Position verification method and device, electronic device, and storage medium

Abstract

One or more embodiments of the present specification provide a location verification method and device, an electronic device, and a storage medium; the method may include: determining a location to be verified and several reference locations corresponding to the location to be verified, and obtaining the declared locations of the location to be verified and the reference locations; determining an estimated travel distance between the location to be verified and each reference location based on travel-related information between the location to be verified and each reference location, the travel-related information including travel time information between the location to be verified and each reference location; locating the location to be verified based on the declared locations of each reference location and the estimated travel distance obtained, to obtain an estimated location of the location to be verified, and matching the estimated location with the declared location of the location to be verified, so as to determine whether the declared location of the location to be verified has passed the verification based on the matching result.

Description

Technical Field

[0001] One or more embodiments of the present specification relate to the field of positioning technology, and in particular, to a position verification method and device, an electronic device, and a storage medium. Background Art

[0002] In related technologies, the physical location of a place can be uploaded to a specific platform to complete registration. For example, in the application scenario of instant delivery, for the delivery platform, the accurate physical location of the merchant is crucial for order scheduling. Reasonable order scheduling can ensure that the delivery order is delivered before the promised time to avoid timeout.

[0003] Therefore, accurate registration of the physical location is an important link. However, in the registration process, the registered physical location is often deviated due to human factors. For example, the merchant location is registered by the merchant himself, and some wrong merchant locations may be due to human operation errors or merchant location updates. But some errors are caused by merchant location fraud. For example, a merchant with a registered location in a large shopping mall often means better service and higher food quality than in a residential area. Therefore, some merchants may deliberately register the wrong location (such as in a shopping mall) to attract more online consumers who do not have to go in person and only need riders to deliver, which is the problem of location fraud. Summary of the invention

[0004] In view of this, one or more embodiments of the present specification provide a location verification method and device, an electronic device, and a storage medium.

[0005] To achieve the above objectives, one or more embodiments of this specification provide the following technical solutions:

[0006] According to a first aspect of one or more embodiments of this specification, a location verification method is proposed, including:

[0007] Determine a location to be verified and a plurality of reference locations corresponding to the location to be verified, and obtain declared positions of the location to be verified and the reference locations;

[0008] Determine the estimated travel distance between the location to be verified and each reference location according to the travel related information between the location to be verified and each reference location, wherein the travel related information includes the travel time information between the location to be verified and each reference location;

[0009] The place to be verified is located according to the declared position of each reference place and the estimated travel distance obtained in advance to obtain the estimated position of the place to be verified, and the estimated position is matched with the declared position of the place to be verified to determine whether the declared position of the place to be verified has passed the verification based on the matching result.

[0010] Optionally, determining the location to be verified and a plurality of reference locations corresponding to the location to be verified includes:

[0011] Get a location set containing multiple locations within a preset range;

[0012] Each location in the location set is traversed to sequentially use the current location as the location to be verified, and use other locations different from the current location as the reference locations.

[0013] Optionally, determining the estimated travel distance between the location to be verified and each reference location according to the travel related information between the location to be verified and each reference location respectively includes:

[0014] Inputting the trip related information into a distance model, wherein the distance model is obtained by training the historical trip related information of a plurality of historical trips using a supervised learning algorithm, wherein the historical trip related information is annotated with the standard trip distance of the historical trip;

[0015] The estimated travel distance between the location to be verified and each reference location is determined according to the output result of the distance model.

[0016] Optionally, when the location to be verified and the reference location are physical stores, the method further includes:

[0017] Obtain the departure time information of any delivery order undertaken by the delivery party and the arrival time information of the next delivery order;

[0018] The travel time information between the first physical store corresponding to any one of the delivery orders and the second physical store corresponding to the next delivery order is determined according to the store departure time information and the store arrival time information.

[0019] Optionally, locating the location to be verified according to the declared locations of each reference location and the estimated travel distance obtained by estimation to obtain the estimated location of the location to be verified includes:

[0020] Finding a target location, wherein the error between the standard travel distance between the location of the target location and the declared location of each reference location and the estimated travel distance satisfies a preset error condition;

[0021] The position of the target location is used as the estimated position.

[0022] Optionally, the target location is found by:

[0023]

[0024] represents the target location, q represents any location, q * Indicates the location to be verified;

[0025] P = {p i ,i=1...N},p i represents reference location i, and N represents the number of reference locations;

[0026] d s (p i ,q) indicates p i The standard travel distance between the declared position of and the position of q;

[0027] d e (p i ,q * ) indicates p i With q * Estimated travel distance between.

[0028] Optionally, the standard travel distance includes a travel distance between the position of the target location calculated using map data and the declared position of each reference location.

[0029] Optionally, when the location to be verified and the reference location are physical stores, the itinerary-related information between the location to be verified and any physical store further includes at least one of the following:

[0030] The number of delivery orders contained in the area to which the location to be verified belongs, the number of delivery orders contained in the area to which any one of the physical stores belongs, the road condition information between the location to be verified and any one of the physical stores, the current delivery status information of the delivery party traveling from one of the location to be verified and any one of the physical stores to the other, and the delivery preference information of the delivery party.

[0031] According to a second aspect of one or more embodiments of this specification, a position verification device is provided, comprising:

[0032] A determination unit, determining a location to be verified and a plurality of reference locations corresponding to the location to be verified, and obtaining declared positions of the location to be verified and the reference locations;

[0033] An estimation unit, determining an estimated travel distance between the location to be verified and each reference location according to travel related information between the location to be verified and each reference location, wherein the travel related information includes travel time information between the location to be verified and each reference location;

[0034] A verification unit locates the place to be verified according to the declared positions of each reference place and the estimated travel distance obtained by estimation to obtain the estimated position of the place to be verified, and matches the estimated position with the declared position of the place to be verified to determine whether the declared position of the place to be verified has passed the verification according to the matching result.

