Method and device for determining expired interest point, electronic equipment and storage medium

Abstract

The method comprises: acquiring periodic heat data of a detection block constituting a region to be detected; determining the detection block in which the periodic heat data abnormally fluctuates within a first preset time period as an event burst detection block; determining heat prediction data of the detection block in a current time period; wherein the heat prediction data is predicted based on periodic heat data of a period before the current time period; comparing the heat prediction data with actual periodic heat data of the current time period, and determining the detection block in which the comparison result meets a first set condition as a candidate detection block; and determining a region constituted by the candidate detection blocks after the event burst detection block is removed from the candidate detection blocks as a region in which an expired interest point exists.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of map data, and particularly relates to a determination method and device of expired POI, electronic equipment and storage medium. BACKGROUND

[0002] With the development of location-based services (LBS), more and more application software integrates service capabilities related to electronic maps. For example, a user can search for a point of interest (POI) in an electronic map through a map navigation application software or a car-hailing application software, and obtain information related to the POI or plan a navigation route to the POI, wherein the POI in the electronic map represents a residential complex, a shop, a subway station, a public toilet and the like in the real world.

[0003] Since the POI represents a location in the real world, and the state of the location changes over time, the POI in the electronic map needs to be updated according to the actual changes of the place, so that the expired (invalid) POI due to reasons such as closure, relocation and decoration of the place can be timely offline, so as to ensure the accuracy of the POI in the electronic map. As known, there are a large number of POIs in the electronic map, and how to efficiently and accurately mine the expired POI from the large number of POIs is one of the technical problems to be solved by those skilled in the art. SUMMARY

[0004] The present disclosure provides a determination method and device of expired POI, electronic equipment and computer readable storage medium.

[0005] In a first aspect, the present disclosure provides a determination method of expired POI, comprising:

[0006] acquiring periodic heat data of a detection block constituting a detection area to be detected;

[0007] determining the detection block in which the periodic heat data abnormally fluctuates in a first preset time period as an event burst detection block;

[0008] determining heat prediction data of the detection block in a current time period, wherein the heat prediction data is predicted based on periodic heat data of a period before the current time period;

[0009] comparing the heat prediction data with actual periodic heat data of the current time period, and determining the detection block whose comparison result meets a first set condition as a candidate detection block;

[0010] After the event burst detection block is removed from the candidate detection blocks, the remaining candidate detection blocks constitute a region determined as a region with expired points of interest.

[0011] Further, the method further comprises:

[0012] dividing the detection block into a plurality of sub-blocks;

[0013] counting periodic heat data of the sub-blocks;

[0014] determining periodic heat data of a detection block containing the sub-blocks according to the periodic heat data of the sub-blocks.

[0015] Further, the method further comprises:

[0016] acquiring a hotspot road;

[0017] determining a sub-block intersecting with the hotspot road as a candidate sub-block;

[0018] deleting, from sub-blocks constituting a detection block, a candidate sub-block whose periodic heat data satisfies a second set condition;

[0019] counting periodic heat data of the remaining sub-blocks in the detection block;

[0020] determining the periodic heat data of the detection block based on the periodic heat data of the remaining sub-blocks in the detection block.

[0021] Further, determining a sub-block intersecting with the hotspot road as a candidate sub-block comprises:

[0022] determining a hotspot road within a coverage range of the detection block according to point coordinates in the detection block;

[0023] judging whether an end point of a road segment constituting the hotspot road is located in a sub-block of the detection block;

[0024] determining a diagonal segment of a sub-block of the detection block, and judging whether a road segment constituting the hotspot road intersects with the diagonal segment;

[0025] determining, as a candidate sub-block, a sub-block in which an end point of a road segment is located, or a sub-block in which a road segment intersects with a diagonal segment.

[0026] Further, the method further comprises:

[0027] determining a road within the detection block;

[0028] determining periodic heat data of the road intersecting with a sub-block of the detection block within a second preset time period;

[0029] determining the sub-block as a hotspot road sub-block when a difference between periodic thermal data of the sub-block in the second preset time period and periodic thermal data of the road intersecting the sub-block in the second preset time period satisfies a third set condition;

[0030] eliminating the hotspot road sub-block in the detection block, and counting periodic thermal data of the remaining sub-blocks;

[0031] determining the periodic thermal data of the detection block based on the periodic thermal data of the remaining sub-blocks of the detection block.

[0032] Further, determining the detection block as an event burst detection block when the periodic thermal data of the detection block in a first preset time period abnormally fluctuates, includes:

[0033] For each of the detection blocks, clustering the periodic thermal data of the detection block in the first preset time period, and determining one of two classification sets obtained by clustering as an abnormal classification set when the periodic thermal data in the one classification set is more than that in the other classification set;

[0034] determining an abnormal period statistical value according to the abnormal classification set; wherein the abnormal period statistical value is a statistical number of periodic thermal data appearing in the abnormal classification set and being continuous in time period;

[0035] determining whether the abnormal period count sequence is an unstable sequence according to a number of abnormal period statistical values contained in the abnormal period count sequence and a maximum value of the abnormal period statistical values, the abnormal period count sequence containing abnormal period statistical values sorted by time;

[0036] determining the detection block corresponding to the unstable sequence as the event burst detection block.

[0037] Further, determining whether the abnormal period count sequence is an unstable sequence according to a number of abnormal period statistical values contained in the abnormal period count sequence and a maximum value of the abnormal period statistical values, includes:

[0038] determining the abnormal period count sequence as a short period unstable sequence when the number of abnormal period statistical values is less than a first threshold value and the maximum value of the abnormal period statistical values is less than a second threshold value;

[0039] determining the abnormal period count sequence as a long period unstable sequence when the number of abnormal period statistical values is equal to a third threshold value and the maximum value of the abnormal period statistical values is greater than or equal to a fourth threshold value and less than a fifth threshold value.

[0040] Further, determining thermal prediction data of the detection block in a current time period, includes:

[0041] The time series decomposition method and the periodic thermal data of the detection block in the first time period are used to predict the thermal prediction data of the detection block in the current time period.

[0042] Further, the thermal prediction data is compared with the actual periodic thermal data of the current time period, and the detection block satisfying a first set condition in the comparison result is determined as a candidate detection block, including:

[0043] When a weighted value obtained by multiplying the thermal prediction data by a preset weight is greater than the actual periodic thermal data, the detection block is determined as the candidate detection block; wherein the preset weight is less than 1.

[0044] In a second aspect, an expired interest point determination apparatus is provided in an embodiment of the present application, and the apparatus includes:

[0045] A first acquisition module is configured to acquire periodic thermal data of a detection block constituting a detection area to be detected;

[0046] A first determination module is configured to determine the detection block in which the periodic thermal data has an abnormal fluctuation in a first preset time period as an event burst detection block;

[0047] A second determination module is configured to determine thermal prediction data of the detection block in a current time period; wherein the thermal prediction data is predicted based on periodic thermal data of a previous period before the current time period;

[0048] A third determination module is configured to compare the thermal prediction data with actual periodic thermal data of the current time period, and determine the detection block satisfying a first set condition in the comparison result as a candidate detection block;

[0049] A fourth determination module is configured to determine, after excluding the event burst detection block from the candidate detection block, a region constituted by the remaining candidate detection block as a region in which an expired interest point exists.

[0050] The functions can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0051] In one possible design, the structure of the apparatus includes a memory and a processor, the memory is used to store one or more computer instructions supporting the apparatus to execute the corresponding method, and the processor is configured to execute the computer instructions stored in the memory. The apparatus can also include a communication interface for communication between the apparatus and other devices or communication networks.

[0052] In a third aspect, the embodiments of the present disclosure provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method of any one of the above aspects.

[0053] In a fourth aspect, the embodiments of the present disclosure provide a computer readable storage medium for storing computer instructions for the above-mentioned any device, which, when executed by a processor, is configured to implement the steps of the method of any one of the above aspects.

[0054] In a fifth aspect, the embodiments of the present disclosure provide a computer program product comprising computer instructions, which, when executed by a processor, is configured to implement the steps of the method of any one of the above aspects.

