Position prediction method, system, electronic device and storage medium

Abstract

The embodiment of the application provides a position prediction method, system, electronic equipment and storage medium, and belongs to the technical field of artificial intelligence. The position prediction method comprises the following steps: acquiring first position information of a sample object at a plurality of historical moments and second position information of the sample object at a current moment; marking according to a corresponding relationship between the first position information and the second position information to obtain marking information, and determining a target variable according to the marking information; constructing a feature according to the first position information to obtain sample features, inputting the sample features and the target variable into a preset position prediction model, adjusting parameters of the position prediction model according to an output result of the position prediction model; acquiring third position information of a target object at a plurality of historical moments; constructing a feature according to the third position information to obtain target features, and inputting the target features into the position prediction model with the adjusted parameters to obtain a position prediction result, so that the accuracy of position prediction can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a position prediction method and system, an electronic device, and a storage medium. BACKGROUND

[0002] Predicting the spatio-temporal position of the occurrence of a target has many application values, and can improve the processing capacity of data and provide more intelligent services for users.

[0003] In related technologies, statistical methods are usually used to predict the position of a target, for example, by comparing the places with a higher frequency of occurrence at historical time points to the spatio-temporal geographical position of the corresponding time points. However, this method cannot predict various possible results, and therefore has low accuracy in position prediction. SUMMARY

[0004] The main purpose of the embodiments of the present application is to provide a position prediction method and system, an electronic device, and a storage medium, which can improve the accuracy of position prediction.

[0005] To achieve the above purpose, a first aspect of the embodiments of the present application provides a position prediction method, which comprises: obtaining first position information of a sample object at a plurality of historical time points and second position information of the sample object at a current time point; marking according to a corresponding relationship between the first position information and the second position information to obtain marking information, and determining a target variable according to the marking information; constructing features according to the first position information to obtain sample features, inputting the sample features and the target variable into a preset position prediction model, and adjusting parameters of the position prediction model according to an output result of the position prediction model; obtaining third position information of a target object at a plurality of historical time points; constructing features according to the third position information to obtain target features, and inputting the target features into the position prediction model with adjusted parameters to obtain a position prediction result.

[0006] In some embodiments, the marking according to the corresponding relationship between the first position information and the second position information to obtain the marking information, and determining the target variable according to the marking information comprises: marking the first position information that is the same as the second position information to obtain first marking information; marking the first position information that is not the same as the second position information to obtain second marking information; taking the first position information containing the first marking information as the target variable, or taking the first position information containing the second marking information as the target variable.

[0007] In some embodiments, the sample objects have a plurality, and the feature construction according to the first position information obtains a sample feature, including: obtaining a position number represented by the first position information of each of the sample objects and a first occurrence number of the sample objects; obtaining a first object number of the sample objects on each of the first position information and a second occurrence number of each of the first position information; and performing feature conversion according to a plurality of the position number, the first occurrence number, the first object number and the second occurrence number to obtain a sample feature.

[0008] In some embodiments, the sample feature and the target variable are input into a preset position prediction model, and parameters of the position prediction model are adjusted according to an output result of the position prediction model, including: splitting the sample feature and the target variable into training samples and test samples; inputting the training samples into the preset position prediction model to obtain a first output result, and adjusting the parameters of the position prediction model according to the first output result; inputting the training samples into the position prediction model with adjusted parameters to obtain a second output result, and inputting the test samples into the position prediction model with adjusted parameters to obtain a third output result; calculating a stability index according to the second output result and the third output result, and determining whether to continue adjusting the parameters of the position prediction model according to the stability index.

[0009] In some embodiments, the second output result includes a first predicted probability value, and the second output result includes a second predicted probability value; the stability index is calculated according to the second output result and the third output result, including: performing addition calculation on the first predicted probability value and the second predicted probability value to obtain a first value; performing logarithmic calculation on the first predicted probability value and the second predicted probability value to obtain a second value; and obtaining a stability index according to a product of the first value and the second value.

[0010] In some embodiments, the second output result includes a first predicted probability value, and the second output result includes a second predicted probability value, and the first predicted probability value and the second predicted probability value each have a plurality; the stability index is calculated according to the second output result and the third output result, including: dividing a plurality of score segments according to the size of the first predicted probability value and the second predicted probability value; calculating a sub-stability index of the first predicted probability value and the second predicted probability value in the corresponding score segment, respectively; and accumulating the sub-stability index in each of the score segments to obtain a stability index.

[0011] In some embodiments, the feature construction according to the third position information obtains a target feature, and the target feature is input into the position prediction model after the parameter adjustment to obtain a position prediction result, including: performing feature construction according to a plurality of third position information respectively to obtain a plurality of target features; inputting a plurality of target features into the position prediction model after the parameter adjustment respectively to obtain a plurality of target prediction positions and corresponding target prediction probability values; and determining a position prediction result in a plurality of prediction positions according to the size of each target prediction probability value.

[0012] To achieve the above object, a second aspect of the embodiments of the present application provides a position prediction system, the system comprising: a sample object acquisition module, configured to acquire first position information of a sample object at a plurality of historical moments and second position information of the sample object at a current moment; a target variable acquisition module, configured to obtain marked information by marking according to a corresponding relationship between the first position information and the second position information, and determine a target variable according to the marked information; a parameter adjustment module, configured to obtain sample features by feature construction according to the first position information, input the sample features and the target variable into a preset position prediction model, and adjust parameters of the position prediction model according to an output result of the position prediction model; a target object acquisition module, configured to acquire third position information of a target object at a plurality of historical moments; and a position prediction module, configured to obtain target features by feature construction according to the third position information, and input the target features into the position prediction model after the parameter adjustment to obtain a position prediction result.

[0013] To achieve the above object, a third aspect of the embodiments of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the method of the first aspect of the embodiments when executing the computer program.

[0014] To achieve the above object, a fourth aspect of the embodiments of the present application provides a storage medium, the storage medium being a computer readable storage medium, the storage medium storing a computer program, and the computer program being executed by a processor to implement the method of the first aspect of the embodiments.

[0015] The position prediction method, system, electronic device and storage medium provided in the application can be applied in a position prediction system. By executing the position prediction method, the first position information of a sample object at a historical moment and the second position information at a current moment can be trained first, wherein the application embodiment marks the corresponding relationship of the two kinds of position information to obtain marking information, and the marking information is used as a target variable in the process of training the position prediction model. Therefore, sample features can be obtained by constructing features according to the first position information, and the sample features and the target variable are input into a preset position prediction model to adjust the parameters of the position prediction model through training. Subsequently, in the application process, the third position information at the historical moment can be input into the trained position prediction model after feature construction to obtain the position prediction result of the target object. In the application embodiment, the position prediction is converted into a predictable mode by constructing the target variable, which is helpful for the learning of the model, changes the data into a supervised learning mode, and finally predicts whether the object appears at the historical position, thereby improving the accuracy of position prediction. BRIEF DESCRIPTION OF DRAWINGS

[0016] Fig. 1 is a flowchart of the position prediction method provided in the application embodiment;

[0017] Fig. 2 is a flowchart of step S102 in Fig. 1 ;

[0018] Fig. 3 is a flowchart of step S103 in Fig. 1 ;

[0019] Fig. 4 is a flowchart of step S103 in Fig. 1 ;

[0020] Fig. 5 is a flowchart of step S404 in Fig. 4 ;

[0021] Fig. 6 is a flowchart of step S404 in Fig. 4 ;

[0022] Fig. 7 is a flowchart of step S105 in Fig. 1 ;

[0023] Fig. 8 is a functional module schematic diagram of the position prediction system provided in the application embodiment;

[0024] Fig. 9 is a hardware structure schematic diagram of the electronic device provided in the application embodiment. DETAILED DESCRIPTION

[0025] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not intended to limit the present application.

[0026] It should be noted that although the functional modules are divided in the device schematic diagram, and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a manner different from the module division in the device or the sequence in the flowchart. The terms "first", "second", etc. in the specification and claims and the above-described drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0028] First, the terms involved in the present application are analyzed:

[0029] Artificial intelligence (AI): is a new technical science that studies, develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence; artificial intelligence is a branch of computer science, and artificial intelligence aims to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. The research in this field includes robots, language recognition, image recognition, natural language processing and expert systems, etc. Artificial intelligence can simulate the information process of human consciousness and thinking. Artificial intelligence is also the theory, method, technology and application system of using digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0030] Under the premise of providing location services for the target, predicting the spatio-temporal location of the target has more application value, for example, traffic planning can be carried out according to the location information provided by the target, or intelligent push of service items can also be carried out according to the location information provided by the target, therefore, predicting the location of the target can improve the data processing capability and provide more intelligent services for users.

[0031] In the related art, a statistical method is usually used to predict the position of a target, for example, by taking a place with a higher frequency of occurrence at a historical time point as the spatiotemporal geographical position of the corresponding time point. However, the applicant finds that this method cannot predict various possible results and can only make a judgment based on the number of occurrences at a single historical moment. For positions with other numbers of occurrences that are not the most, the method cannot make a prediction. Alternatively, the related art uses a time sequence method to predict the position. However, each time point of each target is not necessarily continuous, and more processing is required, and the effect is not necessarily good. Therefore, the accuracy of position prediction is low.

[0032] Based on this, the embodiments of the present application provide a position prediction method, system, electronic device and storage medium. By executing the position prediction method, the first position information of the sample object at the historical time point and the second position information at the current time point can be trained first. In the embodiments of the present application, the correspondence between the two kinds of position information is marked to obtain the marking information. The marking information is used as a target variable in the process of training the position prediction model. Therefore, the sample features can be obtained by constructing features based on the first position information. The sample features and the target variable are input into the preset position prediction model. The parameters of the position prediction model are adjusted through training. Subsequently, in the application process, the third position information at the historical time point can be input into the trained position prediction model after feature construction to obtain the position prediction result of the target object. In the embodiments of the present application, the position prediction is converted into a predictable mode by constructing the target variable, which helps the learning of the model, changes the data into a supervised learning mode, and finally predicts whether to appear at the historical position, thereby improving the accuracy of position prediction.

[0033] The position prediction method, system, electronic device and storage medium provided by the embodiments of the present application are specifically described as follows. First, the position prediction method in the embodiments of the present application is described.

[0034] The position prediction method in the embodiments of the present application can be described as follows.

[0035] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Artificial intelligence (AI) is the use of digital computers or computer-controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results.

[0036] The artificial intelligence basic technology generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. The artificial intelligence software technology mainly includes computer vision technology, robot technology, biometric identification technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0037] The position prediction method provided by the embodiments of the present application relates to the technical field of artificial intelligence. The position prediction method provided by the embodiments of the present application can be applied to a terminal, can also be applied to a server end, and can also be software running in the terminal or the server end. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server end can be configured as an independent physical server, can also be configured as a server cluster or a distributed system composed of multiple physical servers, can also be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and big data and artificial intelligence platform; and the software can be an application implementing the position prediction method, but is not limited to the above forms.

[0038] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0039] It should be noted that in various specific embodiments of the present application, when relevant processing needs to be performed on data related to the identity or characteristics of the user, such as user information, user behavior data, user history data, and user location information, the user's permission or consent will be obtained first, for example, when accessing the user's stored data and the user's cached data, the user's permission or consent will be obtained first. Moreover, the collection, use and processing of such data will comply with relevant national and regional laws, regulations and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the user's separate permission or separate consent will be obtained through a pop-up window or by jumping to a confirmation page, and after obtaining the user's separate permission or separate consent, the necessary user-related data for the normal operation of the embodiments of the present application will be obtained.

[0040] Fig. 1 is an optional flowchart of the position prediction method provided by the embodiments of the present application, Fig. 1 The method in the method can include but is not limited to steps S101-S105.

[0041] Step S101, obtaining first position information of a sample object at a plurality of historical time points and second position information of the sample object at a current time point;

[0042] Exemplarily, the position prediction method in the embodiments of the present application can be applied in a position prediction system, which can be a terminal device, for example, a smart phone, a personal computer or a server, or a system composed of multiple devices, which is not specifically limited here.

