The present disclosure relates to the field of geographic information processing technologies, and in particular, to a clustering method for a point of interest and a related apparatus.
A point of interest (POI) refers to a location area in which a user frequently stays for a long time, for example an area that is of great importance to the user, such as a home, an office, or a frequently-visited supermarket.
Track information of daily activities of the user may be acquired by using a locating function, such as a Wi-Fi network, a Global Positioning System (GPS), and a base station identity (ID) of a terminal such as a mobile phone. The track information includes a large quantity of locating coordinate points with a locating deviation. The study on how to extract a POI of the user from the track information is of great value to context awareness and a location based service (LBS), and is also a research focus in the academia.
Currently, there is a method for exploring a POI based on multiple GPS track information of a user, and a main idea of the method is: first modeling multiple historical location data of the user by using a tree-like hierarchical pattern, and then proposing, based on the tree-like hierarchical pattern, an inference model that performs searching based on a hypertext subject, and establishing a link from a user to a location for a single visit of an individual.
However, in the foregoing method, a stay point of the user is extracted by using spatial and temporal dimensions. The stay point can represent only a single visit of the user, but cannot represent a POI place that is of great importance to the user. In addition, when a POI is being explored, only historical location data of the user is referred to, and therefore reliability and reference value of the explored POI is low.
Embodiments of the present disclosure provide a clustering method for a point of interest and a related apparatus, which are used to improve reliability and reference value of a POI.
A first aspect of the present disclosure provides a clustering method for a point of interest. The method includes acquiring a locating point set of a user within a preset period. The method also includes generating a stay point set according to the foregoing locating point set. Each stay point in the foregoing stay point set represents one hot area, and the foregoing hot area meets the following conditions: a distance between geographic locations of any two locating points in the foregoing hot area is less than a higher locating precision in locating precisions of the foregoing two locating points, and a maximum value of a time interval between locating points in the foregoing hot area is greater than a preset time threshold. The method also includes calculating a confidence level of each stay point in the foregoing stay point set, where a lower average speed corresponding to movement states of all locating points in a hot area represented by a stay point indicates a higher confidence level of the stay point. The method also includes obtaining a trusted stay point from the foregoing stay point set by means of screening according to the confidence level of each stay point in the foregoing stay point set, where a confidence level of the foregoing trusted stay point is greater than a preset confidence level threshold. The method also includes clustering density-connected trusted stay points to form a point of interest, where the foregoing density-connected indicates that ranges of hot areas represented by two trusted stay points are directly connected or indirectly connected each other.
Based on the first aspect of the present disclosure, in a first possible implementation manner, the foregoing generating a stay point set according to the foregoing locating point set includes: determining a hot area that meets the foregoing conditions; and determining a geometric center point of the foregoing hot area as a stay point that represents the foregoing hot area.
Based on the first aspect of the present disclosure or the first possible implementation manner of the first aspect of the present disclosure, in a second possible implementation manner, the foregoing calculating a confidence level of each stay point in the foregoing stay point set includes: acquiring movement states of all locating points included in a hot area represented by each stay point; and calculating the confidence level of each stay point in the foregoing stay point set by using a formula
and according to the movement states of all locating points included in the hot area represented by each stay point, where Confi represents a confidence level of a stay point i, n represents n possible movement states, Wk represents a confidence level weight of the kth movement state, and nk represents a quantity of locating points that are in a hot area represented by the stay point i and whose movement states are the kth movement state, where each movement state is corresponding to a confidence level weight, and a lower movement speed corresponding to a movement state indicates a larger confidence level weight of the movement state.
Based on the second possible implementation manner of the first aspect of the present disclosure, in a third possible implementation manner, the foregoing acquiring movement states of all locating points included in a hot area represented by each residence point is as follows: acquiring the movement states of all locating points in the foregoing locating point set according to sensor data on a terminal of the user or a change of signal strength and a change of a signal quantity of a Wi-Fi network.
Based on the first aspect, or the first possible implementation manner of the first aspect of the present disclosure, or the second possible implementation manner of the first aspect of the present disclosure, or the third possible implementation manner of the first aspect of the present disclosure, in a fourth possible implementation manner, the foregoing clustering density-connected trusted stay points to form a point of interest includes: determining a closed area that is formed by sequentially connecting all leaf stay points among all density-connected trusted stay points as an area of a point of interest, where a sum of confidence levels of all stay points within a preset radius coverage range whose center is a leaf stay point is not greater than a preset threshold.