[0055] The technical solutions provided by the embodiments of the present disclosure can include the following beneficial effects:

[0056] When the expired POI in the to-be-detected area is mined, such as a region that has been relocated or demolished, the embodiments of the present disclosure divide the to-be-detected area into multiple detection blocks, and for each detection block, the periodic heat data in each time period in the detection block is counted, and then the periodic heat data in the past multiple time periods is used to predict the heat prediction value in the current time period, and the heat prediction value is compared with the actual periodic heat detection data in the current time period to screen out candidate detection blocks from the multiple detection blocks in the to-be-detected area, and then the event burst detection block that may have a sudden event is eliminated from the candidate detection blocks, and the region where the remaining candidate detection blocks are located is determined as a region where there may be an expired POI. The embodiments of the present disclosure divide the to-be-detected area into detection blocks with smaller granularity, and determine the region in the to-be-detected area where there may be an expired POI based on the periodic heat data in the detection block. Compared with the way of predicting the expired POI by training a model in the prior art, the embodiments of the present disclosure can improve the timeliness of detecting the expired POI, reduce the detection cost, and improve the detection accuracy of the expired POI based on the statistical real heat data.

[0057] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0058] Other features, objects, and advantages of the present disclosure will become more apparent from the following detailed description of the non-limiting embodiments, taken in conjunction with the accompanying drawings. In the drawings:

[0059] Figure 1 A flowchart of a method for determining an expired POI according to an embodiment of the present disclosure is shown;

[0060] Figure 2This diagram illustrates the process of mining the target region containing expired POIs in the region to be detected according to an embodiment of the present disclosure.

[0061] Figure 3 This diagram illustrates an application scenario for mining expired Points of Interest (POIs) in electronic maps according to an embodiment of the present disclosure.

[0062] Figure 4 A structural block diagram of an apparatus for determining expired points of interest according to an embodiment of the present disclosure is shown.

[0063] Figure 5 This is a schematic diagram of the structure of an electronic device suitable for implementing a method for determining expired points of interest according to an embodiment of the present disclosure. Detailed Implementation

[0064] In the following, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement them. Furthermore, for clarity, portions unrelated to the description of the exemplary embodiments have been omitted from the drawings.

[0065] In this disclosure, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of features, figures, steps, behaviors, components, parts or combinations thereof disclosed in this specification, and are not intended to exclude the possibility of the presence or addition of one or more other features, figures, steps, behaviors, components, parts or combinations thereof.

[0066] It should also be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0067] The details of the embodiments of this disclosure are described in detail below through specific examples.

[0068] Figure 1 A flowchart illustrating a method for determining expired points of interest according to an embodiment of this disclosure is shown. Figure 1 As shown, the method for determining expired points of interest includes the following steps:

[0069] In step S101, periodic thermal data of the detection blocks constituting the area to be detected are acquired;

[0070] In step S102, the detection block in which the periodic thermal data fluctuates abnormally within the first preset time period is identified as an event outbreak detection block;

[0071] In step S103, the thermal prediction data of the detection block for the current time period is determined; wherein, the thermal prediction data is predicted based on the periodic thermal data of previous periods.

[0072] In step S104, thermal prediction data of the detection block in the current time period is determined; wherein the thermal prediction data is predicted based on the periodic thermal data of the period before the current time period under the first set condition.

[0073] In step S105, thermal prediction data of the detection block in the current time period is determined; wherein the thermal prediction data is predicted based on the periodic thermal data of the period before the current time period.

[0074] In the embodiment, the to-be-detected region can correspond to a divided region such as a city, a town, a township, a village, etc. in the electronic map, or can be a region such as a business circle circled in a certain divided region. Of course, it can be understood that the to-be-detected region can also be an arbitrary circled region on the electronic map, which can be determined according to actual needs, and is not specifically limited here.

[0075] The embodiment of the present disclosure proposes a determination method of a target region for the to-be-detected region in the electronic map, in order to determine whether the labeled POI (Point of Interest) in the to-be-detected region has expired. Through the method, the target region in which the expired POI labeled on the electronic map is located can be mined from the to-be-detected region. The expired POI can be, for example, a labeled POI that has been relocated or demolished.

[0076] In the embodiment of the present disclosure, the to-be-detected region can be divided into a grid form according to the map latitude and longitude coordinates. Each grid obtained by the division can be referred to as a detection block. Therefore, one to-be-detected region can include multiple detection blocks. In some embodiments, the to-be-detected region on the map can be gridded by multiplying the map longitude and latitude by 4000 and then dividing by 3, and taking the integer. According to the spherical characteristics, the length of each grid after gridding is about 75 meters. The width and height of the grids with different longitude and latitude can also be different.

[0077] The thermal data can be the number of active devices in a predetermined time range and a regional range, which can be understood as devices in a powered-on state, such as user equipment, vehicles, and the like. The periodic thermal data of the detection block can be understood as the number of active devices in the region where the detection block is located in a set time period. In some embodiments, a time period can be set according to actual conditions, for example, a time period can be set as one month, and a time period can also be divided into multiple sub-periods. When statistical periodic thermal data, the number of active devices in the region where the detection block is located in each sub-period can be counted, and then the periodic thermal data corresponding to multiple sub-periods included in a time period can be summed to obtain the periodic thermal data in a time period. For example, a time period can be set as one month, and a sub-period of a time period can be set as one day. The number of devices appearing in the region where the detection block is located in one day can be counted by counting one device appearing in the region where the detection block is located in one day as 1, and then the thermal data of one day in one month can be summed to obtain the periodic thermal data of one month.

[0078] In some embodiments, the number of active devices in the region where the detection block is located can be determined by the received position positioning signal. The position positioning signal can be a GPS signal. It should be noted that when statistical thermal data of a sub-period, in order to avoid the case that a device staying in the region where the detection block is located for a long time is counted multiple times, multiple position positioning signals received by the same device in a GPS track can be counted as one device.

[0079] The periodic thermal data of all detection blocks constituting the region to be detected can be pre-counted and stored in the above manner. When the above method proposed in the embodiments of the present disclosure is executed, the periodic thermal data corresponding to each time period of all detection blocks constituting the region to be detected in the first preset time period can be directly obtained from the storage location. It should be noted that the first preset time period can include multiple time periods, for example, a time period can be one month, and the first preset time period can be multiple months.

[0080] It should be noted that, in normal cases, the periodic thermal data of a region in a plurality of continuous time periods will show a relatively balanced state, and there may be some small fluctuations occasionally, but there will not be too much change between the plurality of time periods, and it will basically remain in a relatively stable state. In some special cases, the periodic thermal data of a region can change greatly, for example, during a small holiday period, because students go out more due to school holidays, the periodic thermal data of the school region can produce a trough in a short period of time; for another example, during a road repair period of a certain region, due to the inconvenience of surrounding traffic and other reasons, the periodic thermal data can also produce a trough in a relatively long period of time. However, such changes are not caused by the relocation or demolition of the labeled POI in the region, so the region where such changes occur is not the target region to be mined by the embodiments of the present disclosure.

[0081] Therefore, in some embodiments, for each detection block in the to-be-detected region, whether the periodic thermal data in the first time period has an abnormal fluctuation can be determined by detecting whether the periodic thermal data in the first time period has a large mutation, and the detection block with an abnormal fluctuation is determined as an event burst detection block. It can be understood that such an event burst detection block is due to a burst event, and after the burst event, the region where the event burst detection block is located will return to a normal state, which is not the expired POI region to be mined by the embodiments of the present disclosure, such as the target region of relocation or demolition. Therefore, in subsequent steps, the event burst detection block detected by the abnormal fluctuation of the periodic thermal data can be excluded from the target region.

[0082] In some embodiments, a burst event can be understood as an event that causes the thermal data of a certain region to be abnormal. The burst event can be divided into a short-period burst event and a long-period burst event. The short-period burst event can be understood as a period in which the periodic thermal data is abnormal, and the long-period burst event can be understood as a period in which the periodic thermal data is abnormal. For example, the school holidays of May 1st or November 11th can be considered as a short-period burst event, and road repair can be considered as a long-period burst event because of the long period.