[0043] In the embodiments of the present application, the first position information of the sample object at a plurality of historical time points can be obtained, and the second position information of the sample object at a current time point can be obtained, and these position information can be used as samples for subsequent model training.

[0044] Exemplarily, the sample object is the object of the training process, and the sample object can be a user, a vehicle, a mobile device, etc.

[0045] The first position information and the second position information of the user can be obtained by the position prediction system after authorization of the user, or the user can input the first position information and the second position information in the position prediction system, or the first position information and the second position information can be data generated in the running process of the position prediction system; the vehicle can be a car, a motorcycle, an electric bicycle or the like, for example, the car can record the first position information and obtain the second position information under the authorization of the user, and send the first position information and the second position information to the position prediction system, or the user can input the first position information and the second position information on the car, and send the first position information and the second position information to the position prediction system by the car; the mobile device can record the first position information and obtain the second position information under the authorization of the user, and send the first position information and the second position information to the position prediction system, or the user can input the first position information and the second position information on the mobile device, and send the first position information and the second position information to the position prediction system by the mobile device.

[0046] For example, the first position information is information representing a location position, and the first position information can be coordinate position, which is used to represent the latitude and longitude information of the sample object, and the first position information can be multiple, representing multiple coordinate positions at different historical moments; the second position information is information representing a location position, and the second position information can be coordinate position, which is used to represent the latitude and longitude information of the sample object, and the second position information is one, representing a coordinate position at a current moment. It can be understood that the first position information and the second position information are only position information at different moments.

[0047] In step S102, the first position information and the second position information are marked according to the corresponding relationship between the first position information and the second position information to obtain marked information, and the target variable is determined according to the marked information.

[0048] For example, the first position information and the second position information can be marked according to the corresponding relationship between the first position information and the second position information to obtain marked information, and the target variable can be determined according to the marked information. It can be understood that the target variable summarizes the information that is expected to be predicted from the perspective of the algorithm used to construct the position prediction model, so the target variable is the result of the machine learning algorithm prediction, and the construction of the target variable in the embodiment of the application is helpful to the subsequent training of the model.

[0049] Exemplarily, in the embodiments of the present application, the first position information and the second position information can be marked according to the similarity relationship therebetween. The corresponding marking information can be obtained according to whether the first position information and the second position information are similar. The marking information can be a field, which is used to indicate the similarity relationship between the first position information and the second position information. In some embodiments, in the embodiments of the present application, the first position information and the second position information are marked as 1 if they are the same, and marked as 0 if they are different, to obtain a target sample. The target sample is used for feature conversion to obtain a target variable. Alternatively, the first position information after being marked with the marking information can be directly used as the target variable, which is not limited herein.

[0050] In step S103, a sample feature is obtained by performing feature construction on the first position information. The sample feature and the target variable are input into a preset position prediction model. The parameters of the position prediction model are adjusted according to the output result of the position prediction model.

[0051] Exemplarily, in the embodiments of the present application, a position prediction model is preset. The features obtained by converting the position information are input into the position prediction model, so that the corresponding position prediction result can be obtained. Alternatively, the position information can be directly input into the position prediction model. The position prediction model can perform feature construction on the position information, so as to realize the processing of the data.

[0052] In the training process, the sample feature is obtained by performing feature construction on the first position information. The sample feature is a vector feature, which can be used as the output data of the model. According to the above embodiments, the target variable is constructed in advance in the embodiments of the present application. Therefore, the sample feature and the target variable are input into a preset position prediction model. The parameters of the position prediction model are adjusted according to the output result of the position prediction model.

[0053] It can be understood that the purpose of the marking is to construct the target variable of the model. When the model is trained, the target variable is needed to learn which features the sample will be in the same situation as the previous one at the next test point, that is, the part with the label information of 1. If the label information is 1, the location of the sample at the next specified time point can be indirectly predicted.

[0054] In step S104, the third position information of the target object at a plurality of historical time points is obtained.

[0055] Exemplarily, after the training of the position prediction model is completed in the above steps, the position prediction model can be applied. In the application process, the third position information of the target object at a plurality of historical time points can be obtained.

[0056] Exemplarily, the target object is an object in the application process. The target object can be a user, a transportation tool, a mobile device, etc.

[0057] The third position information of the user can be obtained by the position prediction system after authorization of the user, or the user can input the third position information in the position prediction system, or the third position information can be data generated during operation of the position prediction system; the vehicle can be a car, a motorcycle, an electric bicycle, or the like, for example, the car can obtain the third position information under authorization of the user and send the third position information to the position prediction system, or the user can input the third position information on the car and send the third position information to the position prediction system by the car; the mobile device can obtain the third position information under authorization of the user and send the third position information to the position prediction system, or the user can input the third position information on the mobile device and send the third position information to the position prediction system by the mobile device.

[0058] For example, the third position information is information representing a location position, the third position information can be coordinate position, used to represent latitude and longitude information of the target object, and the third position information can be multiple, representing multiple coordinate positions at different historical moments.

[0059] In step S105, the target feature is obtained by feature construction according to the third position information, and the target feature is input into the position prediction model after adjustment of the parameters to obtain a position prediction result.

[0060] In the application process, the target feature is obtained by feature construction according to the third position information, the target feature is a vector feature, and the output data of the model can be obtained. According to the above embodiments, the position prediction model is pre-trained in the embodiment of the application, and the parameters of the position prediction model are adjusted in the training process. Therefore, the target feature is input into the preset position prediction model in the embodiment of the application, and a position prediction result can be obtained. It should be noted that the position prediction result is the position prediction of the model output, and the position prediction result contains the position information predicted by the model.

[0061] In the embodiment of the application, a supervised learning algorithm is constructed to predict the possible position of the customer at the next time point. Specifically, the position prediction is converted into a predictable way by constructing a target variable, which helps the learning of the model, changes the data into a supervised learning way, and finally predicts whether to appear in the historical position, thereby improving the accuracy of the position prediction.

[0062] It should be noted that the position prediction is different from the conventional supervised learning, and the position appearing at a specified time needs to be predicted, and the conventional supervised learning generally predicts whether (binary classification) and predicts the amount of data (regression). The method proposed in the embodiment of the present application is to construct samples and target variables, change the data into the commonly used supervised learning mode, finally predict whether to appear in the historical position, that is, a binary classification problem, and derive features from the existing little information to support model training. The embodiment of the present application can add some rules in the statistical method to the model training by processing into features, so the information obtained by the statistical rule can be covered in the algorithm, and the final result will be better than the statistical method.

[0063] Please refer to Fig. 2 In some embodiments, step S102 can include steps S201 to S203:

[0064] Step S201, marking the first position information same as the second position information to obtain first marked information;

[0065] Step S202, marking the first position information different from the second position information to obtain second marked information;

[0066] Step S203, taking the first position information containing the first marked information as a target variable, or taking the first position information containing the second marked information as a target variable.

[0067] For example, in the marking process of the embodiment of the present application, the marking information is determined according to whether the first position information is same as the second position information. Specifically, in the embodiment of the present application, the first position information same as the second position information is marked to obtain the first marked information, and the first position information different from the second position information is marked to obtain the second marked information. Finally, the first position information containing the first marked information is taken as a target variable, or the first position information containing the second marked information is taken as a target variable.

[0068] For example, the first marked information and the second marked information are a field added, for example, a flag field added to the position information. The first marked information can be 1, and when the first position information has a flag field of 1, it indicates that the first position information is same as the second position information. The second marked information can be 0, and when the first position information has a flag field of 0, it indicates that the first position information is different from the second position information.

[0069] For example, in the embodiment of the present application, the second position information of a plurality of sample objects can be obtained from historical data, as shown in Table 1:

[0070] No. Sample object (id) Current time Second position information 1 c001 2022 / 7 / 2 10:01 Position x11 2 c001 2022 / 7 / 2 20:01 Position x21 3 c002 2022 / 7 / 10 10:20 Position x31

[0071] Table 1

[0072] The first position information of the sample object at the historical time point is acquired again, and corresponding mark information is assigned. For example, the serial number 1 corresponds to 3 geographic positions of the sample object c001 appearing before the corresponding time (for example, 3 months), the serial number 2 corresponds to 5 geographic positions appearing before the corresponding time (3 months in general), and the serial number 3 corresponds to 5 geographic positions appearing before the corresponding time (4 months in general). The new data set is shown in Table 2:

[0073]

[0074]

[0075] Table 2

[0076] It should be noted that the first position information with the first mark information and / or the second mark information can be used as a target sample, and the target variable can be obtained according to the target sample. For example, the target sample can be directly used as the target variable, or the target sample can be constructed into a vector feature, so as to obtain the target variable.

[0077] Referring to Fig. 3 In some embodiments, there are multiple sample objects, and the step S103 can include steps S301 to S303:

[0078] In step S301, the number of positions represented by the first position information of each sample object and the first occurrence number of the sample object are acquired.

[0079] In step S302, the first object number of the sample object on each first position information and the second occurrence number of each first position information are acquired.

[0080] In step S303, the feature conversion is performed according to multiple positions, the first occurrence number, the first object number and the second occurrence number, to obtain the sample feature.

[0081] For example, in the process of feature construction in the embodiment of the present application, multi-dimensional construction can be performed according to the specific circumstances of the sample object and the first position information. Specifically, in the embodiment of the present application, the feature construction can be performed from the dimension of the sample, including acquiring the number of positions represented by the first position information of each sample object and the first occurrence number of the sample object. In addition, the feature construction can also be performed from the dimension of the position, including acquiring the first object number of the sample object on each first position information and the second occurrence number of each first position information.

[0082] Finally, in the embodiments of the present application, the sample features can be constructed from the dimension of the sample, the dimension of the location, or the dimension of the sample plus the location. When the feature construction is performed according to the dimension of the sample, the feature conversion can be performed according to the number of locations and the first occurrence times to obtain the sample features; when the feature construction is performed according to the dimension of the location, the feature conversion can be performed according to the number of first objects and the second occurrence times to obtain the sample features; when the feature construction is performed according to the dimension of the sample plus the location, the feature conversion can be performed according to the number of locations, the first occurrence times, the number of first objects, and the second occurrence times to obtain the sample features.

[0083] For example, in the embodiments of the present application, the sample features are obtained by performing feature conversion on the corresponding first location information according to multiple of the number of locations, the first occurrence times, the number of first objects, and the second occurrence times.

[0084] For example, the number of locations can include the number of locations of the sample object in history, the number of locations of the sample object in history on weekdays and holidays, the number of locations of the sample object in history on Monday to Sunday, the number of locations of the sample object in history on Monday to Sunday, the number of locations of the sample object in history at different time periods, etc.

[0085] For example, the first occurrence times can include the number of times of the sample object in history, the number of times of the sample object in history on weekdays and holidays, the number of times of the sample object in history at different time periods, etc.

[0086] For example, the number of first objects can include how many sample objects have accessed each location in history, the number of sample objects appearing in history on weekdays and holidays, the number of sample objects appearing in different weeks and time periods, etc.

[0087] For example, the second occurrence times can include the number of times of each location in history, the number of times of each location in history on weekdays and holidays, the number of times of each location in different weeks and time periods, etc.

[0088] It should be noted that in the embodiments of the present application, the sample features and the target variables obtained according to the above steps can be associated with the id of the sample object to obtain a wide table, which contains three information: id, sample features, and target variables. The data of the wide table is used to input into the location prediction model for training.

[0089] Please refer to Fig. 4 In some embodiments, step S103 can include steps S401 to S404:

[0090] Step S401, splitting the sample features and the target variables into training samples and test samples;

[0091] Step S402, input the training sample into the preset position prediction model to obtain a first output result, and adjust the parameters of the position prediction model according to the first output result.

[0092] Step S403, input the training sample into the position prediction model after the parameter adjustment to obtain a second output result, input the test sample into the position prediction model after the parameter adjustment to obtain a third output result.

[0093] Step S404, calculate a stability index according to the second output result and the third output result, and determine whether to continue adjusting the parameters of the position prediction model according to the stability index.

[0094] For example, the embodiment of the present application can carry out model training after the training data is established. In the embodiment of the present application, the training data is split according to a certain proportion, including splitting the sample features and the target variable into a training set and a test set, the samples in the training set are training samples, the samples in the test set are test samples, the training sample is input into the preset position prediction model to obtain a first output result, and the parameters of the position prediction model are adjusted according to the first output result. In order to further improve the stability of the model, the training sample is input into the position prediction model after the parameter adjustment to obtain a second output result, the test sample is input into the position prediction model after the parameter adjustment to obtain a third output result, a stability index is calculated according to the second output result and the third output result, and it is determined whether to continue adjusting the parameters of the position prediction model according to the stability index.

[0095] It should be noted that the stability index is a population stability index (PSI), and the stability of the model can be quantified by the PSI index.

[0096] In an embodiment, the sample features and the target variable are split according to a certain proportion, for example, according to a proportion of 7 to 3, to obtain a training set and a test set. Under the premise of meeting the requirements of the embodiment of the present application, other proportions can also be used for splitting.

[0097] In an embodiment, the algorithm of the embodiment of the present application can select common machine learning algorithms (such as randomForest, GBDT, lightGBM, Xgboost, etc.), adjust the parameters (such as loss function, tree depth, learning rate, etc.) of the algorithm to obtain the optimal result, and finally test the stability of the model to obtain the final model.

[0098] It should be noted that the output result of the position prediction model includes multiple predicted positions and corresponding predicted probability values, and therefore the second output result in the above embodiment includes the first predicted probability value and the second output result includes the second predicted probability value.

[0099] Referring to Fig. 5 In some embodiments, step S404 can include steps S501 to S503:

[0100] Step S501, according to the first predicted probability value and the second predicted probability value, a first value is obtained by addition calculation;

[0101] Step S502, according to the first predicted probability value and the second predicted probability value, a second value is obtained by logarithmic calculation;

[0102] Step S503, according to the product of the first value and the second value, a stability index is obtained.

[0103] For example, the application discloses a specific formula for calculating the stability index. In the above embodiment, the first predicted probability value of the predicted position in the training sample is obtained according to the training sample, and the second predicted probability value of the predicted position in the test sample is obtained according to the test sample. It can be understood that the first predicted probability value is an actual proportion, and the second predicted probability value is an expected proportion.

[0104] According to the first predicted probability value and the second predicted probability value, a first value is obtained by addition calculation, so the first value is:

[0105] The first value = sum ((the first predicted probability value - the second predicted probability value);

[0106] According to the first predicted probability value and the second predicted probability value, a second value is obtained by logarithmic calculation, so the first value is:

[0107] The second value = ln (the first predicted probability value / the second predicted probability value);

[0108] Finally, according to the product of the first value and the second value, a stability index is obtained, so the formula of the stability index is:

[0109] psi = sum ((the first predicted probability value - the second predicted probability value) * ln (the first predicted probability value / the second predicted probability value)).

[0110] In some embodiments, the second output result includes the first predicted probability value, the second output result includes the second predicted probability value, and the first predicted probability value and the second predicted probability value both have multiple values.

[0111] Referring toFig. 6 In some embodiments, step S404 can include steps S601-S603:

[0112] Step S601, dividing a plurality of score segments according to the sizes of the first prediction probability value and the second prediction probability value;

[0113] Step S602, respectively calculating the first prediction probability value and the second prediction probability value, and the sub-stability indexes under the corresponding score segments;

[0114] Step S603, accumulating the sub-stability indexes under each score segment to obtain the stability index.

[0115] For example, in the embodiments of the present application, the sizes of the first prediction probability value and the second prediction probability value are divided into a plurality of score segments, and the final total stability index is calculated according to the sub-stability indexes under different score segments. The calculation of the sub-stability indexes can refer to the calculation process in steps S501-S503 described above, and will not be repeated here.

[0116] For example, in the embodiments of the present application, the value obtained by multiplying the probability value predicted by the position prediction model by 100 is divided into 10 segments for list statistics. However, the multiplication by 100 is not required as long as it meets the requirements of the embodiments of the present application. Subsequently, the first prediction probability value is distributed in the corresponding score segments. It can be understood that the sum of the first prediction probability values under each score segment is equal to 1. Similarly, the second prediction probability value is also distributed in the corresponding score segments. Similarly, the sum of the second prediction probability values under each score segment is equal to 1. Thus, in the embodiments of the present application, the comparison between the experimental sample and the prediction sample under different prediction probability values can be known.

[0117] After calculating the sub-stability indexes under each score segment, the embodiments of the present application accumulate the sub-stability indexes under each score segment to obtain the stability index. It should be noted that the sub-stability indexes can be directly accumulated to obtain the stability index, or the sub-stability indexes can be multiplied by corresponding coefficients according to the different score segments for weighting to obtain the stability index.

[0118] In the embodiments of the present application, the first prediction probability value is described as A%, and the second prediction probability value is described as B%. Once you can obtain the table of sub-stability indexes under each score segment, as shown in Table 3:

[0119]

[0120] Table 3

[0121] The sub-stability indexes corresponding to each score segment in the last column of Table 3 are accumulated to obtain the final stability index.

[0122] In determining whether the model meets the requirements according to the stability index, the embodiment of the present application determines whether to continue to adjust the parameters of the position prediction model according to the size of the stability index. For example, when the stability index is less than 0.1, it indicates that the model is relatively stable on the test sample; when the stability index is between 0.1 and 0.25, it indicates that the stability of the model on the test sample is general, and whether it can be used can be combined with the actual situation; when the stability index is greater than 0.25, it indicates that the model is unstable on the test sample, and needs to be retrained, and the parameters of the position prediction model are continued to be adjusted.

[0123] It can be understood that the stability described above can be designed to comprehensively weigh several different time windows of test samples. The time window refers to the corresponding time of sample data, for example, the training sample data time window is January to June 2022, and the test sample data time window is July 2022.

[0124] Please refer to Fig. 7 In some embodiments, step S105 can include steps S701 to S703:

[0125] Step S701, respectively constructing features according to the plurality of third position information to obtain a plurality of target features;

[0126] Step S702, respectively inputting the plurality of target features into the position prediction model after adjusting the parameters to obtain a plurality of target predicted positions and corresponding target predicted probability values;

[0127] Step S703, determining a position prediction result in the plurality of predicted positions according to the size of each target predicted probability value.