A second aspect of the present disclosure provides a clustering apparatus for a point of interest. The apparatus includes: an acquiring unit, configured to acquire a locating point set of a user within a preset period. The apparatus also includes a stay point generating unit, configured to generate a stay point set according to the locating point set acquired by the foregoing acquiring unit. Each stay point in the foregoing stay point set represents one hot area, and the foregoing hot area meets the following conditions: a distance between geographic locations of any two locating points in the foregoing hot area is less than a higher locating precision in locating precisions of the foregoing two locating points, and a maximum value of a time interval between locating points in the foregoing hot area is greater than a preset time threshold. The apparatus also includes a calculating unit, configured to calculate a confidence level of each stay point in the foregoing stay point set, where a lower average speed corresponding to movement states of all locating points in a hot area represented by a stay point indicates a higher confidence level of the stay point. The apparatus also includes a filtering unit, configured to obtain a trusted stay point from the foregoing stay point set by means of screening according to the confidence level that is of each stay point in the foregoing stay point set and is calculated by the foregoing calculating unit, where a confidence level of the foregoing trusted stay point is greater than a preset confidence level threshold. The apparatus also includes a clustering unit, configured to cluster density-connected trusted stay points to form a point of interest, where the foregoing “density-connected” indicates that ranges of hot areas represented by two trusted stay points are directly connected or indirectly connected.
Based on the second aspect of the present disclosure, in a first possible implementation manner, the foregoing stay point generating unit includes: a first determining unit, configured to determine a hot area that meets the foregoing conditions; and a second determining unit, configured to determine a geometric center point of the foregoing hot area as a stay point that represents the foregoing hot area.
Based on the second aspect of the present disclosure or the first possible implementation manner of the second aspect of the present disclosure, in a second possible implementation manner, the foregoing calculating unit includes: an acquiring subunit, configured to acquire movement states of all locating points included in a hot area represented by each stay point; and a calculating subunit, configured to calculate the confidence level of each stay point in the foregoing stay point set by using a formula
and according to the movement states that are of all locating points included in the hot area represented by each stay point and are acquired by the foregoing acquiring subunit, where Wk represents a confidence level of a stay point i, n represents n possible movement states, represents a confidence level weight of the kth movement state, and nk represents a quantity of locating points that are in a hot area represented by the stay point i and whose movement states are the kth movement state, where each movement state is corresponding to a confidence level weight, and a lower movement speed corresponding to a movement state indicates a larger confidence level weight of the movement state.
Based on the second possible implementation manner of the second aspect of the present disclosure, in a third possible implementation manner, the foregoing acquiring subunit is configured to acquire the movement states of all locating points included in the hot area represented by each residence point according to sensor data on a terminal of the user or a change of signal strength and a change of a signal quantity of a Wi-Fi network.
Based on the second aspect of the present disclosure, or the first possible implementation manner of the second aspect of the present disclosure, or the second possible implementation manner of the second aspect of the present disclosure, or the third possible implementation manner of the second aspect of the present disclosure, in a fourth possible implementation manner, the clustering unit is configured to cluster all density-connected trusted stay points to form a point of interest, and determine a closed area that is formed by sequentially connecting all leaf stay points among all density-connected trusted stay points as an area of a point of interest, where a sum of confidence levels of all stay points within a preset radius coverage range whose center is a leaf stay point is not greater than a preset threshold.
It can be learned from the foregoing technical solutions that the embodiments of the present disclosure have the following advantages.
It can be learned from the foregoing that, in the embodiments of the present disclosure, one stay point is formed by multiple locating points according to a locating precision and a time threshold, and most layover points, locating jump points, and pass-through points can be filtered out under the constraint of the locating precision and the time threshold. In addition, a confidence level of a stay point is calculated with reference to movement states of locating points of the stay point, and a stay point whose confidence level is relatively low is screened out according to the confidence level of the stay point, so that some on-the-way noise stay points (such as pseudo stay points generated at the time of staying in a traffic-jammed section, waiting for a traffic light, and slowly walking) can further be filtered out. Therefore, a point of interest that is finally formed by clustering density-connected trusted stay points has higher reliability and higher reference value.