[0083] In some embodiments, in the process of excavating the target region, for each detection block in the to-be-detected region, the periodic heat data in the current time period can also be predicted by using the periodic heat data in the past multiple time periods in the first preset time period, i.e., heat prediction data. As described above, in the case where no relocation or demolition event occurs, the periodic heat data of a region in a period of time will not change greatly, and there will usually be some small fluctuations, but will generally remain within a certain range. If one or more labeled POIs are relocated or demolished, the periodic heat data of the region where the relocated or demolished labeled POI is located will change greatly compared to before the relocation or demolition. Therefore, the periodic heat data of the region where the detection block is located in the current time period under normal circumstances (i.e., in the case where no POI relocation or demolition occurs) is predicted by using the periodic heat data in the past multiple time periods in the first preset time period, i.e., heat prediction data. In the case where the actual periodic heat data in the current time period is consistent with or not greatly different from the heat prediction data, it can be determined that no labeled POI expiration has occurred in the detection block, and in the case where the actual periodic heat data in the current time period is greatly different from the heat prediction data, it can be considered that labeled POI expiration has occurred in the detection block in the current time period, and therefore the detection block can be determined as a candidate detection block.

[0084] It should be noted that the actual periodic heat data of the region where the detection block is located in the current time period can be obtained in the manner described above, i.e., the heat data in the multiple sub-periods included in the current time period can be first counted (which can be determined by counting the number of active devices detected in the sub-periods), and the heat data in the multiple sub-periods in the current time period is added to obtain the actual periodic heat data in the current time period.

[0085] In some embodiments, the first set condition can be set by comparing the difference between the actual periodic heat data of the region where the detection block is located in the current time period and the heat prediction value, i.e., by determining whether the comparison result of the heat prediction data and the actual periodic heat data satisfies the first set condition to determine whether the detection block is a candidate detection block. If the difference is large, e.g., greater than a preset value, it can be considered that the heat prediction data and the actual periodic heat data of the detection block in the current time period satisfy the first set condition, and if the difference is small, e.g., less than or equal to a preset value, it can be considered that the heat prediction data and the actual periodic heat data of the detection block in the current time period do not satisfy the first set condition.

[0086] In some embodiments, the candidate detection block determined in the above manner can be understood as a detection block in which the actual periodic heat data in the current time period has changed greatly compared to the past multiple time periods, and the event burst detection block mentioned above can also be included in the detection block. However, the event burst detection block will return to the normal state after the end of the burst event, that is, the event burst detection block is not the area in which the expired POI to be mined in the embodiments of the present disclosure, and therefore the event burst detection block can be excluded from the candidate detection block to obtain the target detection block. The area in which one or more target detection blocks are located is the area in which the expired POI exists and is to be mined in the embodiments of the present disclosure.

[0087] When the expired POI, such as a relocated or demolished area, in the to-be-detected area is mined, the embodiments of the present disclosure divide the to-be-detected area into multiple detection blocks, and for each detection block, the periodic heat data in each time period in the detection block is counted. Then, the heat prediction value in the current time period is predicted based on the periodic heat data in the past multiple time periods, and the candidate detection block is selected from the multiple detection blocks in the to-be-detected area by comparing the heat prediction value with the actual periodic heat detection data in the current time period. Then, the event burst detection block in which a burst event can occur is excluded from the candidate detection block, and the area in which the remaining candidate detection block is located is determined as the area in which the expired POI can exist. Compared with the way of predicting the expired POI by training a model in the prior art, the embodiments of the present disclosure can improve the detection timeliness of the expired POI, reduce the detection cost, and improve the detection accuracy of the expired POI based on the counted real heat data by dividing the to-be-detected area into detection blocks with smaller granularity and determining the area in which the expired POI can exist in the to-be-detected area based on the periodic heat data in the detection block.

[0088] The application examples of the determination method of the expired POI proposed in the embodiments of the present disclosure are described below.

[0089] If it is necessary to update the labeled POI in A city in the electronic map, in order to delete the expired POI from the electronic map, the determination method of the expired POI proposed in the embodiments of the present disclosure can be used to mine the expired POI in A city.

[0090] First, the area corresponding to A city in the electronic map is determined as the to-be-detected area, and the to-be-detected area can be grid-divided, and the grid obtained by the division is used as the detection block constituting the to-be-detected area.

[0091] The periodic heat data of each detection block can be pre-processed and stored. The periodic heat data of a detection block depends on the number of active devices in the detection block within a set time period, for example, the number of active devices in the detection block can be counted every day, and the periodic heat data of the detection block within a time period can be determined based on the number of active devices every day, such as the number of active devices within a time period.

[0092] Then, the detection blocks with abnormal fluctuations in periodic heat data within a first preset time period, for example, the last 36 months, are determined as time burst detection blocks based on the periodic heat data of all detection blocks constituting the A city. The abnormal fluctuations in periodic heat data can be understood as follows: within a long period of time, the periodic heat data in a detection block should remain stable. If the periodic heat data appears fluctuations such as peaks or troughs at a certain time point or time period, it can be determined that the periodic heat data in the detection block may have been caused by a sudden event, resulting in abnormal fluctuations at a certain time point or time period, while the periodic heat data at other time periods remains stable. Such detection blocks may have experienced some mutations in periodic heat data in a short period of time due to some sudden events such as school holidays, road repairs, etc., but such mutations are not caused by expired POIs, so such detection blocks can be excluded.

[0093] Secondly, the heat prediction data of each detection block in the A city in the current time period is counted, which can be predicted based on the previous periodic heat data. The heat prediction data of the same detection block is compared with the actual periodic heat data (counted by the number of active devices). If the heat prediction data predicted based on the previous periodic heat data is not much different from the actual periodic heat data, it can be considered that the actual periodic heat data of the detection block in the current time period remains stable compared with the periodic heat data in the previous time periods, i.e., the periodic heat data in the current time period has not changed much compared with the previous time periods. If the periodic heat data in the current time period changes greatly compared with the previous time periods, it can be considered that the heat data of the detection block in the current time period has changed significantly, and the reason for the significant change may be due to some mutation events or due to the appearance of expired POIs in the region. Therefore, such detection blocks are determined as candidate detection blocks, and the event burst detection blocks are excluded from the candidate detection blocks, and the remaining candidate detection blocks can be determined as regions with expired POIs.

[0094] Finally, the annotated POIs in the regions with expired POIs can be checked, and the expired POIs can be deleted from the electronic map after verification.

[0095] In an optional implementation of the embodiment, the method further comprises:

[0096] dividing the detection block into a plurality of sub-blocks;

[0097] counting periodic thermal data of the sub-blocks;

[0098] determining periodic thermal data of the detection block containing the sub-blocks according to the periodic thermal data of the sub-blocks.

[0099] In the optional implementation, in order to obtain more accurate results, the region to be detected is divided into a plurality of detection blocks, and the plurality of detection blocks are divided into a plurality of sub-blocks. As described above, the detection blocks can be obtained by grid division of the region to be detected, and each grid is a detection block. In addition, the detection blocks can be divided into finer sub-blocks, and the periodic thermal data of the sub-blocks is counted first, and then the periodic thermal data corresponding to each sub-block in the detection block is added to obtain the periodic thermal data in the detection block. The detection blocks are divided into a plurality of sub-blocks in order to facilitate the elimination of the influence of the road on the thermal data.

[0100] In some embodiments, the region to be detected on the map can be gridded by multiplying the map longitude and latitude by 4000 and then dividing by 3 and rounding. According to the spherical characteristics, the length of each grid after gridding is about 75 meters, and the width and height of the grids with different longitude and latitude can also be different.

[0101] In other embodiments, each detection block can be divided into 3x3=9 sub-blocks, and the length of each square block is about 25 meters.