[0035] Optionally, the determining unit is specifically configured to:

[0036] Get a location set containing multiple locations within a preset range;

[0037] Each location in the location set is traversed to sequentially use the current location as the location to be verified, and use other locations different from the current location as the reference locations.

[0038] Optionally, the estimation unit is specifically used for:

[0039] Inputting the trip related information into a distance model, wherein the distance model is obtained by training the historical trip related information of a plurality of historical trips using a supervised learning algorithm, wherein the historical trip related information is annotated with the standard trip distance of the historical trip;

[0040] The estimated travel distance between the location to be verified and each reference location is determined according to the output result of the distance model.

[0041] Optionally, the estimation unit is further used for:

[0042] Obtain the departure time information of any delivery order undertaken by the delivery party and the arrival time information of the next delivery order;

[0043] The travel time information between the first physical store corresponding to any one of the delivery orders and the second physical store corresponding to the next delivery order is determined according to the store departure time information and the store arrival time information.

[0044] Optionally, the verification unit is specifically used for:

[0045] Finding a target location, wherein the error between the standard travel distance between the location of the target location and the declared location of each reference location and the estimated travel distance satisfies a preset error condition;

[0046] The position of the target location is used as the estimated position.

[0047] Optionally, the target location is found by:

[0048]

[0049] represents the target location, q represents any location, q * Indicates the location to be verified;

[0050] P = {p i ,i=1...N},p i represents reference location i, and N represents the number of reference locations;

[0051] d s (p i ,q) indicates p i The standard travel distance between the declared position of and the position of q;

[0052] d e (p i ,q * ) indicates p i With q * Estimated travel distance between.

[0053] Optionally, the standard travel distance includes a travel distance between the position of the target location calculated using map data and the declared position of each reference location.

[0054] Optionally, when the location to be verified and the reference location are physical stores, the itinerary-related information between the location to be verified and any physical store further includes at least one of the following:

[0055] The number of delivery orders contained in the area to which the location to be verified belongs, the number of delivery orders contained in the area to which any one of the physical stores belongs, the road condition information between the location to be verified and any one of the physical stores, the current delivery status information of the delivery party traveling from one of the location to be verified and any one of the physical stores to the other, and the delivery preference information of the delivery party.

[0056] According to a third aspect of one or more embodiments of this specification, an electronic device is provided, including:

[0057] processor;

[0058] a memory for storing processor-executable instructions;

[0059] The processor implements the position verification method as described in any of the above embodiments by running the executable instructions.

[0060] According to a fourth aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium on which computer instructions are stored. When the instructions are executed by a processor, the steps of the method described in any one of the above embodiments are implemented.

[0061] In the technical solution of this specification, there is a declared position of the place to be verified, that is, the physical position declared for the place to be verified. It should be noted that the physical position is not necessarily accurate and may deviate from the actual physical position. Therefore, several reference places are obtained to verify the declared position of the place to be verified using the declared positions of these reference places.

[0062] Specifically, the travel time between two locations is related to the travel distance between the two locations. Based on the above correlation between the travel time and the travel distance, the travel time can be used to estimate the travel distance between any two locations. Then, the travel distance between the location to be verified and each reference location can be estimated to obtain the estimated travel distance, and the estimated travel distance obtained is used as the actual travel distance between the location to be verified and each reference location (understood as the real travel distance), so that the location to be verified is located based on the declared position and the actual travel distance of each reference location to obtain an estimated position (the estimated position is considered to be the calculated real physical position at this time). It can be understood that if the error between the declared position of the location to be verified and the actual physical position is small, the declared position should match the estimated position. Therefore, whether the declared position passes the verification can be determined based on whether the estimated position matches the declared position. By verifying the physical location in the above manner, it is possible to avoid manual on-site verification at the location, thereby reducing costs while ensuring the accuracy of the verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flowchart of a position verification method provided by an exemplary embodiment.

[0064] Figure 2 It is a schematic diagram of a store location error provided by an exemplary embodiment.

[0065] Figure 3 It is a schematic diagram of combining spatiotemporal data provided by an exemplary embodiment.

[0066] Figure 4 It is a schematic diagram of a map travel distance provided by an exemplary embodiment.

[0067] Figure 5 It is a structural schematic diagram of a device provided by an exemplary embodiment.

[0068] Figure 6 It is a block diagram of a position verification device provided by an exemplary embodiment. DETAILED DESCRIPTION

[0069] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of this specification. Instead, they are merely examples of devices and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0070] It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0071] Taking the application scenario of online instant delivery as an example, online orders (such as food, medicine, etc.) are delivered to users from the merchant's physical stores (such as restaurants, pharmacies, supermarkets, etc.) by delivery personnel (such as riders) within a short time (such as 30 minutes). If the user receives the goods for more than the time limit, the platform will compensate the user. In such a platform, the actual physical location of the physical store is crucial to assigning orders to the most suitable riders, that is, order scheduling. Reasonable order scheduling can ensure that the order is delivered before the promised time to avoid timeouts. In fact, the locations of these stores are registered by the merchants on the delivery platform themselves, so they are not all their accurate physical locations. Some erroneous location records may be due to manual errors or merchant location updates, but some are caused by deliberate manipulation, that is, location fraud.

[0072] The above-mentioned location error problem will seriously affect the business of online instant delivery platforms, such as causing improper order allocation, delivery route deviation, and unnecessary overtime compensation for users. Therefore, delivery platforms need to identify these merchants with wrong store locations and find their real locations.

[0073] In the related art, the delivery platform may require the rider to report the wrong location when delivering, but the rider's reporting method increases the rider's burden and has high labor costs. In response to this, this specification provides a location verification solution that can reduce costs while ensuring the accuracy of verification.

[0074] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of a location verification method provided by an exemplary embodiment. Figure 1As shown, the method may include the following steps:

[0075] Step 102: determine a location to be verified and a plurality of reference locations corresponding to the location to be verified, and obtain declared positions of the location to be verified and the reference locations.

[0076] The declared location of the place to be verified is the physical location declared for the place to be verified. It should be noted that the physical location is not necessarily accurate and may deviate from the actual physical location. In real business scenarios, the physical location registered or registered for a certain location is not necessarily accurate. Taking the online instant delivery scenario as an example, the physical location of the merchant's physical store is manually registered by the merchant (for example, adding a text address, uploading the GPS location of a mobile device, or manually selecting a point on a map), and then converted into GPS coordinates through the system. Therefore, the location of the merchant's physical store can easily deviate due to human operation or human forgery.

[0077] For example, Figure 2 An example of a business with a serious location error is shown. Figure 2 As shown in the figure, the solid dots represent the registered location of the store of merchant A, while the hollow dots represent the actual location of the store of merchant A. When a new delivery order appears for merchant A, it is assumed that the delivery platform has two riders, who are located at the registered locations of merchant B and merchant C respectively. Then, if the location of merchant A is wrong, the delivery platform will mistakenly assign the delivery order to the rider at merchant C, while the rider at merchant B is actually closer to the location of merchant A's store.

[0078] In this regard, the location verification scheme of this manual targets multiple locations and uses the travel distance between two locations to perform mutual cross-validation to verify whether there are errors in the corresponding locations. It should be noted that the distance used in this manual is the travel distance, that is, the distance, not the displacement. For example, if the rider delivers on foot, then the travel distance is the rider's walking distance; if the rider uses an electric bicycle for delivery, then the travel distance is the rider's riding distance. Figure 2 The principle of cross validation is illustrated with examples.

[0079] Use the travel distance from a merchant store to other nearby merchant stores to cross-verify whether the merchant's store location is suspicious. Figure 2As shown in the figure, assume that the registered location of the store of merchant A is wrong, and the registered locations of the stores of merchants B and C are accurate. Then, based on the registered locations of the three merchant stores of A, B, and C and the travel distances between the merchant stores, it can be found that: (1) the actual travel distance between A and B is much shorter than the travel distance calculated based on the registered location of A; (2) the actual travel distance between A and C is much longer than the travel distance calculated based on the registered location of A; (3) the actual travel distance between B and C is similar to the travel distance calculated based on the registered locations of B and C. Therefore, in practical applications, the above cross-validation travel distance method can be used. If the above deviation in travel distance occurs, it can be inferred that the registered location of the store of merchant A is wrong.

[0080] Based on the above cross-validation principle, in order to further improve the accuracy of validation, the distance between mutually validated locations should be avoided to be too far, because long-distance travel may bring large variation noise to the data. In view of the above problem, a smaller range can be set in advance, and the locations that need to be verified included in the range are classified into a location set to perform the above cross-validation steps for each location in the set in turn. In other words, each location in the location set can be traversed, and a cross-validation step is performed once for each location in the location set. In each cross-validation step, the location currently polled is the location to be verified, and other locations are reference locations. Controlling the preset range within a suitable value (such as within 200m, 500m and 800m, etc.) can avoid the above-mentioned problem of large noise, that is, to ensure that the location to be verified and other nearby locations can verify the correctness of the location for each other, thereby helping positioning. Based on the setting of the above preset range, in the process of determining the location to be verified and several reference locations corresponding to the location to be verified, a location set including multiple locations within the preset range can be obtained, and then each location in the location set can be traversed to sequentially use the current location as the location to be verified and other locations different from the current location as reference locations. Of course, the specific value of the preset range can be flexibly set according to actual conditions, and this specification does not limit this.

[0081] Alternatively, in addition to the above-mentioned method of dividing the location set and traversing each location in the set for cross-validation, the declared position of the reference location may not be used, and the actual physical location of the reference location (such as manual on-site survey, or a location that has passed verification) may be used. In this case, it is only necessary to use the actual physical materials of these reference locations to verify the declared position of the location to be verified, without the need to implement the above-mentioned traversal operation.

[0082] Step 104, determining an estimated travel distance between the location to be verified and each reference location based on travel-related information between the location to be verified and each reference location, wherein the travel-related information includes travel time information between the location to be verified and each reference location.

[0083] From the above cross-validation principle, it can be seen that in order to ensure the accuracy of the travel distance comparison in the above cross-validation process, the actual travel distance between the two stores should be sufficiently accurate. At the same time, in order to avoid manual on-site survey of the actual travel distance between the two stores, the method of estimating the actual travel distance (hereinafter referred to as estimated travel distance) can be used to reduce costs. Therefore, the accuracy of the estimated travel distance should be improved as much as possible.

[0084] As an exemplary embodiment, the actual travel distance may be estimated using the correlation between the travel-related information and the travel distance. In one case, the travel-related information includes travel time information. Specifically, the travel time between two locations is correlated with the travel distance between the two locations. For example, when using the same means of transportation, the longer the travel distance between the two locations, the longer the travel time consumed, that is, the travel time is positively correlated with the travel distance.

