinate is chosen as initial cluster center；

By geography information point in the clusters number, the corresponding coordinate of the initial cluster center and cluster geography information point set Positioning coordinate be set as the input value of K-means algorithm, obtain the clusters number cluster centre.

8. method described in -7 according to claim 1, which is characterized in that the method is believed by the corresponding geography of most probable value After breath point is determined as the point of geography information locating for the user, further includes:

The history visiting user's mark of the corresponding geography information point of the most probable value is added to the location information of the user；

The location information of the user with history visiting user's mark is added to the sample of the bayes predictive model In data acquisition system；

New bayes predictive model is generated using the sample data training in the sample data sets.

9. a kind of determining device of geography information point, which is characterized in that described device includes:

Module is obtained, is configured to obtain the location information of user, wherein the location information includes that user positions coordinate；

Computing module, be configured to using the user position coordinate as in advance train bayes predictive model input value, And the user is obtained according to the bayes predictive model and is in each geography information point at least one geography information point Probability value, wherein the bayes predictive model is obtained using the essential information of geography information point as sample data training, In, the essential information includes the history location information of the positioning coordinate of geography information point, history visiting user；

Determining module is configured to the corresponding geography information point of most probable value being determined as geography information locating for the user Point；Wherein

The acquisition module, is further used for: it filters out the user of user's positioning time within a preset period of time and positions coordinate, and Establish original user positioning coordinate set；The abnormal point in the original user positioning coordinate set is rejected, user's positioning is obtained Coordinate set, wherein the abnormal point refers to that the distance moved in the second preset time period is greater than the seat of pre-determined distance threshold value Punctuate；The user is positioned at least one user positioning coordinate in coordinate set and aggregates into one by trajectory clustering algorithm Track centers coordinate；The user is positioned into the flat of the corresponding time point of at least one user's positioning time in coordinate set Equal time point is as the track centers time；

The computing module, is further used for: using the track centers coordinate and the track centers time as preparatory training Bayes predictive model input value, and according to the bayes predictive model obtain the user be at least one geography The probability value of each geography information point in information point.

10. device according to claim 9, which is characterized in that the bayes predictive model parameter includes at least one The history visiting probability of each geography information point in geography information point, wherein the history visiting probability is according to geographical information point Positioning coordinate and the history location information obtain, in which:

11. device according to claim 10, which is characterized in that the positioning coordinate according to geographical information point and described History visiting user's location information of geography information point obtains the history visiting number of the geography information point, comprising:

Choose the historical user in geography information point preset range；

12. device according to claim 9, which is characterized in that the history location information of history visiting user includes History positioning coordinate and history visiting user are located at history positioning time when history positioning coordinate；And

The bayes predictive model parameter includes:

13. device according to claim 12, which is characterized in that the location information of the user further includes the user position User's positioning time when the user positions coordinate；And

The computing module, is further used for:

14. device according to claim 9, which is characterized in that the bayes predictive model parameter includes geography information The positioning probability of point, wherein the positioning probability is obtained according to the distance between the geography information point and cluster centre, In, the cluster centre is clustered to obtain by least one geography information point.

15. device according to claim 14, which is characterized in that the cluster centre is gathered by least one geography information point Class obtains, comprising:

16. device according to claim 15, which is characterized in that described device further includes update module, is configured to: