CN115759419A

CN115759419A - Position prediction method, system, electronic device and storage medium

Info

Publication number: CN115759419A
Application number: CN202211460612.XA
Authority: CN
Inventors: 黄移军
Original assignee: Ping An Life Insurance Company of China Ltd
Current assignee: Ping An Life Insurance Company of China Ltd
Priority date: 2022-11-17
Filing date: 2022-11-17
Publication date: 2023-03-07

Abstract

The embodiment of the application provides a position prediction method, a position prediction system, electronic equipment and a 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 the current moment; marking according to the corresponding relation between the first position information and the second position information to obtain marking information, and determining a target variable according to the marking information; 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; acquiring third position information of the target object at a plurality of historical moments; and performing feature construction according to the third position information to obtain target features, and inputting the target features into the position prediction model after the parameters are adjusted to obtain a position prediction result, so that the accuracy of position prediction can be improved.

Description

Position prediction method, system, electronic device and storage medium

Technical Field

The present application relates to the field of artificial intelligence technologies, and in particular, to a method, a system, an electronic device, and a storage medium for location prediction.

Background

The position of the space-time location where the target appears is predicted, so that the application value is high, the data processing capacity can be improved, and more intelligent service can be provided for users.

In the related art, the position of the target is usually predicted by a statistical method, for example, by counting a place with a relatively high frequency of occurrence at a historical time point as a spatio-temporal geographic position of a corresponding time point, but this method cannot predict various possible results, and thus the accuracy of position prediction is low.

Disclosure of Invention

The embodiment of the present application mainly aims to provide a position prediction method, a position prediction system, an electronic device, and a storage medium, which can improve the accuracy of position prediction.

In order to achieve the above object, a first aspect of an embodiment of the present application provides a location prediction method, where the method includes: acquiring first position information of a sample object at a plurality of historical moments and second position information of the sample object at the current moment; marking according to the corresponding relation between the first position information and the second position information to obtain marking information, and determining a target variable according to the marking information; performing feature construction according to the first position information to obtain a sample feature, inputting the sample feature 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; acquiring third position information of the target object at a plurality of historical moments; and performing feature construction according to the third position information to obtain target features, and inputting the target features into the position prediction model after parameters are adjusted to obtain a position prediction result.

In some embodiments, the labeling according to the correspondence between the first location information and the second location information to obtain labeled information, and determining a target variable according to the labeled information includes: marking the first position information which is the same as the second position information to obtain first marking information; marking the first position information different from the second position information to obtain second marking information; the first position information including the first marker information is taken as a target variable, or the first position information including the second marker information is taken as the target variable.

In some embodiments, there are a plurality of sample objects, and the performing feature construction according to the first position information to obtain a sample feature includes: acquiring the number of positions represented by the first position information of each sample object and the first occurrence number of the sample object; acquiring the number of first objects of the sample object on each piece of first position information and the second occurrence frequency of each piece of first position information; and performing feature conversion according to the position number, the first occurrence number, the first object number and the second occurrence number to obtain sample features.

In some embodiments, the inputting the sample characteristics and the target variables into a preset position prediction model, and adjusting parameters of the position prediction model according to an output result of the position prediction model includes: splitting the sample features and the target variables into training samples and test samples; inputting the training sample 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 sample into the position prediction model after parameter adjustment to obtain a second output result, and inputting the test sample into the position prediction model after parameter adjustment to obtain a third output result; and calculating to obtain a stability index according to the second output result and the third output result, and determining whether to continuously adjust the parameters of the position prediction model according to the stability index.

In some embodiments, the second output result comprises a first predicted probability value, the second output result comprises a second predicted probability value; the calculating to obtain the stability index according to the second output result and the third output result includes: adding the first prediction probability value and the second prediction probability value to obtain a first numerical value; carrying out logarithmic calculation according to the first prediction probability value and the second prediction probability value to obtain a second numerical value; and obtaining a stability index according to the product of the first numerical value and the second numerical value.

In some embodiments, the second output result comprises a first predicted probability value, the second output result comprises a second predicted probability value, the first predicted probability value and the second predicted probability value are each in plurality; the calculating to obtain the stability index according to the second output result and the third output result includes: dividing a plurality of score segments according to the first prediction probability value and the second prediction probability value; respectively calculating the first prediction probability value and the second prediction probability value, and sub-stability indexes under the corresponding score segments; and accumulating according to the sub-stability indexes under each fractional segment to obtain a stability index.

In some embodiments, the performing feature construction according to the third location information to obtain a target feature, and inputting the target feature into the location prediction model after adjusting parameters to obtain a location prediction result, includes: respectively carrying out feature construction according to the third position information to obtain a plurality of target features; respectively inputting the target characteristics into the position prediction model after the parameters are adjusted to obtain a plurality of target prediction positions and corresponding target prediction probability values; and determining a position prediction result in the plurality of prediction positions according to the magnitude of each target prediction probability value.

To achieve the above object, a second aspect of an embodiment of the present application proposes a position prediction system, which includes: the device comprises a sample object acquisition module, a storage module and a processing module, wherein the sample object acquisition module is used for acquiring first position information of a sample object at a plurality of historical moments and second position information of the sample object at the current moment; the target variable acquisition module is used for marking according to the corresponding relation between the first position information and the second position information to obtain marking information, and determining a target variable according to the marking information; the parameter adjusting module is used for carrying out 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; the target object acquisition module is used for acquiring third position information of the target object at a plurality of historical moments; and the position prediction module is used for carrying out feature construction according to the third position information to obtain target features, and inputting the target features into the position prediction model after parameters are adjusted to obtain a position prediction result.

In order to achieve the above object, a third aspect of an embodiment of the present application provides an electronic device, where the electronic device includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the method described in the foregoing first aspect of the embodiment.

In order to achieve the above object, a fourth aspect of the embodiments of the present application proposes a storage medium, which is a computer-readable storage medium, and the storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the method described in the foregoing first aspect of the embodiments.

The position prediction method, the position prediction system, the electronic equipment and the storage medium can be applied to the position prediction system. By executing the position prediction method, training can be performed according to first position information of a sample object at a historical time and second position information of the sample object at a current time, wherein the embodiment of the application marks a corresponding relationship between the two kinds of position information to obtain marked information, and the marked information is used as a target variable in a process of training the position prediction model, so that characteristic construction can be performed according to the first position information to obtain sample characteristics, the sample characteristics and the target variable are input into a preset position prediction model, parameters of the position prediction model are adjusted through training, and then characteristic construction can be performed according to third position information at the historical time in an application process and then input into the trained position prediction model to obtain a position prediction result of the target object. In the embodiment of the application, the position prediction is converted into a predictable mode by constructing the target variable, so that the learning of a model is facilitated, data is converted into a supervised learning mode, whether the position occurs at a historical position is predicted finally, and the accuracy of the position prediction is improved.

Drawings

Fig. 1 is a flowchart of a location prediction method provided in an embodiment of the present application;

fig. 2 is a flowchart of step S102 in fig. 1;

FIG. 3 is a flowchart of step S103 in FIG. 1;

fig. 4 is a flowchart of step S103 in fig. 1;

fig. 5 is a flowchart of step S404 in fig. 4;

fig. 6 is a flowchart of step S404 in fig. 4;

fig. 7 is a flowchart of step S105 in fig. 1;

FIG. 8 is a functional block diagram of a location prediction system provided in an embodiment of the present application;

fig. 9 is a schematic hardware structure diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

It is noted that while functional block divisions are provided in device diagrams and logical sequences are shown in flowcharts, in some cases, steps shown or described may be performed in sequences other than block divisions within devices or flowcharts. The terms first, second and the like in the description and in the claims, and the drawings described above, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

Unless defined otherwise, 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 terminology used herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the application.

First, several terms referred to in the present application are resolved:

artificial Intelligence (AI): is a new technical science for researching and developing theories, methods, technologies and application systems for simulating, extending and expanding human intelligence; artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produces a new intelligent machine that can react in a manner similar to human intelligence, and research in this field includes robotics, language recognition, image recognition, natural language processing, and expert systems, among others. The artificial intelligence can simulate the information process of human consciousness and thinking. Artificial intelligence is also a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results.

On the premise that the target provides the location service, predicting the position of the space-time location where the target appears has more application value, for example, traffic planning can be performed according to the location information provided by the target, or intelligent pushing of service items can be performed according to the location information provided by the target, so that the data processing capacity can be improved and more intelligent service can be provided for the user by predicting the position of the target.

In the related art, a statistical method is usually adopted to predict the position of a target, for example, a place with relatively high occurrence frequency of historical time points is counted as a space-time geographic position of a corresponding time point, however, the applicant finds that the method cannot predict various possible results, can only judge according to the occurrence frequency of a single historical time point, and cannot predict other positions with not the most occurrence frequency; alternatively, in the related art, the position is predicted by a time-series method, but each time point of each object is not necessarily continuous, and thus, many processes are required, and the effect is not necessarily good, and thus the accuracy of the position prediction is low.

Based on this, the present application provides a position prediction method, a system, an electronic device, and a storage medium, where by executing the position prediction method, a sample object may be trained according to first position information of the sample object at a historical time and second position information of the sample object at a current time, where in the present application, a correspondence between the two kinds of position information is labeled to obtain labeled information, and the labeled information is used as a target variable in a process of training the position prediction model, so that a feature configuration may be performed according to the first position information to obtain a sample feature, the sample feature and the target variable are input into a preset position prediction model, a parameter of the position prediction model is adjusted through training, and then, according to third position information at the historical time, the feature configuration may be performed in an application process and then input into the trained position prediction model to obtain a position prediction result of the target object. In the embodiment of the application, the position prediction is converted into a predictable mode by constructing the target variable, so that the learning of a model is facilitated, data is converted into a supervised learning mode, whether the position occurs in a historical position is predicted finally, and the accuracy of the position prediction is improved.

The position prediction method, the system, the electronic device, and the storage medium provided in the embodiments of the present application are specifically described in the following embodiments, and first, the position prediction method in the embodiments of the present application is described.

The position prediction method in the embodiment of the present application can be explained by the following embodiment.

The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.

The artificial intelligence base technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

The embodiment of the application provides a position prediction method, and relates to the technical field of artificial intelligence. The position prediction method provided by the embodiment of the application can be applied to a terminal, a server side and software running in the terminal or the server side. In some embodiments, the terminal may be a smartphone, tablet, laptop, desktop computer, or the like; the server side can be configured into an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, and cloud servers for providing basic cloud computing services such as cloud service, a cloud database, cloud computing, cloud functions, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN (content delivery network) and big data and artificial intelligence platforms; the software may be an application or the like implementing the position prediction method, but is not limited to the above form.

The application is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

It should be noted that, in various embodiments of the present application, when data related to the identity or characteristic of a user, such as user information, user behavior data, user history data, and user location information, is processed in a related manner, permission or consent of the user is obtained first, for example, when data stored by the user and a cache data access request of the user are obtained, permission or consent of the user is obtained first. Moreover, the collection, use, and processing of such data, etc., will comply with relevant laws and regulations and standards in the relevant countries and regions. In addition, when the embodiment of the present application needs to acquire sensitive personal information of a user, individual permission or individual consent of the user is obtained through a pop-up window or a jump to a confirmation page, and after the individual permission or individual consent of the user is definitely obtained, necessary user-related data for enabling the embodiment of the present application to operate normally is acquired.

Fig. 1 is an alternative flowchart of a location prediction method provided in an embodiment of the present application, and the method in fig. 1 may include, but is not limited to, steps S101 to S105.

Step S101, acquiring first position information of a sample object at a plurality of historical moments and second position information of the sample object at the current moment;

for example, the location prediction method in the embodiment of the present application may be applied to a location prediction system, where the location prediction system may be a terminal device, for example, a smart phone, a personal computer, or a server, and the location prediction system may also be a system composed of multiple devices, which is not limited in this respect.

In the embodiment of the application, first position information of the sample object at a plurality of historical moments can be acquired, second position information of the sample object at the current moment can be acquired, and the position information is used as a sample so as to perform model training in the following process.

Illustratively, the sample object is the subject of a training process, and the sample object may be a user, a vehicle, a mobile device, or the like.