[0085] Based on the above correlation between travel time and travel distance, the travel time can be used to estimate the travel distance between any two locations. Then, the travel distance between the location to be verified and each reference location can be estimated to obtain an estimated travel distance, and the estimated travel distance obtained is used as the actual travel distance between the location to be verified and each reference location (understood as the real travel distance), so that the declared position and the actual travel distance of each reference location are used as the basis for locating the location to be verified to obtain an estimated position (at this time, the estimated position is considered to be the calculated real physical position). It can be understood that if the error between the declared position of the location to be verified and the actual physical position is small, the declared position should match the estimated position. Therefore, whether the declared position passes the verification can be determined based on whether the estimated position matches the declared position. By verifying the physical location in the above manner, it is possible to avoid manual on-site verification at the location, thereby reducing costs while ensuring the accuracy of the verification.

[0086] Among them, a supervised learning algorithm can be used to learn the correlation between the above travel time and travel distance, so as to train a distance model for estimating the actual travel distance. For the training process of the supervised learning algorithm, the input sample data is called a "training set". The sample data in the training set has a clear identification or result (i.e., a sample label). When using the supervised learning algorithm to establish a distance model, the supervised learning algorithm establishes a learning process, compares the predicted results with the actual results of the "training set", and continuously adjusts the prediction model until the prediction results of the model reach an expected accuracy. For example, common supervised learning algorithms include logistic regression, neural networks, decision trees, support vector machines, Bayesian classifiers, and the like.

[0087] Therefore, the historical itinerary-related information (such as the above-mentioned travel time information) of multiple historical itineraries can be obtained as sample data, and the historical itinerary-related information can be annotated (i.e., the sample data can be labeled), and the corresponding label is the standard travel distance of the historical itinerary obtained. Among them, in order to make the prediction accuracy of the trained distance model higher, the accuracy of the label should be guaranteed. In this regard, map data (the location information in the map data is relatively accurate and has passed verification) can be used to calculate the travel distance between the two locations involved in the historical itinerary as the standard travel distance. Alternatively, the standard travel distance can also be obtained by manual on-site survey. Of course, any other method can also be used, as long as the standard travel distance can accurately represent the actual travel distance.

[0088] Based on the training of the distance model, for the determination of the estimated travel distance in step 104, the travel-related information can be input into the distance model, so as to determine the estimated travel distance between the location to be verified and each reference location according to the output result of the distance model.

[0089] For ease of understanding, the specific implementation methods of the above steps are described in detail below using the online instant delivery scenario as an example.

[0090] In the online instant delivery scenario, users order goods from merchants online (e.g., using a smartphone app or website) through a digital platform (i.e., an online delivery service provider). The platform generates a delivery order for the goods and designates a rider to take over the delivery order. The rider picks up the goods at the merchant’s physical store and then delivers the goods to the user. If the user receives the goods after the time limit (e.g., 30 minutes), the delivery order is considered to have timed out and the platform needs to compensate the user. At the same time, the platform will track the real-time status of the order (i.e., order status data) and record it for accounting purposes and to display it to users for a better user experience. The order status data records the four main events from the rider’s order placement to delivery: (1) accepting the order, (2) arriving at the merchant, (3) leaving the merchant (taking away the goods), and (4) delivering to the user. Since riders receive a large number of orders for delivery every day, especially during peak hours, they usually pick up goods corresponding to multiple orders from different merchants at the same time and then deliver them to users one by one. Therefore, for two adjacent orders undertaken by the rider, the travel time of the rider between the merchant stores corresponding to the two orders can be obtained based on the departure time of one order and the arrival time of the next order. In other words, when the above-mentioned location to be verified and the reference location are physical stores, the travel time information between any two physical stores (the first physical store and the second physical store) can be obtained in the following way: obtain the departure time information of any delivery order undertaken by the delivery party and the arrival time information of the next delivery order, and determine the travel time information between the first physical store corresponding to the delivery order and the second physical store corresponding to the next delivery order (of the delivery order) based on the departure time information and the arrival time information.

[0091] Then, for a large number of merchant pairs, the travel time between a pair of consecutively visited merchant stores is correlated with the travel distance between the merchant stores. Based on this correlation, the travel time can be used to estimate the travel distance between any two merchants. Compared with the travel distance obtained by the registered location, the travel distance obtained based on the travel time is more accurate. Based on this, the travel time between any two merchants' physical stores can be inferred by the arrival time and departure time obtained from the order status data. Then, the travel time between two merchants is used to infer the travel distance between the two merchants, so as to use the travel distance from one merchant to a nearby merchant to cross-verify whether the merchant has a suspicious registered location.

[0092] The following is an introduction to the method of collecting order status data.

[0093] Manual collection strategy that relies on the rider’s smartphone. The rider can manually report the order status through the delivery client program installed on the mobile phone during the delivery process. For example, the time information of arrival at the merchant store, leaving the merchant store, and delivery to the user, etc.

[0094] Automatic collection strategy based on the physical Beacon check-in system. In order to automatically obtain the time when the rider arrives at and leaves the merchant store, Beacon devices can be deployed in the merchant store. Since the store is equipped with Beacon devices, the ID of the Beacon device corresponds to the merchant ID one by one. The Beacon device continuously broadcasts Bluetooth signals to the outside, and when the rider's smartphone is close to the Beacon device, it can detect the Bluetooth signal it emits. Then, the rider's smartphone will upload the timestamp of the Beacon device and the Beacon device ID to the server (with the rider's authorization), which means that the rider is around the store corresponding to the Beacon device at this time, so the rider's arrival and departure time can be automatically obtained.

[0095] Automatic collection strategy based on virtual beacon check-in system. The virtual beacon system can be deployed on the delivery client program on the merchant side, which is part of the application used by the merchant to manage the orders received on the platform. After obtaining the merchant's consent, the virtual beacon Bluetooth broadcast module has been added to the merchant's application. When the rider approaches the merchant's store, the rider's smartphone will receive a virtual beacon message generated by the merchant's smartphone. The collection mechanism is the same as the physical beacon device mentioned above. The virtual beacon message does not contain continuous GPS tracking to protect the merchant's privacy, but relies on the dynamic ID to be mapped with the merchant ID on the server, which corresponds to the merchant, for virtual beacon arrival detection.

[0096] Based on the above collection strategy, the following data in the time dimension can be obtained: T1, manually collected timestamp (Man) input by the rider; T2, timestamp from physical Beacon (Phy). According to the records from the physical Beacon device, the accurate timestamp when the rider passes the merchant can be automatically obtained. The first timestamp of the same merchant monitored by the rider is regarded as the rider's arrival at the merchant, and the last timestamp is regarded as the rider leaving the merchant; T3, timestamp from virtual Beacon (Vir). The method of obtaining timestamps from virtual Beacon devices is the same as that of physical Beacon devices.

[0097] Since the travel time information is related to the location corresponding to the trip, in addition to the time dimension, it is also necessary to collect data in the spatial dimension, that is, the location information of each merchant store. Based on the above collection strategy, the following data in the spatial dimension can be obtained: L1, the GPS location of the rider (GPS). When the rider reports arrival or departure on his application (i.e., manually collects timestamps) or his application detects a Beacon signal (i.e., automatically collects timestamps based on a physical or virtual Beacon system), the application will obtain its GPS record. At this time, the rider's location can be considered as the location of the merchant store; L2, the merchant's registered location (Reg), which can be obtained by matching the merchant ID uploaded by the rider (from a manually reported or detected Beacon device) with the merchant ID in the database. Each merchant ID is associated with a unique location registered by the merchant; L3, the merchant's actual location, which can be obtained by manually going to the field for field research, such as regularly visiting some merchant stores with abnormal orders.

[0098] Based on the above-mentioned data in time and space dimensions, location errors can be identified and the real location of the merchant store can be inferred. Figure 3 As shown, different combinations of spatiotemporal measurements can be made for the collected temporal and spatial data, and for each combination a distance model can be trained that estimates the travel distances between nearby merchants. Merchants are then located based on the travel distances output by the model and compared with the registered locations of these merchants to discover merchant stores with incorrect locations.

[0099] Reg+Man, uses manually entered order status data (T1) and merchant registration location (L2). Reg+Vir, uses records obtained from virtual Beacon devices (T3) and merchant registration location (L2). Reg+Phy, uses records obtained from physical Beacon devices (T2) and merchant registration location (L2). GPS+Phy, uses records obtained from physical Beacon devices (T2) and rider GPS location (L1). GPS+Man, uses rider GPS location (L1) and manually entered order status data (T1). GPS+Vir, uses rider GPS location (L1) and records obtained from virtual Beacon devices (T3).

[0100] Given the inherent homogeneity of spatiotemporal data, the distance estimation model can take any of the five types of measurement data mentioned above as input and output the travel distance. Travel distance is related to different types of features, such as time features, spatial features, and personalized features. Time features mainly reflect time-related information, including travel time, peak hours, latest delivery time, etc. Spatial features reflect regional information, including location, number of orders per unit area, delivery difficulty, etc. Personalized features reflect the behavior and preferences of riders, such as the number of orders carried, the number of historical overdue orders, and historical travel distance.

[0101] For example, time information may include at least one of the following: travel time, peak time period, longest delivery time, shortest estimated travel time, historical average delivery time, team historical delivery time, promised delivery time, time slice, merchant historical meal delivery time, merchant historical meal pickup time, merchant historical meal delivery time. Spatial information may include at least one of the following: delivery difficulty, possible delay ratio, merchant popularity, number of completed orders in the last 30 minutes, mall ID. Personalized information may include at least one of the following: number of orders, rider level, number of overtime orders, rider order acceptance time, average historical travel distance, rider popularity, merchant historical delayed meal delivery rate.

[0102] In summary, when the location to be verified and the reference location are physical stores, the itinerary-related information between the location to be verified and any physical store may also include at least one of the following: the number of delivery orders contained in the area to which the location to be verified belongs, the number of delivery orders contained in the area to which any physical store belongs, the road condition information between the location to be verified and any physical store, the current delivery status information of the delivery party traveling from one of the location to be verified and any physical store to the other, and the delivery preference information of the delivery party.

[0103] Based on the features selected above, since GBDT can learn the nonlinear relationship between features and labels, the distance model can be obtained based on GBDT (Gradient Boosting Decision Tree). Moreover, as a decision tree model, GBDT can explain the importance of features, thereby revealing the rider's mobile behavior pattern. The following introduces the labels, outputs, and loss functions of the distance model.

[0104] Tags: map travel distance d map (p,q), the travel distance between merchants p and q calculated in the map application can be called the "map travel distance" d map(p,q), and use it as a label to train a supervised GBDT model. Considering that different modes of travel (for example, walking or electric bicycles) correspond to different travel distances, two models can be trained separately: one is a walking model for shorter distances (for example, <100m), and the other is an electric bicycle model for longer distances (for example, 100m). In this regard, dbike can be set as the long-distance threshold to determine which model should be selected for the sample data. Of course, in order to ensure that the prediction of the distance model is accurate enough, the rider’s GPS data can also be collected during the delivery process with the rider’s authorization, so that the actual travel distance can be obtained using the GPS data as a label.

[0105] Output: Model estimated distance d model (p,q), the distance between merchant p and merchant q predicted by the distance model is recorded as the model estimated distance d model (p,q).

[0106] Loss function: Use MAE (Mean Absolute Error) as the loss function for training the model.

[0107] Step 106, locate the place to be verified according to the declared position of each reference place and the estimated travel distance to obtain the estimated position of the place to be verified, and match the estimated position with the declared position of the place to be verified to determine whether the declared position of the place to be verified passes the verification based on the matching result.

[0108] When locating the location to be checked based on the declared positions of each reference location and the estimated travel distance obtained by estimation, an error condition can be set, and then the target location can be searched so that the error between the standard travel distance between the location of the target location and the declared positions of each reference location and the estimated travel distance satisfies the preset error condition, thereby taking the location of the target location as the estimated location. For example, the error condition can be set so that the error is within the error threshold, or the target location with the smallest error is selected as the estimated location. Of course, the specific form of the error condition can be flexibly set according to the actual situation, and this specification does not limit this.

[0109] Specifically, you can find your target location in the following ways:

[0110]

[0111] represents the target location, q represents any location, q * Indicates the location to be verified;

[0112] P = {p i ,i=1...N},pi represents reference location i, and N represents the number of reference locations;

[0113] d s (p i ,q) indicates p i The standard travel distance between the declared position of and the position of q;

[0114] d e (p i ,q * ) indicates p i With q * Estimated travel distance between.

[0115] Similarly, the standard travel distance may include the travel distance between the location of the target location calculated using the map data and the declared location of each reference location.

[0116] For ease of understanding, the following explanation is continued using the example of the above distance model.

[0117] Based on the distance estimated by the model trained above, each merchant store can be located and checked to see if it deviates significantly from the registered location. * , called the "target merchant" (the registered location may be fraudulent), given a set of nearby merchants P = {p i ,i=1…N}, called “reference merchants”. model (p i ,q * ) is obtained from the merchant q through the distance model * To the merchant i The goal is to estimate the distance based on the locations P and d of these reference merchants. model (p i ,q * ) Estimate the location of a target merchant Then the calculated With q * Compare them to determine whether there is a significant difference between the two (for example, >200m) based on the comparison results. The above process is performed for each merchant in turn, that is, traversing each merchant, each merchant is once the target merchant and also the reference merchant for nearby merchants. In particular, let G = (V, E) be the road network, where V is the set of road intersections, E = {e v1，v2 :v1,v2∈V} is a set of road segments. Figure 4 As shown, given two merchants p1 and p2, d map (p1,p2) is the map travel distance between the two. For example, d map (p1,p2) equals dmap (p1,v1)+|e v1，v2 |+d map (v2,p2), where |e v1，v2 | is the length of the road section, d map (p1,v1)i is the side length |e p1，v1 |. It should be noted that since riders rarely take detours, this manual specifies the shortest route when calculating the map trip distance. Therefore, d map (p1,p2) corresponds to the route (p1,v1,v2,p2) rather than (p1,v4,v3,p2).

[0118] For the above-defined goal, the problem can be described as: given a set of reference merchants (P = {p i , i=1…N}) and the model estimated distance {d model (p i ,q * ),i=1…N}, the merchant location problem is to find Let the model estimate the distance d model (p i ,q * ) and map travel distance The error is the smallest, for all p i ∈P, that is

[0119]

[0120] This best estimated position Should be registered with the merchant location * If not, it is detected as a suspicious merchant and manually verified. The above problem is based on the travel distance of merchants on the road network. There is only a limited travel distance between merchants, and there are no edges between merchants. Considering that the problem is based on the travel distance between merchants on the road network, this specification refers to it as "graph-based multi-edge positioning".

[0121] In order to solve the multi-lateral positioning problem based on the graph, this specification designs a segment search algorithm, the basic idea of ​​which is to scan all segments and search for the optimal position for each segment. The solution in the algorithm is globally optimal in polynomial time, which makes it possible to extend to periodic large-scale detection, for example, once a day or a week, to reduce or minimize the delay of fraud detection.

[0122] Road segment search algorithm:

[0123] Input: Road network G = (V, E); Reference merchant P = {p i , i=1…N}; model estimates distance d model (pi ,q * )

[0124] Output: estimated position

[0125] min_g←∞

[0126] for do

[0127]

[0128]

[0129] if g(q) <min_g then

[0130]

[0131] Here, g(q) is a convex function about q(d(u,p)), which can be solved directly by Newton's method. Each edge e must be iterated, so the computational complexity of the algorithm is θ(|E|), which is a polynomial.

[0132] It should be noted that merchants near the target merchant can verify the correctness of each other's location, thereby helping with positioning. Merchant pairs that are farther away have fewer rider tracks and therefore provide little information. More importantly, long-distance trips may bring greater variation noise to the data. Therefore, the distance model and positioning algorithm can be run in a small range.

[0133] The above small-scale setting can be achieved by clustering merchants into small subsets. Distance estimation and positioning are performed within each cluster. All merchants can be clustered based on the following conditions: the urban area and suburbs are divided; there is a significant difference in the distance within and between clusters, for example, the distance within a cluster is 2KM, and the distance between clusters is at least 5KM; the coverage radius of a single cluster is no more than 2KM.