For a more complete understanding of the present invention, and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings.
To describe the technical solutions in the embodiments of the present disclosure more clearly, the following briefly introduces the accompanying drawings required for describing the embodiments. The accompanying drawings in the following description show merely some embodiments of the present disclosure, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.
Embodiments of the present disclosure provide a clustering method for a point of interest and a related apparatus.
To make the disclosure objectives, features, and advantages of the present disclosure clearer and more comprehensible, the following clearly describes the technical solutions in the embodiments of the present disclosure with reference to the accompanying drawings in the embodiments of the present disclosure. The embodiments described are merely a part rather than all of the embodiments of the present disclosure. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present disclosure without creative efforts shall fall within the protection scope of the present disclosure.
The following describes a clustering method for a point of interest provided in an embodiment of the present disclosure. Referring to
101. Acquire a locating point set of a user within a preset period.
The locating point set in this embodiment of the present disclosure includes one or more locating points, and the locating point is used to indicate location information of the user.
Optionally, the locating point is a GPS location point represented by a longitude value and a latitude value, and a clustering apparatus for a point of interest acquires the locating point set of the user within a preset period from GPS location data of the user.
102. Generate a stay point set according to the foregoing locating point set.
Each stay point in the foregoing stay point set represents one hot area, where the foregoing hot area meets the following conditions: a distance between geographic locations of any two locating points in the foregoing hot area is less than a higher locating precision in locating precisions of the foregoing two locating points, and a maximum value of a time interval between locating points in the foregoing hot area is greater than a preset time threshold.
For example, it is assumed that a locating point set {Pj, Pj+1, . . . , Pj+L} constitutes a hot area in which a stay point may be formed, and then the locating point set {Pj, Pj+1, . . . , Pj+L} needs to meet the following two conditions.
1. A distance between geographic locations of any two locating points in {Pj, Pj+1, . . . , Pj+L} is less than a spatial threshold Dth, where Dth is not a fixed value, and it varies as related locating points change, and is a maximum value in locating precisions of two locating points, that is, Dth(Pn,Pm)=max(Radius(Pn),Radius(Pm)). For example, it is assumed that a locating precision of a locating point P1 is 5 meters, a locating precision of a locating point) P2 is 10 meters, and then for the locating point P1 and the locating point P2, Dth(P1,P2) is equal to a higher locating precision in locating precisions of the locating point P1 and the locating point P2, that is, Dth(P1,P2) is equal to 10 meters. If a locating precision of a locating point P3 is 15 meters, for the locating point P1 and the locating point P3, Dth(P1, P3) is equal to a higher locating precision in locating precisions of the locating point P1 and the locating point P3, that is, Dth(P1,P2) is equal to 15 meters.
2. Because different locating points are obtained by means of measurement at different time points, a time interval exists between every two locating points. For example, it is assumed that the locating point P1 is obtained by means of measurement at 10:00, the locating point P2 is obtained by means of measurement at 10:03, and then a time interval between the locating point P1 and the locating point P2 is three minutes. In this embodiment of the present disclosure, the {Pj, Pj+1, . . . , Pj+L} further needs to meet the following condition: a maximum value of a time interval between locating points in the {Pj, Pj+1, . . . , Pj+L} is greater than a preset time threshold Tth (for example, a value of Tth may be 5 minutes, 7 minutes, or 10 minutes), that is, in the {Pj, Pj+1, . . . , Pj+L}, a time interval between a locating point that is firstly obtained by means of measurement and a locating point that is lastly obtained by means of measurement is greater than Tth, where Tth may be set according to an actual requirement, which is not limited herein.
Optionally, after a hot area that meets the foregoing conditions is determined, a geometric center point of the foregoing hot area is determined as a stay point that represents the foregoing hot area, where the stay point may represent all locating points in the hot area. Certainly, another manner may also be used to determine a stay point corresponding to a hot area. For example, a center of gravity of a hot area is used as a stay point that represents the hot area, where the center of gravity of the hot area is related to a distribution of locating points within the hot area.
103. Calculate a confidence level of each stay point in the foregoing stay point set.
A lower average speed corresponding to movement states of all locating points in a hot area represented by a stay point indicates a higher confidence level of the stay point.