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 operation process of the position prediction system; the vehicle may be an automobile, a motorcycle, an electric bicycle, or the like, for example, the automobile may record the first location information and obtain the second location information under the authorization of the user, and send the first location information and the second location information to the location prediction system, and the user may also input the first location information and the second location information on the automobile and send the first location information and the second location information to the location prediction system by the automobile; the mobile device may record the first location information and obtain the second location information under the authorization of the user, and send the first location information and the second location information to the location prediction system, or the user may input the first location information and the second location information on the mobile device, and the mobile device sends the first location information and the second location information to the location prediction system.

For example, the first location information is information characterizing a location, the first location information may be a coordinate location, which represents latitude and longitude information of the sample object, and the first location information may be multiple, which characterizes multiple coordinate locations at different historical times; the second position information is information representing a position of the location, and the second position information may be a coordinate position representing latitude and longitude information of the sample object, and one of the second position information represents a coordinate position at the current time. It is understood that the first location information and the second location information are only location information at different times.

Step S102, marking is carried out according to the corresponding relation between the first position information and the second position information to obtain marking information, and a target variable is determined according to the marking information;

for example, in the embodiment of the present application, the marking may be performed according to a corresponding relationship between the first location information and the second location information, so as to obtain marked marking information, and the target variable may be determined according to the marking information. It will be appreciated that the target variables summarize the information that it is desired to predict the result from the perspective of the algorithm used to construct the position prediction model, and thus the target variables are the results predicted by the machine learning algorithm.

For example, in the embodiment of the present application, the marking may be performed according to a similarity relationship between the first location information and the second location information, and corresponding marking information is performed according to whether the first location information and the second location information are similar, where the marking information may be a field for indicating the similarity relationship between the first location information and the second location information. In some embodiments, in the embodiments of the present application, the first position information and the second position information are labeled as 1, and different ones are labeled as 0, so as to obtain a target sample, and perform feature conversion based on the target sample, so as to obtain a target variable, or the first position information after labeling the information may be directly used as the target variable, which is not limited herein.

Step S103, carrying out 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;

for example, in the embodiment of the present application, a position prediction model is preset, and the feature obtained by converting the position information is input into the position prediction model, so that a corresponding position prediction result can be obtained, or the position information can be directly input into the position prediction model, and the feature of the position information can be constructed in the position prediction model, thereby implementing the processing of data.

In the training process, the embodiment of the present application performs feature construction according to the first position information to obtain a sample feature, where the sample feature is a vector feature and can be used to make output data of a model, and it can be known from the above embodiment that a target variable is constructed in advance in the embodiment of the present application, so that the embodiment of the present application inputs the sample feature and the target variable into a preset position prediction model, and adjusts a parameter of the position prediction model according to an output result of the position prediction model.

It can be understood that the purpose of the label is to construct a target variable of the model, and when the model is trained, the target variable needs to be present to learn which features, the position of the sample at the next test point is the same as the previous situation, i.e. the part with the label information of 1, and if the position is 1, the position of the sample at the next specified time point can be indirectly predicted.

Step S104, acquiring third position information of the target object at a plurality of historical moments;

for example, after the training of the position prediction model is completed in the above steps, the embodiment of the application may apply the position prediction model, and in the applying process, third position information of the target object at a plurality of historical time points may be obtained.

Illustratively, the target object is an object of an application process, which may be a user, a vehicle, a mobile device, and the like.

The third location information of the user may be obtained by the location prediction system after authorization of the user, or the user may also input the third location information in the location prediction system, or the third location information may be data generated in the operation process of the location prediction system; the vehicle may be an automobile, a motorcycle, an electric bicycle, or the like, for example, the automobile may 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 may also input the third location information on the automobile and send the third location information to the location prediction system by the automobile; the mobile device may obtain the third location information under the authorization of the user, and send the third location information to the location prediction system, or the user may input the third location information on the mobile device, and the third location information is sent to the location prediction system by the mobile device.

For example, the third location information is information representing a location position, the third location information may be a coordinate position representing latitude and longitude information of the target object, and the third location information may be multiple, and represents multiple coordinate positions at different historical times.