[0102] In an optional implementation of the embodiment, the step of counting the periodic thermal data of the sub-blocks further comprises the following steps:

[0103] counting the number of active devices of the sub-blocks in each sub-period in units of sub-periods;

[0104] determining the periodic thermal data of the sub-blocks in a time period according to the number of devices; wherein the time period includes a plurality of sub-periods.

[0105] In the optional implementation, as described above, the periodic thermal data is the thermal data of a time period, and a time period can include a plurality of sub-periods. Therefore, when counting the periodic thermal data of the sub-blocks, the thermal data of the sub-periods can be counted, and then the thermal data of the plurality of sub-periods included in a time period is added to obtain the periodic thermal data of the sub-blocks in a time period.

[0106] In some embodiments, one time period can be one month, and one sub-period can be one day.

[0107] In some other embodiments, the number of devices receiving position positioning signals such as GPS signals from devices in the area where the sub-block is located, such as user devices, vehicle devices, etc., can be understood as the number of active devices in the sub-block, and the heat data of the sub-block in one day can be determined according to the number of active devices.

[0108] Through the above manner of the embodiments of the present disclosure, the heat data obtained by periodically counting the devices in the area in the process of returning GPS can be more real.

[0109] In one optional implementation of the present embodiment, the method further comprises the following steps:

[0110] Obtaining a hotspot road;

[0111] Determining a sub-block intersecting with the hotspot road as a candidate sub-block;

[0112] From the sub-blocks constituting the detection block, deleting the candidate sub-blocks whose periodic heat data meet the second set condition;

[0113] Counting the periodic heat data of the remaining sub-blocks in the detection block;

[0114] Determining the periodic heat data of the detection block based on the periodic heat data of the remaining sub-blocks in the detection block.

[0115] In the optional implementation, the hotspot road can be understood as a road whose periodic heat data is larger than that of the surrounding roads. In some embodiments, a threshold value can be set, and the road whose periodic heat data exceeds the threshold value can be determined as the hotspot road. In some other embodiments, the periodic heat data of the road can be determined by the traffic volume on the road in the time period, and the traffic volume can be determined by the number of people and vehicles passing through the road.

[0116] Therefore, whether each road is a hotspot road can be determined in advance, and when counting the periodic heat data of the detection block, the hotspot road is obtained, the sub-block intersecting with the hotspot road among the sub-blocks included in the detection block is determined as a candidate sub-block, and the candidate sub-blocks whose periodic heat data meet the second set condition are deleted from the sub-blocks constituting the detection block.

[0117] In some embodiments, the second set condition can be set as that the sum of periodic heat data of the candidate sub-block in the second preset time period is less than N times of the sum of periodic heat data of all roads intersecting with the candidate sub-block in the second preset time period, N≤2, that is, when the sum of periodic heat data of the candidate sub-block in the second preset time period is less than N times of the sum of periodic heat data of all roads intersecting with the candidate sub-block in the second preset time period, the candidate sub-block is determined as the candidate sub-block satisfying the second set condition.

[0118] After the candidate sub-blocks satisfying the second set condition are removed from the sub-blocks included in the detection block, the periodic heat data of the detection block can be obtained by statistical analysis of the periodic heat data of the remaining sub-blocks. The periodic heat data of the sub-blocks can be obtained by statistical analysis in the manner described above.

[0119] In an optional implementation of the embodiment, the step of determining the sub-blocks intersecting with the hotspot road as the candidate sub-blocks further includes the following steps:

[0120] determining the hotspot road within the coverage of the detection block according to the point coordinates in the detection block;

[0121] judging whether the end points of the road segments constituting the hotspot road are located in the sub-blocks of the detection block;

[0122] determining the diagonal segments of the sub-blocks of the detection block, and judging whether the road segments constituting the hotspot road intersect with the diagonal segments;

[0123] determining the sub-blocks in which the end points of the road segments are located, or the sub-blocks in which the road segments intersect with the diagonal segments, as the candidate sub-blocks.

[0124] In the optional implementation, one purpose of dividing the detection block into multiple sub-blocks is to determine the road covered by the detection block by using the intersection between the sub-blocks and the road.

[0125] The following illustrates the determination process of the intersection between the road and the sub-block.

[0126] Suppose that the road curve p in the electronic map is composed of a plurality of line segments, which can be represented as The point coordinates in the detection block are known; all the road curves in the area where the detection block is located are captured through the point coordinates in the detection block;

[0127] For any sub-block j in the detection block i, if 1≤j≤9, the four vertex coordinates of the sub-block j are taken to form two diagonal line segments, which are represented as

[0128] The line segments corresponding to each road curve p in the area where the detection block is located are determined. whether the end points of the line segment are within the region where the sub-block j is located, if the end points of the line segment are within the region where the sub-block j is located, the road p intersects with the sub-block j, otherwise if the end points of the line segment are not within the region where the sub-block j is located, the road p does not intersect with the sub-block j. In addition, it can also be determined whether the line segment intersects with the diagonal line segment , if the line segment intersects with the diagonal line segment , the road p intersects with the sub-block j, otherwise if the line segment does not intersect with the diagonal line segment , the road p does not intersect with the sub-block j.

[0129] Through the above manner, the sub-blocks intersecting with the hot road can be determined, and the sub-blocks are determined as the candidate sub-blocks. Through the determination process of the road intersecting with the sub-block as described above, all intersecting roads intersecting with the sub-block in the region where the detection block is located can be determined. Then, it can also be determined whether the intersecting road is the hot road by comparing the periodic thermal data of the intersecting road with the periodic thermal data of the sub-block, for example, if the periodic thermal data of the intersecting road is much greater than the periodic thermal data of the sub-block, the intersecting road can be considered as the hot road. The sub-block intersecting with the hot road can be determined as the hot road sub-block.

[0130] Considering that the periodic thermal data on the hot road intersecting with the hot road sub-block has certain influence on the periodic thermal data of the detection block, therefore, before the periodic thermal data of the detection block is counted, the hot road sub-block can be removed from the detection block, and then the periodic thermal data of the remaining sub-blocks is used to count the periodic thermal data of the detection block. In this way, the influence of the periodic thermal data on the hot road on the periodic thermal data in the detection block can be removed, and the accuracy of the periodic thermal data of the detection block is further improved.

[0131] In an optional implementation of the embodiment, the method further includes the following steps:

[0132] determining the roads in the detection block;

[0133] determining the periodic thermal data of the roads intersecting with the sub-block of the detection block in a second preset time period;

[0134] when the difference between the periodic thermal data of the sub-block in the second preset time period and the periodic thermal data of the road intersecting with the sub-block in the second preset time period meets a third set condition, the sub-block is determined as the hot road sub-block;

[0135] After the hot road sub-blocks in the detection block are culled, the periodic heat data of the remaining sub-blocks is counted;

[0136] Based on the periodic heat data of the remaining sub-blocks of the detection block, the periodic heat data of the detection block is determined.

[0137] In the optional implementation, all roads in the detection block can be captured by using the point coordinates of the detection block in the electronic map. The second preset time period can include multiple time periods, and in some embodiments, the second preset time period is shorter than the first preset time period.

[0138] The hot road sub-block can be understood as a sub-block intersecting with a hot road, and the hot road can be understood as a road with periodic heat data greater than that of the surrounding roads.

[0139] The periodic heat data of a road is the traffic volume aggregated in a time period. In road network data, a road can be understood as a road segment without branches, and vehicles can only travel from the beginning to the end of the road segment without leaving from the middle, so the traffic volume of a road is a unique value. As described above, the traffic volume of a road can also be calculated from the number of devices corresponding to the received GPS signals.

[0140] The periodic heat data of a road in the second preset time period can include the periodic heat data in multiple time periods in the second preset time period. The periodic heat data of a road in a time period can be obtained by counting the heat data of the road in each sub-period in the time period. The sub-period can be, for example, one day (i.e., 24 hours from 0 o'clock), and the heat data of the road in each day can be determined by counting the number of devices passing through the road, and the number of devices can be determined by the received GPS signals. In order to avoid the case that a single device stays on the road for a long time, the GPS signals in a GPS track corresponding to the same device in a day are counted only once.