In an implementation manner, the clustering apparatus for a point of interest acquires movement states of all locating points included in a hot area represented by each stay point in the foregoing stay point set, and calculates the confidence level of each stay point in the foregoing stay point set according to the movement states of all locating points included in the hot area represented by each stay point in the foregoing stay point set and by using a formula
where Confi represents a confidence level of a stay point i, n represents n possible movement states, Wk represents a confidence level weight of the kth movement state, and nk represents a quantity of locating points that are in a hot area represented by the stay point i and whose movement states are the kth movement state, where each movement state is corresponding to a confidence level weight, and a lower movement speed corresponding to a movement state indicates a larger confidence level weight of the movement state. For example, it is assumed that a possible movement state of the user is stationary, walking, or riding, and then for the three movement states, because movement speeds that are corresponding to the movement states and are in ascending order are sequentially as follows: stationary, walking, and riding, confidence level weights that are of the movement states and are in descending order are sequentially as follows: stationary, walking, and riding.
Optionally, the clustering apparatus for a point of interest acquires the movement states of all locating points in the foregoing locating point set according to sensor data (such as an acceleration and data acquired from a gyroscope) on a terminal (such as a mobile phone, a tablet, and a vehicle-mounted terminal) of the foregoing user or a change of signal strength and a change of a signal quantity of a Wi-Fi network; or the clustering apparatus for a point of interest may acquire the movement states of all locating points included in the hot area represented by each residence point from another locating device (such as a server), which is not limited herein.
104. Obtain a trusted stay point from the foregoing stay point set by means of screening according to the confidence level of each stay point in the foregoing stay point set.
The confidence level of the foregoing trusted stay point is greater than a preset confidence level threshold.
In this embodiment of the present disclosure, the clustering apparatus for a point of interest obtains the trusted stay point (that is, a stay point whose confidence level is greater than the preset confidence level threshold) from the foregoing stay point set by means of screening, and removes a stay point whose confidence level is not greater than the preset confidence level threshold.
105. Cluster density-connected trusted stay points to form a POI.
The foregoing “density-connected” indicates that ranges of hot areas represented by two trusted stay points are directly connected or indirectly connected each other. For example, it is assumed that a coverage range of a trusted stay point P1 intersects with a coverage range of a trusted stay point P2, and then ranges of hot areas represented by the trusted stay point P1 and the trusted stay point P2 are directly connected each other. In this case, the trusted stay point P1 and the trusted stay point P2 are density-connected. Alternatively, it is assumed that a coverage range of a trusted stay point P1 intersects with a coverage range of a trusted stay point P3, a coverage range of a trusted stay point P2 intersects with a coverage range of a trusted stay point P4, and the coverage range of the trusted stay point P3 intersects with the coverage range of the trusted stay point P4, and then ranges of hot areas represented by the trusted stay point P1 and the trusted stay point P2 are indirectly connected each other. In this case, the trusted stay point P1 and the trusted stay point P2 are also density-connected.
Optionally, if the trusted stay point P1 (which may be a leaf stay point, or may be an inner stay point) is density-reachable from an inner stay point o, and the trusted stay point P2 (which may be a leaf stay point, or may be an inner stay point) is density-reachable from the inner stay point o, the trusted stay point P1 and the trusted stay point P2 are density-connected. The following describes the “inner stay point”, the “leaf stay point”, and the “density-reachable”.
Inner stay point: a sum of trusted levels of all stay points within preset radius coverage whose center is the stay point is greater than a preset threshold.
Leaf stay point: a sum of trusted levels of all stay points within preset radius coverage whose center is the stay point is not greater than a preset threshold.
Density-reachable: in a given series of trusted stay points p1, p2, . . . , pn, where pi (0<i<n) must be an inner stay point, pn may be a leaf stay point or an inner stay point, P=pi, and Q=pn, it is assumed that pi is directly density-reachable from pi-1 (1<i<(n+1)), and then a stay point Q is density-reachable from an inner stay point P.
Directly density-reachable: if a trusted stay point P (which may be a leaf stay point, or may be an inner stay point) is located within radius coverage of a trusted stay point Q, and the trusted stay point Q is an inner stay point, the trusted stay point P is directly density-reachable from the trusted stay point Q.