[0128] For example, when predicting the position according to the position prediction model after adjusting the parameters, different prediction results can be obtained according to different third position information, and the final position prediction result is determined according to the prediction probability value of each result. Specifically, the embodiment of the present application can respectively construct features according to a plurality of third position information to obtain a plurality of target features, and the construction process of the target feature is similar to the above-mentioned sample feature construction process, which will not be repeated here. Subsequently, the plurality of target features are respectively input into the position prediction model after adjusting the parameters to obtain a plurality of target predicted positions and corresponding target predicted probability values. It can be understood that each target predicted position has a corresponding target predicted probability value, and finally, a position prediction result is determined in the plurality of predicted positions according to the size of each target predicted probability value.

[0129] It can be understood that the target prediction position is a third target position corresponding to a target prediction probability value output by the position prediction model. In the embodiment of the present application, the position prediction model is used to predict and calculate the prediction probability of the input position information, and select a position information with the maximum probability as the final position prediction result, that is, the corresponding geographical position with the maximum probability value is taken as the predicted geographical position of the target object at this time point.

[0130] For example, when there are multiple target objects with different ids, the third position information corresponding to each target object at historical time points is shown in Table Four:

[0131] No. Target object (id) Current time Third position information 1 c001 2022 / 7 / 2 10:01 Position x12 1 c001 2022 / 7 / 2 10:01 Position x13 1 c001 2022 / 7 / 2 10:01 Position x11 2 c001 2022 / 7 / 2 20:01 Position x21 2 c001 2022 / 7 / 2 20:01 Position x22 2 c001 2022 / 7 / 2 20:01 Position x23 2 c001 2022 / 7 / 2 20:01 Position x24 2 c001 2022 / 7 / 2 20:01 Position x25 3 c002 2022 / 7 / 10 10:20 Position x31 3 c002 2022 / 7 / 10 10:20 Position x32 3 c002 2022 / 7 / 10 10:20 Position x33 3 c002 2022 / 7 / 10 10:20 Position x34

[0132] Table Four

[0133] The target features obtained from the third position information are input into the position prediction model after adjusting the parameters, to obtain different target prediction positions and corresponding target prediction probability values, as shown in Table Five:

[0134] No. Target object (id) Current time Third position information Target prediction probability value 1 c001 2022 / 7 / 2 10:01 Position x12 0.8 1 c001 2022 / 7 / 2 10:01 Position x13 0.61 1 c001 2022 / 7 / 2 10:01 Position x11 0.12 2 c001 2022 / 7 / 2 20:01 Position x21 0.13 2 c001 2022 / 7 / 2 20:01 Position x22 0.25 2 c001 2022 / 7 / 2 20:01 Position x23 0.36 2 c001 2022 / 7 / 2 20:01 Position x24 0.02 2 c001 2022 / 7 / 2 20:01 Position x25 0.9 3 c002 2022 / 7 / 10 10:20 Position x31 0.2 3 c002 2022 / 7 / 10 10:20 Position x32 0.18 3 c002 2022 / 7 / 10 10:20 Position x33 0.09 3 c002 2022 / 7 / 10 10:20 Position x34 0.9

[0135] Table Five

[0136] Finally, in the embodiment of the present application, the corresponding position information with the maximum target prediction probability value in Table Five is selected as the position prediction result, and the final position prediction result is shown in Table Six:

[0137] No. Target object (id) Current time Position prediction result Target prediction probability value 1 c001 2022 / 7 / 2 10:01 Position x12 0.8 2 c001 2022 / 7 / 2 20:01 Position x25 0.9 3 c002 2022 / 7 / 10 10:20 Position x34 0.9

[0138] Table Six

[0139] Please refer to Fig. 8 The embodiment of the present application also provides a position prediction system, which can implement the above position prediction method. The position prediction system comprises:

[0140] The sample object acquisition module 801 is configured to acquire the first position information of the sample object at a plurality of historical time points and the second position information of the sample object at a current time point.

[0141] The target variable acquisition module 802 is configured to mark the corresponding relationship between the first position information and the second position information to obtain marking information, and determine the target variable according to the marking information.

[0142] The parameter adjustment module 803 is configured to construct sample features according to the first position information, input the sample features and the target variable into a preset position prediction model, and adjust the parameters of the position prediction model according to the output result of the position prediction model.

[0143] The target object acquisition module 804 is configured to acquire third position information of the target object at a plurality of historical time points;

[0144] The position prediction module 805 is configured to perform feature construction according to the third position information to obtain a target feature, and input the target feature into the position prediction model after the adjustment of the parameters to obtain a position prediction result.

[0145] For example, the position prediction system in the embodiments of the present application can perform the position prediction method in the above embodiments. The position prediction system can be a terminal device, for example, a smart phone, a personal computer or a server. The position prediction system can also be a system composed of a plurality of devices, which is not specifically limited here.

[0146] In the process of performing the position prediction method, the position prediction system in the embodiments of the present application can acquire first position information of a sample object at a plurality of historical time points and acquire second position information of the sample object at a current time point, and use the position information as a sample for subsequent model training.

[0147] For example, the sample object is the object of the training process, and the sample object can be a user, a vehicle, a mobile device, etc.

[0148] For example, the first position information and the second position information of the user can be obtained by the position prediction system after authorization of the user, or the user can input the first position information and the second position information in the position prediction system, or the first position information and the second position information can be data generated during the operation of the position prediction system. The vehicle can be a car, a motorcycle or an electric bicycle, etc. For example, the car can record the first position information and acquire the second position information under the authorization of the user, and send the first position information and the second position information to the position prediction system. The user can also input the first position information and the second position information on the car and send them to the position prediction system by the car. The mobile device can record the first position information and acquire the second position information under the authorization of the user, and send the first position information and the second position information to the position prediction system. The user can also input the first position information and the second position information on the mobile device and send them to the position prediction system by the mobile device.