And step S105, performing feature construction according to the third position information to obtain target features, and inputting the target features into the position prediction model after the parameters are adjusted to obtain a position prediction result.

In an application process, the embodiment of the present application performs feature construction according to third location information to obtain a target feature, where the target feature is a vector feature and can be used to make output data of a model, and it can be known from the above embodiment that a location prediction model is trained in advance in the embodiment of the present application, and parameters of the location prediction model are adjusted in the training process, so that the embodiment of the present application inputs the target feature into the preset location prediction model to obtain a location prediction result, where it is to be noted that the location prediction result is a location prediction condition output by the model, and the location prediction result includes location information predicted by the model.

In the embodiment of the application, the positions of the clients which are possibly appeared at the next time point are predicted by constructing a supervised learning algorithm. Specifically, the position prediction is converted into a predictable mode by constructing a target variable, so that the model learning is facilitated, the data is converted into a supervised learning mode, whether the position occurs in the historical position is predicted finally, and the accuracy of the position prediction is improved.

It should be noted that the position prediction is different from the conventional supervised learning, and the prediction is the position occurring in a certain time, and the conventional supervised learning predicts whether (binary classification) and the prediction data amount (regression) in general. The method provided by the embodiment of the application is to construct a sample and a target variable, change data into a common supervised learning mode, finally predict whether the sample and the target variable appear at a historical position, namely, the problem of two-classification, derive characteristics from the existing little information and support model training. According to the embodiment of the application, some rules in the statistical method can be processed into features to be added into model training, so that information obtained by the statistical rules can be covered in an algorithm, and the final result is superior to that of the statistical method.

Referring to fig. 2, in some embodiments, step S102 may include steps S201 to S203:

step S201, marking first position information which is the same as the second position information to obtain first marking information;

step S202, marking first position information different from the second position information to obtain second marking information;

in step S203, the first position information including the first marker information is set as a target variable, or the first position information including the second marker information is set as a target variable.

For example, in the marking process, the marking information is determined according to whether the first position information is the same as the second position information. Specifically, in the embodiment of the present application, first position information that is the same as second position information is marked to obtain first marked information, first position information that is different from the second position information is marked to obtain second marked information, and finally, the first position information including the first marked information is used as a target variable, or the first position information including the second marked information is used as a target variable.

Illustratively, the first flag information and the second flag information are an added field, for example, a flag field added to the location information, and the first flag information may be 1, and when the first location information carries a field with a flag of 1, it indicates that the first location information is the same as the second location information; the second flag information may be 0, indicating that the first position information is not identical to the second position information when the first position information has a field of flag 0.

For example, in the embodiment of the present application, second position information of a plurality of sample objects may be obtained from historical data, as shown in table 1:

serial number	Sample object (id)	Current time of day	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

TABLE 1

Then, the first position information of the sample object at the historical time is obtained, and corresponding mark information is given, 3 geographical positions of the sample object c001 corresponding to the sequence number 1 before the corresponding time (for example, 3 months), 5 geographical positions of the sample object c 2 before the corresponding time (for example, 3 months in a unified manner), and 5 geographical positions of the sample object c 3 before 4 months before the corresponding time are obtained, so that a new data set is obtained as shown in table 2:

TABLE 2

It should be noted that, the first position information with the first label information and/or the second label information may be used as a target sample, and a target variable may be obtained according to the target sample, for example, the target sample may be directly used as the target variable, or alternatively, the target sample may be subjected to feature construction, so as to be converted into a vector feature, so as to obtain the target variable.

Referring to fig. 3, in some embodiments, the number of sample objects is multiple, and step S103 may include steps S301 to S303:

step S301, acquiring the number of positions represented by the first position information of each sample object and the first occurrence frequency of the sample object;

step S302, acquiring the number of first objects of the sample object on each first position information and the second occurrence frequency of each first position information;

step S303, performing feature conversion according to the position number, the first occurrence number, the first object number and the second occurrence number to obtain sample features.