Optionally, all density-connected trusted stay points are clustered to form a POI. In addition, a closed area that is formed by sequentially connecting all leaf stay points among all the density-connected trusted stay points as an area of the POI, or an area of the POI may be a minimum circular area or square area that includes all the density-connected trusted stay points, which is not limited herein.
It can be learned from the foregoing that, in this embodiment of the present disclosure, one stay point is formed by multiple locating points according to a locating precision and a time threshold, and most layover points, locating jump points, and pass-through points can be filtered out under the constraint of the locating precision and the time threshold. In addition, a confidence level of a stay point is calculated with reference to movement states of locating points of the stay point, and a stay point whose confidence level is relatively low is screened out according to the confidence level of the stay point, so that some on-the-way noise stay points (such as pseudo stay points generated at the time of staying in a traffic-jammed section, waiting for a traffic light, and slowly walking) can further be filtered out. Therefore, a point of interest that is finally formed by clustering density-connected trusted stay points has higher reliability and higher reference value.
For a better understanding of the technical solutions of the present disclosure, the following describes the clustering method for a point of interest in this embodiment of the present disclosure by using a specific application scenario.
It is assumed that a track of a locating point set of a user within a preset period is shown in
Step 1: A clustering apparatus for a point of interest determines, from the locating point set that is shown in
Step 2: The clustering apparatus for a point of interest calculates a confidence level of each stay point according to movement states (such as stationary, walking, and riding) of the user at all locating points that form one stay point, where if an average speed corresponding to movement states of the user at all locating points of one stay point is lower, a confidence level of the stay point is higher, for example, if a percentage of locating points, of a stay point, whose movement states are stationary (that is, a movement speed is 0) in all locating points that form the stay point is higher, a confidence level of the stay point is higher; otherwise, if a percentage of locating points, of a stay point, whose movement states are riding (that is, a movement speed is far greater than 0) in all locating points that form the stay point is higher, a confidence level of the stay point is lower; and after calculating the confidence level of each stay point, removes, from the stay point set shown in
Step 3: Cluster the foregoing stay points based on confidence levels, identify a maximum stay point set in which all stay points are density-connected (that is, any two stay points in the set are density-connected), and cluster residence points in the maximum stay point set to form a POI, where an area of the POI may be represented by using a closed polygon that is formed by sequentially connecting all leaf stay points in the maximum stay point set, such as a schematic diagram of a POI that is obtained by means of clustering and is shown in
An embodiment of the present disclosure further provides a clustering apparatus for a point of interest. As shown in
The acquiring unit 301 is configured to acquire a locating point set of a user within a preset period.
The locating point set in this embodiment of the present disclosure includes one or more locating points, and the locating point is used to indicate location information of the user.
Optionally, the locating point is a GPS location point represented by a longitude value and a latitude value, and the clustering apparatus for a point of interest acquires the locating point set of the user within the preset period from GPS location data of the user.
The stay point generating unit 302 is configured to generate a stay point set according to the locating point set acquired by the acquiring unit 301, where each stay point in the foregoing stay point set represents one hot area, and the foregoing hot area meets the following conditions: a distance between geographic locations of any two locating points in the foregoing hot area is less than a higher locating precision in locating precisions of the foregoing two locating points, and a maximum value of a time interval between locating points in the foregoing hot area is greater than a preset time threshold.
Optionally, on the basis of the embodiment shown in
The calculating unit 303 is configured to calculate a confidence level of each stay point in the foregoing stay point set, where a lower average speed corresponding to movement states of all locating points in a hot area represented by a stay point indicates a higher confidence level of the stay point.
Optionally, on the basis of the embodiment shown in
and according to the movement states that are of all locating points included in the hot area represented by each stay point and are acquired by the foregoing acquiring subunit, where Confi represents a confidence level of a stay point i, n represents n possible movement states, Wk represents a confidence level weight of the kth movement state, and nk represents a quantity of locating points that are in a hot area represented by the stay point i and whose movement states are the kth movement state, where each movement state is corresponding to a confidence level weight, and a lower movement speed corresponding to a movement state indicates a larger confidence level weight of the movement state.