[0141] The counting method of the periodic heat data of a sub-block has been described above and will not be repeated here. The periodic heat data of a sub-block in the second preset time period can include the periodic heat data corresponding to each time period in the second preset time period.

[0142] By comparing the periodic heat data of the sub-block and the roads intersecting with the sub-block in the second preset time period, it is determined whether the intersecting road is a hot road. In some embodiments, if the sum of the periodic heat data of all roads intersecting with the sub-block in the second preset time period is much greater than the sum of the periodic heat data of the sub-block in the second preset time period, the sub-block can be considered as a hot road sub-block.

[0143] In some embodiments, the third set condition can be set as that the sum of the periodic thermal data of the sub-block in the second preset time period is less than N times of the sum of the periodic thermal data of all roads intersecting the sub-block in the second preset time period, N≤2, that is, when the sum of the periodic thermal data of the sub-block in the second preset time period is less than N times of the sum of the periodic thermal data of all roads intersecting the sub-block in the second preset time period, the sub-block is determined as a hotspot road sub-block.

[0144] After the hotspot road sub-blocks satisfying the third set condition are removed from the sub-blocks included in the detection block, the periodic thermal data of the detection block can be obtained by statistical analysis of the periodic thermal data of the remaining sub-blocks. The periodic thermal data of the sub-blocks can be obtained by statistical analysis in the manner described above, which will not be described here again.

[0145] In an optional implementation of the present embodiment, the step S102, i.e., the step of determining the detection block in which the periodic thermal data in the first preset time period abnormally fluctuates as an event burst detection block, further includes the following steps:

[0146] For each detection block, the periodic thermal data in the first preset time period of the detection block is clustered, and the one of the two classification sets obtained by clustering in which the periodic thermal data is more is determined as an abnormal classification set;

[0147] An abnormal period statistical value is determined according to the abnormal classification set; wherein the abnormal period statistical value is the statistical number of the periodic thermal data appearing in the abnormal classification set and being continuous in time period;

[0148] According to the number of the abnormal period statistical values contained in the abnormal period count sequence and the maximum value of the abnormal period statistical values, it is determined whether the abnormal period count sequence is an unstable sequence, the abnormal period count sequence containing the abnormal period statistical values sorted by time;

[0149] The detection block corresponding to the unstable sequence is determined as the event burst detection block.

[0150] In the optional implementation, for each detection block in the to-be-detected region, it is judged whether the detection block has occurred a burst event in the first preset time period, and if the detection block has occurred a burst event, the detection block is considered as an event burst detection block in the first preset time period. It is considered that the great change of the periodic thermal data of the detection block in the first preset time period is caused by the burst event, rather than the expired POI in the detection block. Therefore, in order to exclude the event burst detection block from the target detection block, it is necessary to judge whether each detection block in the to-be-detected region is an event burst detection block.

[0151] In some embodiments, the emergency event can be divided into a short-period emergency event and a long-period emergency event. The short-period emergency event can be understood as a period in which the periodic thermal data is abnormal, and the time period is short, while the long-period emergency event can be understood as a period in which the periodic thermal data is abnormal, and the time period is long. Therefore, in the embodiments of the present disclosure, the periodic thermal data corresponding to all time periods in the first preset time period of the detection block is first divided into two clusters, and two classification sets are obtained. It can be understood that, as described above, under the normal state, the periodic thermal data corresponding to each time period of the same detection block in the first preset time period will basically remain stable and will not change greatly; and under the abnormal state, the periodic thermal data will change greatly. Through the above clustering, a small part of the periodic thermal data that is inconsistent with most of the periodic thermal data is clustered into a class, and most of the periodic thermal data is clustered into another class. Considering that the duration of the abnormal state will be less than the duration of the normal state, the set with a smaller number after clustering can be determined as an abnormal classification set, and the periodic thermal data of the short-period emergency event and the long-period emergency event can be determined by further detecting the periodic thermal data in the abnormal classification set.

[0152] In the process of detecting the short-period emergency event and the long-period emergency event, the number of the periodic thermal data that is continuous in time in the abnormal classification set can be counted, and the number is arranged in time sequence to form an abnormal period count sequence. The abnormal period count sequence includes a plurality of abnormal period statistical values, and each abnormal period statistical value represents the number of the periodic thermal data that is classified into the abnormal classification set and is continuous in time. For example, after the periodic thermal data of detection block A in the past 36 months is divided into two clusters, a first classification set and a second classification set are obtained. The first classification set includes 30 periodic thermal data, and the second classification set includes 6 periodic thermal data. Therefore, the second classification set can be determined as the abnormal classification set. Assuming that the time periods corresponding to the periodic thermal data included in the abnormal classification set are the past 2-5 months and the past 34-35 months, respectively, the abnormal period count sequence corresponding to the abnormal classification set can be represented as {4, 2}, that is, the abnormal period count sequence includes an abnormal period statistical value 4 corresponding to the 2-5 months and an abnormal period statistical value 2 corresponding to the 34-35 months.

[0153] In some embodiments, whether the abnormal period count sequence corresponds to an unstable sequence can be determined by the number of the abnormal period statistical values in the abnormal period count sequence and the maximum value of the abnormal period statistical values. If it is an unstable sequence, it can be understood that an emergency event occurs in the detection block in the first preset time period. Therefore, the detection block corresponding to the unstable sequence can be determined as an event emergency detection block.

[0154] In an optional implementation of the embodiment, the step of determining whether the abnormal period count sequence is an unstable sequence according to the number of abnormal period statistical values contained in the abnormal period count sequence and the maximum value of the abnormal period statistical values further comprises the following steps:

[0155] The abnormal period count sequence is determined as a short-period unstable sequence when the number of abnormal period statistical values is less than a first threshold value and the maximum value of the abnormal period statistical values is less than a second threshold value.

[0156] The abnormal period count sequence is determined as a long-period unstable sequence when the number of abnormal period statistical values is equal to a third threshold value and the maximum value of the abnormal period statistical values is greater than or equal to a fourth threshold value and less than a fifth threshold value.

[0157] In the optional implementation, the event burst detection block can be determined by detecting short-period burst events and long-period burst events. In some embodiments, the first threshold value, the second threshold value, the third threshold value, the fourth threshold value and the fifth threshold value can be set according to experience, and the detection block satisfying the threshold condition is determined as the event burst detection block.

[0158] In some embodiments, the detection condition of the short-period burst event can be set as follows: according to the abnormal period count sequence of the periodic thermal data of all time periods in a first preset time period, the abnormal period count sequence in which the number of abnormal period statistical values is less than a first threshold value and the maximum value of the abnormal period statistical values is less than a second threshold value is determined as a short-period unstable sequence, that is, the detection block has occurred a short-period burst event.

[0159] In other embodiments, the detection condition of the long-period burst event can be set as follows: according to the abnormal period count sequence of the periodic thermal data of all time periods in a first preset time period, the abnormal period count sequence in which the number of abnormal period statistical values is equal to a third threshold value and the maximum value of the abnormal period statistical values is greater than or equal to a fourth threshold value and less than a fifth threshold value is determined as a long-period unstable sequence, that is, the detection block has occurred a long-period burst event.

[0160] The detection block that neither satisfies the short-period burst event detection condition nor satisfies the long-period burst event detection condition can be considered as a detection block that has not occurred a burst event.

[0161] The following illustrates a detection method of a burst event in the embodiment of the disclosure.

[0162] The priori assumption is that the duration of the burst event is much smaller than the time in the normal state, and whether the detection block is an event burst detection block is determined in the following way:

[0163] The periodic thermal data of the detection block in the past N months (N is a hyperparameter and is set to N> 36, N corresponds to the first preset time period mentioned above) is used, and the Kmeans clustering algorithm is used to perform two-part clustering on the periodic thermal data of the N months. The classification set with a smaller number of classifications is marked as an abnormal classification set, and the other classification set with a larger number of classifications is marked as a normal classification set.