[0149] For example, the first position information is information representing a location position, and the first position information can be a coordinate position, which is used to represent the latitude and longitude information of the sample object. There can be multiple first position information, representing multiple coordinate positions at different historical time points. The second position information is information representing a location position, and the second position information can be a coordinate position, which is used to represent the latitude and longitude information of the sample object. There is one second position information, representing a coordinate position at a current time point. It can be understood that the first position information and the second position information are only position information at different time points.

[0150] Exemplarily, in the embodiments of the present application, the first location information and the second location information can be marked according to the corresponding relationship between the first location information and the second location information, so as to obtain marked label information, and the target variable can be determined according to the label information. It can be understood that the target variable generalizes the information that is expected to be predicted from the perspective of the algorithm for constructing the location prediction model, so the target variable is the result of the machine learning algorithm prediction, and the construction of the target variable is helpful to the subsequent training of the model.

[0151] Exemplarily, in the embodiments of the present application, the first location information and the second location information can be marked according to the similarity relationship between the first location information and the second location information, and corresponding label information can be made according to whether the first location information and the second location information are similar. The label information can be a field, which is used to indicate the similarity relationship between the first location information and the second location information. In some embodiments, the first location information and the second location information are marked as 1 if they are the same, and marked as 0 if they are different, to obtain a target sample, and feature conversion is performed based on the target sample to obtain a target variable. Alternatively, the first location information after being marked with the label information can be directly used as the target variable, which is not limited herein.

[0152] Exemplarily, in the embodiments of the present application, a location prediction model is pre-set, and the features converted from the location information are input into the location prediction model to obtain a corresponding location prediction result. Alternatively, the location information can be directly input into the location prediction model, and the location prediction model can construct features from the location information to realize data processing.

[0153] In the training process, the sample features are obtained by constructing features from the first location information, and the sample features are a kind of vector features, which can be used as the output data of the model. As known from the above embodiments, the target variable is pre-constructed in the embodiments of the present application, so the sample features and the target variable are input into the pre-set location prediction model, and the parameters of the location prediction model are adjusted according to the output result of the location prediction model.

[0154] It can be understood that the purpose of marking is to construct the target variable of the model, and the target variable is needed for the model to learn which features the sample will be in the same situation as the previous one at the next test point, i.e., the part with the label information of 1. If it is 1, the location of the sample at the next specified time point can be indirectly predicted.

[0155] Exemplarily, after the training of the location prediction model is completed in the above steps, the location prediction model can be applied, and the third location information of the target object at a plurality of historical time points can be obtained in the application process.

[0156] Exemplarily, the target object is an object of an application process, and the target object can be a user, a vehicle, a mobile device, etc.

[0157] In the application process, the third location information of the user can be obtained by the location prediction system after authorization of the user, or the user can input the third location information in the location prediction system, or the third location information can be data generated in the running process of the location prediction system; the vehicle can be a car, a motorcycle or an electric bicycle, etc., for example, the car can obtain the third location information under the authorization of the user, and send the third location information to the location prediction system, and the user can input the third location information on the car and send it to the location prediction system by the car; the mobile device can obtain the third location information under the authorization of the user, and send the third location information to the location prediction system, and the user can input the third location information on the mobile device and send it to the location prediction system by the mobile device.

[0158] Exemplarily, the third location information is information representing the location of a place, and the third location information can be coordinate position, which is used to represent the latitude and longitude information of the target object, and the third location information can be multiple, representing multiple coordinate positions at different historical moments.

[0159] In the application process, the target feature is obtained by constructing the feature according to the third location information, and the target feature is a vector feature, which can be the output data of the model. According to the above embodiments, the location prediction model is pre-trained in the embodiment of the application, and the parameters of the location prediction model are adjusted in the training process, so that the target feature is input into the pre-set location prediction model in the embodiment of the application, and the location prediction result can be obtained. It should be noted that the location prediction result is the location prediction of the model output, and the location prediction result contains the location information predicted by the model.

[0160] In the embodiment of the application, a supervised learning algorithm is constructed to predict the possible location of the customer at the next time point. Specifically, the position prediction is converted into a predictable way by constructing a target variable, which is helpful for the learning of the model, and the data is changed into a supervised learning way, and finally whether to appear in the historical position is predicted, so as to improve the accuracy of the position prediction.

[0161] It should be noted that the position prediction is different from the conventional supervised learning, and the position at a specified time needs to be predicted, and the conventional supervised learning generally predicts whether (binary classification) and predicts the amount of data (regression). The system provided in the embodiment of the present application is to construct samples and target variables, change the data into the conventional supervised learning mode, finally predict whether to appear in the historical position, that is, a binary classification problem, and derive features from the existing little information to support model training. The embodiment of the present application can add some rules in the statistical method to the model training by processing into features, so that the information obtained by the statistical rule can be covered in the algorithm, and the final result will be better than the statistical method.

[0162] The specific implementation of the position prediction system is basically the same as the specific embodiment of the position prediction method described above, and will not be repeated here. The position prediction system can also be provided with other functional modules to realize the position prediction method in the above embodiment under the premise of meeting the requirements of the embodiment of the present application.

[0163] The embodiment of the present application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the position prediction method described above when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0164] Please refer to Fig. 9 , Fig. 9 The hardware structure of the electronic device of another embodiment is illustrated, which includes:

[0165] The processor 901 can be implemented in the form of a general-purpose CPU (Central Processing Unit, central processor), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the embodiments of the present application;

[0166] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 902 can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are saved in the memory 902 and called and executed by the processor 901 to implement the position prediction method of the embodiment of the present application;

[0167] The input / output interface 903 is used to realize information input and output;

[0168] The communication interface 904 is configured to realize the communication interaction between the device and other devices, and can realize the communication through a wired manner (for example, a USB, a network cable, or the like) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, or the like).

[0169] The bus 905 is configured to transmit information between various components (for example, the processor 901, the memory 902, the input / output interface 903, and the communication interface 904) of the device.

[0170] The processor 901, the memory 902, the input / output interface 903, and the communication interface 904 are connected to each other through the bus 905 to realize the communication connection between the device.

[0171] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to realize the position prediction method.