Optionally, the foregoing acquiring subunit is configured to acquire the movement states of all locating points included in the hot area represented by each residence point according to sensor data (such as an acceleration and data acquired from a gyroscope) on a terminal of the foregoing user or a change of signal strength and a change of a signal quantity of a Wi-Fi network.
The filtering unit 304 is configured to obtain a trusted stay point from the foregoing stay point set by means of screening according to the confidence level that is of each stay point in the foregoing stay point set and is calculated by the calculating unit 303, where a confidence level of the foregoing trusted stay point is greater than a preset confidence level threshold.
The clustering unit 305 is configured to cluster density-connected trusted stay points to form a point of interest, where the “density-connected” indicates that ranges of hot areas represented by two trusted stay points are directly connected or indirectly connected each other.
Optionally, the clustering unit 305 is further configured to cluster all density-connected trusted stay points to form a point of interest, and determine that a closed area that is formed by sequentially connecting all leaf stay points among all density-connected trusted stay points as an area of a point of interest, where a sum of confidence levels of all stay points within a preset radius coverage range whose center is a leaf stay point is not greater than a preset threshold.
It should be noted that the clustering apparatus for a point of interest in this embodiment of the present disclosure may be a terminal (such as a mobile phone, a tablet, a notebook computer, a desktop computer, or another terminal that has a locating function), or the clustering apparatus for a point of interest may be an apparatus that is independent of a terminal and can communicate with the terminal in a wired or wireless manner, which is not limited herein.
It should be noted that the clustering apparatus for a point of interest in this embodiment of the present disclosure may be a clustering apparatus for a point of interest in the foregoing method embodiment, and can be configured to implement all technical solutions in the foregoing method embodiment. A function of each function module of the clustering apparatus for a point of interest may be implemented according to a method described in the foregoing method embodiment. For a specific implementation process, reference may be made to related descriptions in the foregoing embodiment, and details are not described herein again.
It can be learned from the foregoing that, in this embodiment of the present disclosure, one stay point is formed by multiple locating points according to a locating precision and a time threshold, and most layover points, locating jump points, and pass-through points can be filtered out under the constraint of the locating precision and the time threshold. In addition, a confidence level of a stay point is calculated with reference to movement states of locating points of the stay point, and a stay point whose confidence level is relatively low is screened out according to the confidence level of the stay point, so that some on-the-way noise stay points (such as pseudo stay points generated at the time of staying in a traffic-jammed section, waiting for a traffic light, and slowly walking) can further be filtered out. Therefore, a point of interest that is finally formed by clustering density-connected trusted stay points has higher reliability and higher reference value.
An embodiment of the present disclosure further provides a computer storage medium, where the computer storage medium stores a program, and the program performs a part or all of steps recorded in the foregoing method embodiment.
The following describes another clustering apparatus for a point of interest that is provided in this embodiment of the present disclosure. Referring to
The processor 504 performs the following steps: acquiring a locating point set of a user within a preset period; generating a stay point set according to the foregoing locating point set, where each stay point in the foregoing stay point set represents one hot area, and the foregoing hot area meets the following conditions: a distance between geographic locations of any two locating points in the foregoing hot area is less than a higher locating precision in locating precisions of the foregoing two locating points, and a maximum value of a time interval between locating points in the foregoing hot area is greater than a preset time threshold; calculating a confidence level of each stay point in the foregoing stay point set, where a lower average speed corresponding to movement states of all locating points in a hot area represented by a stay point indicates a higher confidence level of the stay point; obtaining a trusted stay point from the foregoing stay point set by means of screening according to the confidence level of each stay point in the foregoing stay point set, where a confidence level of the foregoing trusted stay point is greater than a preset confidence level threshold; and clustering density-connected trusted stay points to form a point of interest, where the foregoing “density-connected” indicates that ranges of hot areas represented by two trusted stay points are directly connected or indirectly connected each other.
Optionally, the locating point is a GPS location point represented by a longitude value and a latitude value, and the processor 504 is configured to acquire the locating point set of the user within the preset period from GPS location data of the user.
Optionally, after determining a hot area that meets the foregoing conditions, the processor 504 determines a geometric center point of the foregoing hot area as a stay point that represents the foregoing hot area, where the stay point may represent all locating points in the hot area. Certainly, another manner may also be used to determine a stay point corresponding to a hot area. For example, a center of gravity of a hot area is used as a stay point that represents the hot area, where the center of gravity of the hot area is related to a distribution of locating points within the hot area.