[0164] Suppose that the number of consecutive normal months in the time series is m, and the number of consecutive abnormal months is n. The original time series is sorted into a periodic count sequence c seq ={…,m i ,n i ,m i+1 ,…},an abnormal periodic count sequence n seq ={n1,n2,…},and a normal periodic count sequence m seq ={m1,m2,…}。

[0165] In the abnormal classification set, the following two types of event patterns are identified:

[0166] Short cycle event: len(n seq )<b 1 , max(n seq )<b 2 ;

[0167] Long cycle event: len(n seq )=b 3 , b 4 <max(n seq )<b 5 ;

[0168] Where b 1 ~ b 5 are the first threshold to the fifth threshold mentioned above, which are hyperparameters. It can be understood that the above hyperparameters can be obtained by experimental parameter tuning in actual application process, and the values of the above hyperparameters can be different in different application scenarios. For example, the above hyperparameters can be as follows: b 1 =5, b 2 =3, b 3 =3, b 4 =2, b 5 =12.

[0169] The event sequence of the short cycle event and the long cycle event is marked as an unstable sequence.

[0170] In an optional implementation of the embodiment, step S103, i.e., the step of determining the thermal prediction data of the detection block in the current time period, further includes the following steps:

[0171] The time series decomposition method and the periodic thermal data of the detection block in the first time period are used to predict the thermal prediction data of the detection block in the current time period.

[0172] In this optional implementation, the time series decomposition method uses an additive model or a multiplicative model to split the original sequence into four parts: long-term trend T, seasonal variation S (explicit period, periodic fluctuation with fixed amplitude and length), cycle variation C (implicit period, fluctuation with no strict rules on cycle length), and irregular variation L. The time series Y can be expressed as a function of the above four parts, i.e., Y = F(T, S, C, L). F(·) can be expressed as an additive model or a multiplicative model. The additive model is Y = T + S + C + L, and the multiplicative model is Y = T × S × C × L.

[0173] The time series decomposition method includes the following steps:

[0174] 1) The moving average method is used to eliminate long-term trends and periodic changes to obtain a sequence TC. Then the monthly (seasonal) average method is used to obtain a seasonal index S.

[0175] 2) A scatter plot is made, and a suitable curve model is used to fit the long-term trend of the sequence to obtain a long-term trend T.

[0176] 3) The cycle factor C is calculated. The cycle variation factor C can be obtained by dividing the sequence TC by T.

[0177] 4) After the time series T, S, and C are decomposed, the remaining is the irregular variation, i.e., I = Y / (TSC).

[0178] When the time series decomposition method is applied to the embodiments of the present disclosure, the time series decomposition method can be used to predict the periodic thermal data of the current time period t = N + 1 according to the periodic thermal data of each time period (e.g., N time periods) in the past first preset time. That is, the thermal prediction data of the current time period. Assuming that the selected time series decomposition method model is a multiplicative model: Y = T × S × C × I, Y represents the thermal prediction data of the time period t, it is assumed that the first preset time period is 36 months, and the current time period is the current month. The long-term trend is assumed to be a linear model T = ax + b, the long-term trend sequence is calculated using a 12-month window moving average method, and the sequence parameters a and b are solved using the least squares method; the seasonal index S is calculated using the moving average trend elimination method; and the cycle variation component C is calculated using the residual method. t t t t t t t t t ​​​​​​​​Since the time series decomposition method is prior art, specific details are not repeated here. The time series model fitted based on the periodic heat data of the past 36 months can directly solve the heat prediction value Y of the current month N+1 .

[0179] In an optional implementation of the embodiment, the step S104, i.e., comparing the heat prediction data with the actual periodic heat data of the current time period, determining the detection block satisfying the first set condition as a candidate detection block, further includes the following steps:

[0180] When the weighted value obtained by multiplying the heat prediction data by a preset weight is greater than the actual periodic heat data, the detection block is determined as a candidate detection block; wherein the preset weight is less than 1.

[0181] In the optional implementation, as described above, the periodic heat data of the same detection block in each time period within the first preset time period will not change greatly in the normal state, so in the normal state, the difference between the periodic heat data of the current time period predicted by the time series decomposition method based on the periodic heat data of each time period within the first preset time period and the actual periodic heat data of the current time period will not be too large. Experiments show that when the weighted value obtained by multiplying the heat prediction data by a preset weight is greater than the actual periodic heat data, the detection block may have an expired POI region within the first preset time period, so the detection block is determined as a candidate detection block. The preset weight is a hyperparameter, and the preset weight is less than 1. In this way, one or more candidate detection blocks in the to-be-detected region can be determined. Due to the existence of the event, there may be an event burst detection block in the one or more candidate detection blocks, so after the event burst detection block in the candidate detection block is put forward, the remaining candidate detection block is the target detection block, i.e., there is a target region of the expired POI.

[0182] Figure 2 An implementation schematic diagram of a process of mining a target region of an expired POI in a to-be-detected region according to an embodiment of the present disclosure is shown. As shown in Figure 2 , the implementation process of mining the expired POI in the to-be-detected region includes the following steps:

[0183] (I) Grid division

[0184] For the to-be-detected region in the electronic map, grid division can be performed first, for example, the map longitude and latitude coordinates are multiplied by 4000 and then divided by 3 to obtain the grid of the map data corresponding to the to-be-detected region, and each grid is referred to as a detection block. Then each detection block is subdivided into 3x3 sub-blocks, each with a side length of about 25 meters, to facilitate the removal of road influence.

[0185] After the grid division is completed, the heat data of the detection block can be summarized by day, and the summarized heat data is added by month to obtain the periodic heat data of the detection block, referred to as detection block heat. In this summary method, the same device can be counted only once in the same day, and can be counted multiple times in different days.

[0186] (II) High heat road removal and periodic heat data statistics

[0187] After the grid division is completed, the heat data of the detection block can be summarized by day, and the summarized heat data is added by month to obtain the periodic heat data of the detection block, referred to as detection block heat. In this summary method, the same device can be counted only once in the same day, and can be counted multiple times in different days.

[0188] In order to remove the sub-blocks of the high heat road from the detection block, the periodic heat data of the remaining sub-blocks of the detection block can be added to obtain the periodic heat data representing the detection block, and the specific method is as follows:

[0189] 1) It is known that the road curve p is composed of several line segments It is known that the coordinates of the points in the detection block; for each detection block, all roads within the coverage of the detection block are grabbed through the coordinates of the points in the detection block.

[0190] 2) For any sub-block j in detection block i, 1≤j≤9, the coordinates of the four vertices of the sub-block are taken, and two diagonal line segments are formed

[0191] 3) Determine whether the endpoints of the line segment are within the range of the sub-block j, if so, the road p intersects the sub-block j, otherwise the road p does not intersect the sub-block j; determine whether and have intersection points, if there are intersection points, the road p intersects the sub-block j, otherwise the road p does not intersect the sub-block j.

[0192] 4) If the sub-block contains at least one road, and the sum of the periodic heat data of the sub-block in the last M months (M is a hyperparameter and can be set as M≥6) is less than Q times the sum of the periodic heat data of all intersecting roads in the last M months (Q is a hyperparameter and can be set as Q≤2), the sub-block is regarded as a hot road sub-block and is removed from the detection block.

[0193] 5) Statistics are summarized using the remaining sub-blocks to obtain the periodic thermal data of the detection block.

[0194] (III) Event detection

[0195] The prior assumption is that the duration of the event within the detection block is much smaller than the time in normal state. The specific method of event detection is as follows:

[0196] 1) For the periodic thermal data of the past N months (N is a hyperparameter and can be set as N≥36), use Kmeans clustering algorithm for binary clustering, and mark the small number of categories as abnormal category set, and mark the large number of categories as normal category set.

[0197] 2) Assuming that the number of consecutive normal months in the time series is m and the number of consecutive abnormal months is n, the original time series is sorted into periodic count sequence c seq ={…,m i ,n i ,m i+1 ,…}, abnormal periodic count sequence n seq ={n1,n2,…}, normal periodic count sequence m seq ={m1,m2,…}.