[0134] Among them, the hierarchical clustering algorithm BIRCH can be used to cluster merchants. BIRCH is a widely used clustering algorithm that clusters from bottom to top, starting from a single node, and iteratively searches for a pair of points with the smallest distance between clusters. If the distance is close enough, they are merged. When all iterations are completed, clusters with less than 4 merchants are merged into nearby clusters, because at least three points are required to locate the fourth point. This manual defines the center point pˉ of merchant cluster P as its average position. The radius R(P) of cluster P is the farthest travel distance from the merchant to the center point. Given two merchant clusters P i and P j , the distance between clusters D out (P i ,P j ) is defined as Pi and P j The minimum travel distance between merchants.

[0135] Merchant clustering:

[0136] Input: Merchant set N = {p i}

[0137] Output: Merchant cluster collection

[0138] where P i ←{p i}

[0139] while do

[0140] Find P i , With minimum intra-class distance

[0141] if D out (P i ,P j )>αthen

[0142] Break;

[0143] P i ,P j Merge into a cluster P *

[0144] If |P i |<4, merge P i To the nearest cluster.

[0145] It can be seen from the above embodiments that the present specification automatically senses and corrects outdoor merchant location errors based on the order status data of the online instant delivery platform, thereby eliminating the need for additional labor costs and being able to complete position correction without the rider noticing, facilitating low-cost large-scale deployment.

[0146] Corresponding to the above method embodiments, this specification also provides corresponding device embodiments.

[0147] Figure 5 is a schematic structural diagram of a device provided by an exemplary embodiment. Figure 5At the hardware level, the device includes a processor 502, an internal bus 504, a network interface 506, a memory 508, and a non-volatile memory 510, and may also include hardware required for other services. One or more embodiments of this specification may be implemented based on software, such as the processor 502 reading the corresponding computer program from the non-volatile memory 510 into the memory 508 and then running it. Of course, in addition to the software implementation, one or more embodiments of this specification do not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0148] Please refer to Figure 6 , the position verification device can be used for example Figure 5 The device shown in the figure is used to implement the technical solution of this specification. The position verification device may include:

[0149] A determination unit 61 determines a location to be verified and a plurality of reference locations corresponding to the location to be verified, and obtains declared positions of the location to be verified and the reference locations;

[0150] An estimation unit 62 determines an estimated travel distance between the location to be verified and each reference location according to travel related information between the location to be verified and each reference location, wherein the travel related information includes travel time information between the location to be verified and each reference location;

[0151] The verification unit 63 locates the place to be verified according to the declared positions of each reference place and the estimated travel distance obtained by estimation to obtain the estimated position of the place to be verified, and matches the estimated position with the declared position of the place to be verified to determine whether the declared position of the place to be verified passes the verification according to the matching result.

[0152] Optionally, the determining unit 61 is specifically configured to:

[0153] Get a location set containing multiple locations within a preset range;

[0154] Each location in the location set is traversed to sequentially use the current location as the location to be verified, and use other locations different from the current location as the reference locations.

[0155] Optionally, the estimating unit 62 is specifically configured to:

[0156] Inputting the trip related information into a distance model, wherein the distance model is obtained by training the historical trip related information of a plurality of historical trips using a supervised learning algorithm, wherein the historical trip related information is annotated with the standard trip distance of the historical trip;

[0157] The estimated travel distance between the location to be verified and each reference location is determined according to the output result of the distance model.

[0158] Optionally, the estimating unit 62 is further configured to:

[0159] Obtain the departure time information of any delivery order undertaken by the delivery party and the arrival time information of the next delivery order;

[0160] The travel time information between the first physical store corresponding to any one of the delivery orders and the second physical store corresponding to the next delivery order is determined according to the store departure time information and the store arrival time information.

[0161] Optionally, the verification unit 63 is specifically used for:

[0162] Finding a target location, wherein the error between the standard travel distance between the location of the target location and the declared location of each reference location and the estimated travel distance satisfies a preset error condition;

[0163] The position of the target location is used as the estimated position.

[0164] Optionally, the target location is found by:

[0165]

[0166] represents the target location, q represents any location, q * Indicates the location to be verified;

[0167] P = {p i ,i=1...N},p i represents reference location i, and N represents the number of reference locations;

[0168] d s (p i ,q) indicates p i The standard travel distance between the declared position of and the position of q;

[0169] d e (p i ,q * ) indicates p i With q * Estimated travel distance between.

[0170] Optionally, the standard travel distance includes a travel distance between the position of the target location calculated using map data and the declared position of each reference location.

[0171] Optionally, when the location to be verified and the reference location are physical stores, the itinerary-related information between the location to be verified and any physical store further includes at least one of the following:

[0172] The number of delivery orders contained in the area to which the location to be verified belongs, the number of delivery orders contained in the area to which any one of the physical stores belongs, the road condition information between the location to be verified and any one of the physical stores, the current delivery status information of the delivery party traveling from one of the location to be verified and any one of the physical stores to the other, and the delivery preference information of the delivery party.

[0173] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which may be in the form of a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver, a game console, a tablet computer, a wearable device or a combination of any of these devices.

[0174] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0175] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0176] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0177] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0178] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0179] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0180] It should be understood that although the terms first, second, third, etc. may be used to describe various information in one or more embodiments of this specification, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0181] The above description is merely a preferred embodiment of one or more embodiments of the present specification and is not intended to limit one or more embodiments of the present specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present specification shall be included in the scope of protection of one or more embodiments of the present specification.