Optionally, the processor 504 acquires movement states of all locating points included in a hot area represented by each stay point in the foregoing stay point set; and calculates the confidence level of each stay point in the foregoing stay point set according to the movement states of all locating points included in the hot area represented by each stay point in the foregoing stay point set and by using a formula
where Confi represents a confidence level of a stay point i, n represents n possible movement states, Wk represents a confidence level weight of the kth movement state, and nk represents a quantity of locating points that are in a hot area represented by the stay point i and whose movement states are the kth movement state, where each movement state is corresponding to a confidence level weight, and a lower movement speed corresponding to a movement state indicates a larger confidence level weight of the movement state. For example, it is assumed that a possible movement state of the user is stationary, walking, or riding, and then for the three movement states, because movement speeds that are corresponding to the movement states and are in ascending order are sequentially as follows: stationary, walking, and riding, confidence level weights that are of the movement states and are in descending order are sequentially as follows: stationary, walking, and riding.
Optionally, the processor 504 acquires the movement states of all locating points in the foregoing locating point set according to sensor data (such as an acceleration and data acquired from a gyroscope) on a terminal of the foregoing user or a change of signal strength and a change of signal quantity of a Wi-Fi network; or the processor 504 may acquire the movement states of all locating points included in the hot area represented by each residence point from another locating device (such as a server), which is not limited herein.
Optionally, the processor 504 clusters all density-connected trusted stay points to form a POI, and determines that a closed area that is formed by sequentially connecting all leaf stay points among all the density-connected trusted stay points is an area of the POI, or an area of the POI may be a minimum circular area or square area that includes all the density-connected trusted stay points, which is not limited herein.
It should be noted that the clustering apparatus for a point of interest in this embodiment of the present disclosure may be a terminal (such as a mobile phone, a tablet, a notebook computer, a desktop computer, or another terminal that has a locating function), or the clustering apparatus for a point of interest may be an apparatus that is independent of a terminal and can communicate with the terminal in a wired or wireless manner, which is not limited herein.
It should be noted that the clustering apparatus for a point of interest in this embodiment of the present disclosure may be a clustering apparatus for a point of interest in the foregoing method embodiment, and may be configured to implement all technical solutions in the foregoing method embodiment. Functions of each function module of the clustering apparatus for a point of interest may be implemented according to a method described in the foregoing method embodiment. For a specific implementation process, reference may be made to related descriptions in the foregoing embodiment, and details are not described herein again.
It can be learned from the foregoing that, in this embodiment of the present disclosure, one stay point is formed by multiple locating points according to a locating precision and a time threshold, and most layover points, locating jump points, and pass-through points can be filtered out under the constraint of the locating precision and the time threshold. In addition, a confidence level of a stay point is calculated with reference to movement states of locating points of the stay point, and a stay point whose confidence level is relatively low is screened out according to the confidence level of the stay point, so that some on-the-way noise stay points (such as pseudo stay points generated at the time of staying in a traffic-jammed section, waiting for a traffic light, and slowly walking) can further be filtered out. Therefore, a point of interest that is finally formed by clustering density-connected trusted stay points has higher reliability and higher reference value.
It should be noted that, for brief description, the foregoing method embodiments are represented as a series of actions. However, a person skilled in the art should appreciate that the present disclosure is not limited to the described order of the actions, because according to the present disclosure, some steps may be performed in other orders or simultaneously. In the foregoing embodiments, the description of each embodiment has respective focuses. For a part that is not described in detail in an embodiment, reference may be made to related descriptions in other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the described apparatus embodiment is merely exemplary. For example, the unit division is merely logical function division and may be other division in actual implementation. For example, a plurality of units or components may be combined or integrated into another system, or some features may be ignored or not performed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections may be implemented by using some interfaces. The indirect couplings or communication connections between the apparatuses or units may be implemented in electronic, mechanical, or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each of the units may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in a form of hardware, or may be implemented in a form of a software functional unit.