[0198] 3) In the abnormal category set, the following two types of event patterns are identified:

[0199] Short-period event: len(n seq )<b 1 , max(n seq )<b 2 ;

[0200] Long-period event: len(n seq )=b 3 , b 4 <max(n seq )<b 5 ;

[0201] Where b 1 ~b 5 are hyperparameters and can be set as b 1 =5, b 2 =3, b 3 =3, b 4 =2, b 5 =12.

[0202] The abnormal periodic count sequence of short-period event and long-period event is marked as unstable sequence. The event burst detection block set corresponding to the unstable sequence is denoted as Set event .

[0203] (iv) Thermal decay

[0204] The specific calculation process of thermal decay includes:

[0205] 1) Obtain the periodic thermal data of the current month as the true value h.

[0206] 2) Use the time series decomposition method to predict the thermal prediction value Y of the current month t = N + 1 according to the thermal data of the past N months (N can be set to 36). N+1 .

[0207] 3) If h < wY N+1 , w is a hyperparameter and can be set to w = 0.6, then mark the detection block as a candidate detection block. The set of marked candidate detection blocks is denoted as Set pred .

[0208] (v) Output logic

[0209] By calculating the difference between the candidate detection block set and the event burst detection block set, the final output result Set prod = Set pred - Set event , that is, the output result includes the target area where the detection block with expired POI exists.

[0210] Figure 3 An application scenario diagram for mining expired POIs in an electronic map according to an embodiment of the present disclosure is shown. As Figure 3 shown, for a to-be-detected area in an electronic map, in the process of judging whether the annotated POI is expired, the server 301 obtains the GPS signals returned by each device 303 in the to-be-detected area in the past 36 months from the return GPS signal receiving device 302, which can include but is not limited to the receiving device corresponding to each application installed on the device, such as the receiving device corresponding to the electronic map application, the receiving device corresponding to the mobile operator, etc. The device 303 in the to-be-detected area will periodically return its GPS signal to the corresponding return GPS signal receiving device 302 according to the use of the application on the device 303. The server collects the GPS signals returned by each device 303 in the to-be-detected area in the past 36 months from each return GPS signal receiving device 302, and statistically counts the number of devices in the to-be-detected area based on the GPS signals in units of days, with only one device counted in a day; then the number of devices counted each day is summed up to obtain the periodic thermal data of each detection block in the to-be-detected area in units of months.

[0211] The server 301 predicts the heat forecast data of the current month based on the periodic heat data of each detection block in the past 36 months, and screens the candidate detection blocks in which the expired POIs are likely to exist from each detection block by comparing with the actual periodic heat data of the current month. The server 301 also screens the event burst detection blocks in which the burst events occur from each detection block of the to-be-detected area based on the periodic heat data of each detection block in the past 6 months, and determines the remaining candidate detection blocks as the target area in which the expired POIs exist after excluding the event burst detection blocks from the candidate detection blocks.

[0212] The following is an apparatus embodiment of the present disclosure, which can be used to perform the method embodiments of the present disclosure.

[0213] Figure 4 A schematic diagram of a framework of a target area determination apparatus according to an embodiment of the present disclosure is shown. As shown in the figure, the apparatus can be realized as part or all of an electronic device by software, hardware or a combination of both. Figure 4

[0214] The target area determination apparatus includes:

[0215] A first acquisition module 401 configured to acquire periodic heat data of a detection block constituting a to-be-detected area;

[0216] A first determination module 402 configured to determine the detection block in which the periodic heat data abnormally fluctuates in a first preset time period as an event burst detection block;

[0217] A second determination module 403 configured to determine heat forecast data of the detection block in a current time period; wherein the heat forecast data is predicted based on the periodic heat data of a period before the current time period;

[0218] A third determination module 404 configured to compare the heat forecast data with actual periodic heat data of the current time period, and determine the detection block in which the comparison result meets a first set condition as a candidate detection block;

[0219] A fourth determination module 405 configured to determine the area constituted by the remaining candidate detection blocks after excluding the event burst detection block from the candidate detection blocks as an area in which an expired POI exists.

[0220] In an optional implementation of the present embodiment, the apparatus further includes:

[0221] A division module configured to divide the detection block into a plurality of sub-blocks;

[0222] A first statistical module configured to statistically acquire periodic heat data of the sub-blocks; ​

[0223] a fifth determining module, configured to determine periodic thermal data of a detection block containing the sub-block according to the periodic thermal data of the sub-block.

[0224] In an optional implementation of the embodiment, the apparatus further includes:

[0225] a second obtaining module, configured to obtain a hotspot road;

[0226] a sixth determining module, configured to determine a sub-block intersecting with the hotspot road as a candidate sub-block;

[0227] a deleting module, configured to delete the candidate sub-block whose periodic thermal data satisfies a second set condition from the sub-blocks constituting the detection block;

[0228] a first counting module, configured to count the periodic thermal data of the remaining sub-blocks in the detection block;

[0229] a seventh determining module, configured to determine the periodic thermal data of the detection block based on the periodic thermal data of the remaining sub-blocks in the detection block.

[0230] In an optional implementation of the embodiment, the sixth determining module includes:

[0231] a first determining submodule, configured to determine the hotspot road within the coverage of the detection block according to the point coordinates in the detection block;

[0232] a judging submodule, configured to judge whether the end points of road segments constituting the hotspot road are located in the sub-blocks of the detection block;

[0233] a second determining submodule, configured to determine diagonal segments of the sub-blocks of the detection block, and judge whether the road segments constituting the hotspot road intersect with the diagonal segments;

[0234] a third determining submodule, configured to determine the sub-blocks in which the end points of the road segments are located, or the sub-blocks in which the road segments intersect with the diagonal segments as the candidate sub-blocks.

[0235] In an optional implementation of the embodiment, the apparatus further includes:

[0236] an eighth determining module, configured to determine a road within the detection block;

[0237] a ninth determining module, configured to determine periodic thermal data of the road intersecting with the sub-block of the detection block within a second preset time period;

[0238] a tenth determining module, configured to determine the sub-block as a hotspot road sub-block when a difference between periodic thermal data of the sub-block in the second preset time period and periodic thermal data of the road intersecting with the sub-block in the second preset time period meets a third set condition;

[0239] a second counting module, configured to count the periodic thermal data of the remaining sub-blocks after excluding the hotspot road sub-blocks in the detection block;

[0240] an eleventh determining module, configured to determine the periodic thermal data of the detection block based on the periodic thermal data of the remaining sub-blocks in the detection block.

[0241] In an optional implementation of the embodiment, the first determining module comprises:

[0242] a clustering submodule, configured to, for each detection block, cluster the periodic thermal data in a first preset time period of the detection block, and determine one of two classification sets obtained by clustering, in which the periodic thermal data is more, as an abnormal classification set;

[0243] a third determining submodule, configured to determine an abnormal period statistical value according to the abnormal classification set; wherein the abnormal period statistical value is a statistical number of periodic thermal data that appears in the abnormal classification set and is continuous in time period;

[0244] a fourth determining submodule, configured to determine whether an abnormal period count sequence is an unstable sequence according to a number of abnormal period statistical values contained in the abnormal period count sequence and a maximum value of the abnormal period statistical values, the abnormal period count sequence containing abnormal period statistical values sorted by time;

[0245] a fifth determining submodule, configured to determine the detection block corresponding to the unstable sequence as the event burst detection block.

[0246] In an optional implementation of the embodiment, the fourth determining submodule comprises:

[0247] a sixth determining submodule, configured to determine the abnormal period count sequence in which the number of abnormal period statistical values is less than a first threshold value and the maximum value of the abnormal period statistical values is less than a second threshold value as a short-period unstable sequence;

[0248] a seventh determining submodule, configured to determine the abnormal period count sequence in which the number of abnormal period statistical values is equal to a third threshold value and the maximum value of the abnormal period statistical values is greater than or equal to a fourth threshold value and less than a fifth threshold value as a long-period unstable sequence.