[0172] The memory is a non-transitory computer readable storage medium, and can be used to store a non-transitory software program and a non-transitory computer executable program. In addition, the memory can include a high-speed random access memory, and can further include a non-transitory memory, for example, at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and the remote memory can be connected to the processor through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0173] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0174] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and can include more or fewer steps than the figures, or combine certain steps, or different steps.

[0175] The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separate, that is, can be located in one place, or can be distributed on multiple network units. According to actual needs, part or all of the modules can be selected to realize the purpose of the embodiments of the present application.

[0176] Those skilled in the art can understand that all or some of the steps in the method disclosed above, the function modules / units in the system and the device can be implemented as software, firmware, hardware or appropriate combination thereof.

[0177] The terms "first", "second", "third", "fourth" etc. (if any) in the description of the application and in the claims that follow are used for distinguishing between similar elements and not necessarily for describing a sequential or chronological order. It is to be understood that the use of these terms herein is to be construed to cover the embodiments of the application whether or not the embodiments are described using the same term. Furthermore, the terms "comprise", "comprising", "include", "including", and "has", "having" and variants thereof are to be construed in a non-exclusive manner when used in this description and in the claims that follow. For example, when used in the context of a process, method, system, product or apparatus, the term "comprising" means that the process, method, system, product or apparatus includes the recited steps or units, but can also include additional steps or units not specifically recited.

[0178] It should be understood that, in the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the relationship between associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that there are three cases: only A, only B, and A and B at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be singular or plural.

[0179] In several embodiments provided by the present application, it should be understood that the disclosed system and method can be implemented in other ways. For example, the above-described system embodiments are only illustrative, for example, the division of the above-mentioned units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed objects can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0180] The units described as separate components above can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0181] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0182] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application, essentially or the part that contributes to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.

[0183] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, and the scope of the rights of the embodiments of the present application is not limited thereto. Any modifications, equivalent replacements and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the rights of the embodiments of the present application.

Claims

1. A position prediction method characterized by, The method comprises: obtaining first position information of a sample object at multiple historical time points and second position information of the sample object at a current time point; labeling the first position information that is the same as the second position information to obtain first labeled information and labeling the first position information that is not the same as the second position information to obtain second labeled information; taking the first position information containing the first labeled information as a target variable or taking the first position information containing the second labeled information as the target variable; performing feature construction according to the first position information to obtain sample features, inputting the sample features and the target variable into a preset position prediction model, and adjusting parameters of the position prediction model according to an output result of the position prediction model; obtaining third position information of a target object at multiple historical time points; performing feature construction according to the third position information to obtain target features, and inputting the target features into the position prediction model with adjusted parameters to obtain a position prediction result.

2. The position prediction method according to claim 1, characterized by, The sample object has multiple sample objects; the feature construction according to the first position information to obtain sample features comprises: obtaining a number of positions represented by the first position information of each sample object and a first occurrence number of the sample object; obtaining a first object number of the sample object on each first position information and a second occurrence number of each first position information; performing feature conversion according to multiple ones of the number of positions, the first occurrence number, the first object number and the second occurrence number to obtain sample features.

3. The position prediction method according to claim 1, characterized by, The inputting of the sample features and the target variable into a preset position prediction model, the adjustment of parameters of the position prediction model according to an output result of the position prediction model comprises: splitting the sample features and the target variable into training samples and test samples; inputting the training samples into a preset position prediction model to obtain a first output result, and adjusting parameters of the position prediction model according to the first output result; inputting the training samples into the position prediction model with adjusted parameters to obtain a second output result, inputting the test samples into the position prediction model with adjusted parameters to obtain a third output result; calculating a stability index according to the second output result and the third output result, and determining whether to continue adjusting the parameters of the position prediction model according to the stability index.

4. The position prediction method according to claim 3, characterized by, The second output result comprises a first prediction probability value, and the second output result comprises a second prediction probability value; The calculation of a stability index according to the second output result and the third output result comprises: performing addition calculation according to the first prediction probability value and the second prediction probability value to obtain a first value; performing logarithmic calculation according to the first prediction probability value and the second prediction probability value to obtain a second value; obtaining a stability index according to a product of the first value and the second value.

5. The position prediction method according to claim 3, characterized by, The second output result includes a first prediction probability value, and the second output result includes a second prediction probability value, and the first prediction probability value and the second prediction probability value are both multiple; The stability index is calculated according to the second output result and the third output result, including: According to the size of the first prediction probability value and the second prediction probability value, a plurality of score segments are divided; The first prediction probability value and the second prediction probability value are calculated respectively, and the sub-stability index under the corresponding score segment is calculated; The sub-stability indexes under each score segment are accumulated to obtain a stability index.

6. The position prediction method according to claim 1, characterized by, The target feature is constructed according to the third position information, and the target feature is input into the position prediction model after adjusting the parameters to obtain a position prediction result, including: The target feature is constructed according to a plurality of third position information respectively, and a plurality of target features are obtained; A plurality of target prediction positions and corresponding target prediction probability values are obtained by inputting a plurality of target features into the position prediction model after adjusting the parameters respectively; According to the size of each target prediction probability value, a position prediction result is determined in a plurality of prediction positions.

7. A position prediction system characterized by, The system includes: A sample object acquisition module for acquiring first position information of a sample object at a plurality of historical time points and second position information of the sample object at a current time point; A target variable acquisition module for marking the first position information that is the same as the second position information to obtain first marking information, and marking the first position information that is different from the second position information to obtain second marking information; taking the first position information containing the first marking information as a target variable, or taking the first position information containing the second marking information as the target variable; A parameter adjustment module for constructing a sample feature according to the first position information, inputting the sample feature and the target variable into a preset position prediction model, and adjusting the parameters of the position prediction model according to the output result of the position prediction model; A target object acquisition module for acquiring third position information of a target object at a plurality of historical time points; A position prediction module for constructing a target feature according to the third position information, and inputting the target feature into the position prediction model after adjusting the parameters to obtain a position prediction result.

8. An electronic device, comprising: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the position prediction method of any one of claims 1-6 when executing the computer program.

9. A computer readable storage medium, the storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the position prediction method of any one of claims 1-6.

Images