When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, the integrated unit may be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the present disclosure essentially, or the part contributing to the prior art, or all or some of the technical solutions may be implemented in the form of a software product. The software product is stored in a storage medium and includes several instructions for instructing a computer device (which may be a personal computer, a server, or a network device) to perform all or some of the steps of the methods described in the embodiments of the present disclosure. The foregoing storage medium includes: any medium that can store program code, such as a USB flash drive, a removable hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.
The foregoing describes in detail a clustering method for a point of interest and a related apparatus that are provided in the present disclosure. A person of ordinary skill in the art may, based on the idea of the present disclosure, make modifications with respect to the specific implementation manners and the application scope. Therefore, the content of this specification shall not be construed as a limitation to the present disclosure.
While this invention has been described with reference to illustrative embodiments, this description is not intended to be construed in a limiting sense. Various modifications and combinations of the illustrative embodiments, as well as other embodiments of the invention, will be apparent to persons skilled in the art upon reference to the description. It is therefore intended that the appended claims encompass any such modifications or embodiments.
Number | Date | Country | Kind |
---|---|---|---|
2013 1 0552636 | Nov 2013 | CN | national |
This application is a continuation of International Application No. PCT/CN2014/088443, filed on Oct. 13, 2014, which claims priority to Chinese Patent Application No. 201310552636.2, filed on Nov. 7, 2013, both of which are hereby incorporated by reference in their entireties.
Number | Name | Date | Kind |
---|---|---|---|
5844522 | Sheffer | Dec 1998 | A |
8301371 | Sheha et al. | Oct 2012 | B2 |
8335990 | Arrasvuori | Dec 2012 | B2 |
8339399 | Snow | Dec 2012 | B2 |
8358224 | Seder et al. | Jan 2013 | B2 |
20010022558 | Karr, Jr. | Sep 2001 | A1 |
20060181505 | Dort | Aug 2006 | A1 |
20070219706 | Sheynblat | Sep 2007 | A1 |
20080186234 | Alles et al. | Aug 2008 | A1 |
20080268816 | Wormald | Oct 2008 | A1 |
20100069035 | Johnson | Mar 2010 | A1 |
20100189312 | Nielsen | Jul 2010 | A1 |
20110047509 | Arrasvuori | Feb 2011 | A1 |
20110151898 | Chandra et al. | Jun 2011 | A1 |
20120047152 | Purdy | Feb 2012 | A1 |
20120203732 | Oono | Aug 2012 | A1 |
20120239607 | Rao et al. | Sep 2012 | A1 |
20130262479 | Liang et al. | Oct 2013 | A1 |
20130273937 | Nakahara | Oct 2013 | A1 |
20130339498 | Johnson | Dec 2013 | A1 |
Number | Date | Country |
---|---|---|
101742545 | Jun 2010 | CN |
102595323 | Jul 2012 | CN |
102682041 | Sep 2012 | CN |
103218442 | Jul 2013 | CN |
2009116541 | May 2009 | JP |
2011171876 | Sep 2011 | JP |
2007115240 | Oct 2007 | WO |
2008121872 | Oct 2008 | WO |
2011046113 | Apr 2011 | WO |
2012080787 | Jun 2012 | WO |
2012096175 | Jul 2012 | WO |
2013060925 | May 2013 | WO |
Entry |
---|
Ashbrook, D. et al.: “Learning Significant Locations and Predicting User Movement With GPS,” Sixth International Symposium on Wearable Computers, Oct. 2002, pp. 101-108. |
Khetarpaul, S. et al.: “Mining GPS Data to Determine Interesting Locations,” Mar. 28, 2011, pp. 1-6. |
Minh, T., et al.; “Contextual Conditional Models for Smartphone-Based Human Mobility Prediction,” Sep. 2012, pp. 1-10. |
Zheng, Y. et al.: “Mining Interesting Locations and Travel Sequences From GPS Trajectories,” Apr. 20-24, 2009, pp. 1-10, Madrid, Spain. |
Gambs, S. et al.: “Next Place Prediction Using Mobility Markov Chains,” Apr. 10, 2012, pp. 1-6, Bern, Switzerland. |
Number | Date | Country | |
---|---|---|---|
20160253407 A1 | Sep 2016 | US |
Number | Date | Country | |
---|---|---|---|
Parent | PCT/CN2014/088443 | Oct 2014 | US |
Child | 15148365 | US |