[0249] In an optional implementation of the embodiment, the second determining module comprises:

[0250] The prediction submodule is configured to predict the heat forecast data of the detection block in a current time period by using a time series decomposition method and the periodic heat data of the detection block in the first time period.

[0251] In an optional implementation of the embodiment, the third determining module comprises:

[0252] The eighth determining submodule is configured to determine the detection block as a candidate detection block when a weighted value obtained by multiplying the heat forecast data by a preset weight is greater than the actual periodic heat data, wherein the preset weight is less than 1.

[0253] The determination apparatus of the expired interest point in the embodiment corresponds to the determination method of the expired interest point in the foregoing, and specific details can be referred to the description of the determination method of the expired interest point, which will not be repeated here.

[0254] Figure 5 is a structural schematic diagram of an electronic device suitable for implementing the determination method of the expired interest point according to the embodiments of the present disclosure.

[0255] As shown in Figure 5 The electronic device 500 includes a processing unit 501, which can be implemented as a CPU, a GPU, an FPGA, an NPU, or the like. The processing unit 501 can perform various processes in the embodiments of any of the methods of the present disclosure according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage portion 508 into a random access memory (RAM) 503. Various programs and data required for the operation of the electronic device 500 are also stored in the RAM 503. The processing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0256] The following components are connected to the I / O interface 505: an input portion 506 including a keyboard, a mouse, and the like; an output portion 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage portion 508 including a hard disk, and the like; and a communication portion 509 including a network interface card such as a LAN card, a modem, and the like. The communication portion 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as necessary. A removable medium 511 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 510 as necessary, so that a computer program read therefrom is installed in the storage portion 508 as necessary.

[0257] In particular, according to embodiments of the present disclosure, the above-mentioned methods with reference to any of the embodiments of the present disclosure can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program tangibly embodied on a non-transitory computer readable medium, the computer program containing program code for executing any of the methods described in embodiments of the present disclosure. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 509, and / or installed from the removable medium 511.

[0258] The flow and block diagrams in the drawings show the architectural, functional and operational views of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow and block diagrams can represent a module, a segment, or a portion of code which comprises one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may

[0259] The units or modules described in the embodiments of the present disclosure can be implemented by software, or by hardware. The described units or modules can also be provided in a processor, and the names of the units or modules do not constitute a limitation on the units or modules themselves in some cases.

[0260] As another aspect, the present disclosure also provides a computer readable storage medium, which can be the computer readable storage medium included in the apparatus described in the above embodiments, or can exist separately from the apparatus and not be assembled into the apparatus. The computer readable storage medium stores one or more programs for execution by one or more processors to perform the methods described in the present disclosure.

[0261] The above description is merely that of the preferred embodiments of the present disclosure and a description of the technical principles of the present disclosure. It should be understood by those skilled in the art that the inventive scope involved in the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by the combinations of the above technical features or equivalent features without departing from the inventive concept. For example, the technical solutions formed by the mutual replacement of the above features and the technical features with similar functions disclosed in the present disclosure (but not limited to) without departing from the inventive concept.

Claims

1. A method of determining expired points of interest, wherein, The method comprises: acquiring periodic thermal data of a detection block constituting a region to be detected; determining the detection block in which the periodic thermal data abnormally fluctuates within a first preset time period as an event burst detection block, the first preset time period comprising a plurality of time periods in the past; determining thermal prediction data of the detection block in a current time period, wherein the thermal prediction data is predicted based on the periodic thermal data of a period before the current time period; comparing the thermal prediction data with actual periodic thermal data of the current time period, and determining the detection block in which the comparison result satisfies a first set condition as a candidate detection block; after eliminating the event burst detection block from the candidate detection blocks, determining a region constituted by the remaining candidate detection blocks as a region in which an expired point of interest exists.

2. The method of claim 1, wherein, The method further comprises: dividing the detection block into a plurality of sub-blocks; counting periodic thermal data of the sub-blocks; determining periodic thermal data of the detection block containing the sub-blocks according to the periodic thermal data of the sub-blocks.

3. The method of claim 1, wherein, The method further comprises: acquiring a hot road; determining a sub-block intersecting with the hot road as a candidate sub-block; eliminating the candidate sub-block in which the periodic thermal data satisfies a second set condition from the sub-blocks constituting the detection block; counting periodic thermal data of the remaining sub-blocks in the detection block; determining the periodic thermal data of the detection block based on the periodic thermal data of the remaining sub-blocks in the detection block.

4. The method of claim 3, wherein, Determining a sub-block intersecting with the hot road as a candidate sub-block comprises: determining a hot road within the coverage of the detection block according to the point coordinates in the detection block; judging whether the end points of road segments constituting the hot road are located in the sub-blocks of the detection block; determining a diagonal segment of the sub-blocks of the detection block, and judging whether the road segments constituting the hot road intersect with the diagonal segment; determining the sub-block in which the end points of the road segments are located, or the sub-block in which the road segments intersect with the diagonal segment, as the candidate sub-block.

5. The method of claim 1, wherein, The method further comprises: determining a road within the detection block; determining periodic thermal data of the road intersecting with the sub-blocks of the detection block within a second preset time period; determining the sub-block as a hot road sub-block when the difference between the periodic thermal data of the sub-block within the second preset time period and the periodic thermal data of the road intersecting with the sub-block within the second preset time period satisfies a third set condition; counting periodic thermal data of the remaining sub-blocks after eliminating the hot road sub-blocks within the detection block; determining the periodic thermal data of the detection block based on the periodic thermal data of the remaining sub-blocks.

6. The method according to any one of claims 1 to 5, wherein, Determining the detection block in which the periodic thermal data abnormally fluctuates within a first preset time period as an event burst detection block comprises: for each detection block, clustering the periodic thermal data of the detection block within the first preset time period, and determining the one of two classification sets in which the periodic thermal data is more as an abnormal classification set; determining an abnormal period statistical value according to the abnormal classification set; wherein the abnormal period statistical value is a statistical number of periodic thermal data appearing in the abnormal classification set and continuously in time periods; determining whether the abnormal period count sequence is an unstable sequence according to a number of abnormal period statistical values contained in the abnormal period count sequence and a maximum value of the abnormal period statistical values, the abnormal period count sequence containing abnormal period statistical values sorted by time; determining the detection block corresponding to the unstable sequence as the event burst detection block.

7. The method of claim 6, wherein, determining whether the abnormal period count sequence is an unstable sequence according to a number of abnormal period statistical values contained in the abnormal period count sequence and a maximum value of the abnormal period statistical values, the abnormal period count sequence containing abnormal period statistical values sorted by time; determining the abnormal period count sequence as a short period unstable sequence when the number of abnormal period statistical values is less than a first threshold value and the maximum value of the abnormal period statistical values is less than a second threshold value; determining the abnormal period count sequence as a long period unstable sequence when the number of abnormal period statistical values is equal to a third threshold value and the maximum value of the abnormal period statistical values is greater than or equal to a fourth threshold value and less than a fifth threshold value; wherein the first threshold value, the second threshold value, the third threshold value, the fourth threshold value and the fifth threshold value are hyperparameters.

8. The method according to any one of claims 1-5, 7, wherein, determining thermal prediction data of the detection block in a current time period, comprising: predicting the thermal prediction data of the detection block in the current time period by using a time series decomposition method and the periodic thermal data of the detection block in the first preset time period.

9. The method according to any one of claims 1-5, 7, wherein, comparing the thermal prediction data with actual periodic thermal data in the current time period, and determining the detection block as a candidate detection block when a comparison result satisfies a first set condition, comprising: determining the detection block as a candidate detection block when a weighted value obtained by multiplying the thermal prediction data by a preset weight is greater than the actual periodic thermal data; wherein the preset weight is less than 1.

10. A computer program product comprising computer instructions, wherein, The computer instructions are executed by the processor to implement the method of any one of claims 1-9. The computer instructions are executed by the processor to implement the method of any one of claims 1-9.

Images