Claims

1. A position verification method, characterized in that: include: Determine a location to be verified and several reference locations corresponding to the location to be verified, and obtain the declared positions of the location to be verified and the reference locations, wherein the reference location is a location that verifies the declared position of the location to be verified using its own declared position; Determine the estimated travel distance between the location to be verified and each reference location according to the travel related information between the location to be verified and each reference location, wherein the travel related information includes the travel time information between the location to be verified and each reference location; The place to be verified is located according to the declared position of each reference place and the estimated travel distance obtained in advance to obtain the estimated position of the place to be verified, and the estimated position is matched with the declared position of the place to be verified to determine whether the declared position of the place to be verified has passed the verification according to the matching result; the estimated position is the position of the target place where the error between the standard travel distance between the declared position of each reference place and the estimated travel distance meets the preset error condition.

2. The method according to claim 1, characterized in that The step of determining a location to be verified and a plurality of reference locations corresponding to the location to be verified includes: Get a location set containing multiple locations within a preset range; Each location in the location set is traversed to sequentially use the current location as the location to be verified, and use other locations different from the current location as the reference locations.

3. The method according to claim 1, characterized in that The step of determining the estimated travel distance between the location to be verified and each reference location according to the travel related information between the location to be verified and each reference location respectively includes: Inputting the trip related information into a distance model, wherein the distance model is obtained by training the historical trip related information of a plurality of historical trips using a supervised learning algorithm, wherein the historical trip related information is annotated with the standard trip distance of the historical trip; The estimated travel distance between the location to be verified and each reference location is determined according to the output result of the distance model.

4. The method according to claim 1, characterized in that: In the case where the location to be verified and the reference location are physical stores, the method further includes: Obtain the departure time information of any delivery order undertaken by the delivery party and the arrival time information of the next delivery order; The travel time information between the first physical store corresponding to any one of the delivery orders and the second physical store corresponding to the next delivery order is determined according to the store departure time information and the store arrival time information.

5. The method according to claim 1, characterized in that Find the target location by: represents the target location, q represents any location, q * Indicates the location to be verified; P = {p i ,i=1...N},p i represents reference location i, and N represents the number of reference locations; d s (p i ,q) indicates p i The standard travel distance between the declared position of and the position of q; d e (p i ,q * ) indicates p i With q * Estimated travel distance between.

6. The method according to claim 1, characterized in that The standard travel distance includes the travel distance between the position of the target location calculated using map data and the declared position of each reference location.

7. The method according to claim 1, characterized in that In the case where the location to be verified and the reference location are physical stores, the itinerary-related information between the location to be verified and any physical store further includes at least one of the following: The number of delivery orders contained in the area to which the location to be verified belongs, the number of delivery orders contained in the area to which any one of the physical stores belongs, the road condition information between the location to be verified and any one of the physical stores, the current delivery status information of the delivery party traveling from one of the location to be verified and any one of the physical stores to the other, and the delivery preference information of the delivery party.

8. A position verification device, characterized in that: include: a determination unit, determining a location to be verified and a plurality of reference locations corresponding to the location to be verified, and obtaining declared positions of the location to be verified and the reference locations, wherein the reference location is a location for verifying the declared position of the location to be verified using its own declared position; An estimation unit, determining an estimated travel distance between the location to be verified and each reference location according to travel related information between the location to be verified and each reference location, wherein the travel related information includes travel time information between the location to be verified and each reference location; A verification unit locates the place to be verified according to the declared positions of each reference place and the estimated travel distance obtained in advance to obtain the estimated position of the place to be verified, and matches the estimated position with the declared position of the place to be verified to determine whether the declared position of the place to be verified has passed the verification according to the matching result; the estimated position is the position of the target place where the error between the standard travel distance between the declared positions of each reference place and the estimated travel distance meets the preset error condition.

9. The device according to claim 8, characterized in that The determining unit is specifically used for: Get a location set containing multiple locations within a preset range; Each location in the location set is traversed to sequentially use the current location as the location to be verified, and use other locations different from the current location as the reference locations.

10. The device according to claim 8, characterized in that The estimation unit is specifically used for: Inputting the trip related information into a distance model, wherein the distance model is obtained by training the historical trip related information of a plurality of historical trips using a supervised learning algorithm, wherein the historical trip related information is annotated with the standard trip distance of the historical trip; The estimated travel distance between the location to be verified and each reference location is determined according to the output result of the distance model.

11. The device according to claim 8, characterized in that The estimation unit is also used for: Obtain the departure time information of any delivery order undertaken by the delivery party and the arrival time information of the next delivery order; The travel time information between the first physical store corresponding to any one of the delivery orders and the second physical store corresponding to the next delivery order is determined according to the store departure time information and the store arrival time information.

12. The device according to claim 8, characterized in that Find the target location by: represents the target location, q represents any location, q * Indicates the location to be verified; P = {p i ,i=1...N},p i represents reference location i, and N represents the number of reference locations; d s (p i ,q) indicates p i The standard travel distance between the declared position of and the position of q; d e (p i ,q * ) indicates p i With q * Estimated travel distance between.

13. The device according to claim 8, characterized in that The standard travel distance includes the travel distance between the position of the target location calculated using map data and the declared position of each reference location.

14. The device according to claim 8, characterized in that In the case where the location to be verified and the reference location are physical stores, the itinerary-related information between the location to be verified and any physical store further includes at least one of the following: The number of delivery orders contained in the area to which the location to be verified belongs, the number of delivery orders contained in the area to which any one of the physical stores belongs, the road condition information between the location to be verified and any one of the physical stores, the current delivery status information of the delivery party traveling from one of the location to be verified and any one of the physical stores to the other, and the delivery preference information of the delivery party.

15. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor implements the method according to any one of claims 1 to 7 by running the executable instructions.

16. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instruction is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images