For example, in the feature construction process in the embodiment of the present application, multi-dimensional construction may be performed in combination with specific situations according to the sample object and the first position information. Specifically, in the embodiment of the present application, feature construction may be performed from the dimension of the sample, including obtaining 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, feature construction can be performed from the dimension of the position, and the feature construction comprises the steps of obtaining the first object number of the sample object on each first position information and the second occurrence number of each first position information.

Finally, in embodiments of the present application, sample features may be constructed from dimensions of the sample, dimensions of the location, or dimensions of the sample plus the location. When feature construction is carried out according to the dimension of the sample, feature conversion can be carried out according to the position number and the first occurrence number to obtain the sample feature; when feature construction is carried out according to the dimension of the position, feature conversion can be carried out according to the number of the first objects and the second occurrence frequency to obtain sample features; when feature construction is performed according to the dimension of the sample plus the position, feature conversion can be performed according to the number of positions, the first occurrence number, the first object number and the second occurrence number to obtain a sample feature.

In an embodiment of the present application, feature conversion is performed on the corresponding first position information according to a plurality of the position number, the first occurrence number, the first object number, and the second occurrence number, so as to obtain a sample feature.

For example, the number of positions may include the number of positions of the sample object in the history, the number of positions of the sample object in the weekday/holiday history, the number of positions of the sample object in the monday to weekday history, the number of positions of the sample object in different periods of the history, and the like.

For example, the first occurrence number may include the number of occurrences of the sample object in the history, the number of occurrences of the sample object in the weekday/holiday history, the number of occurrences of the sample object in different periods of the history, and the like.

For example, the first number of objects may include how many sample objects have been accessed in the history for each location, the number of sample objects that appear on the history workday/holiday, the occurrence of different days/periods, etc.

For example, the second number of occurrences may include the number of occurrences of each location in history, the number of occurrences of history workday \ holiday, the occurrence of different week \ period, etc.

It should be noted that in the embodiment of the present application, feature association may be performed on the sample feature and the target variable obtained according to the above steps, in combination with the id of the sample object, to obtain a wide table, where the wide table includes three information: id. And the sample characteristics, the target variable and the data of the wide table are used for inputting into a position prediction model for training.

Referring to fig. 4, in some embodiments, step S103 may include steps S401 to S404:

step S401, splitting sample characteristics and target variables into training samples and test samples;

step S402, inputting a training sample 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;

step S403, inputting the training sample into the position prediction model after parameter adjustment to obtain a second output result, and inputting the test sample into the position prediction model after parameter adjustment to obtain a third output result;

and S404, calculating to obtain a stability index according to the second output result and the third output result, and determining whether to continuously adjust the parameters of the position prediction model according to the stability index.

For example, the training of the model may be performed after the training data is established. In the embodiment of the present application, the training is to split data according to a certain proportion, including splitting sample characteristics and a target variable into a training set and a testing set, where the samples in the training set are training samples, the samples in the testing set are testing samples, and 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, and in order to further improve the stability of the model, in the embodiment of the present application, the training samples are input into the position prediction model after parameter adjustment to obtain a second output result, the testing samples are input into the position prediction model after parameter adjustment to obtain a third output result, and a stability index is calculated according to the second output result and the third output result, and whether to continue to adjust parameters of the position prediction model is determined according to the stability index.

It should be noted that the Stability Index is a Population Stability Index (PSI), by which the Stability of the model can be quantified.

In an embodiment, in the embodiment of the present application, the sample characteristics and the target variables are split according to a certain ratio, for example, the training set and the test set are obtained by splitting according to a ratio of 7 to 3, and on the premise of meeting the requirements of the embodiment of the present application, the splitting may be performed according to other ratios.

In an embodiment, the algorithm of the embodiment of the present application may select a common machine learning algorithm (e.g., randomForest, GBDT, lightGBM, xgboost, etc.), obtain an optimal result by adjusting parameters (parameters of the algorithm, such as a loss function, a depth of a tree, a learning rate, etc.), and finally obtain a final model by testing stability of the model.

It should be noted that the output result of the position prediction model includes a plurality of 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.

Referring to fig. 5, in some embodiments, step S404 may include steps S501 to S503:

step S501, adding and calculating according to the first prediction probability value and the second prediction probability value to obtain a first value;

step S502, carrying out logarithmic calculation according to the first prediction probability value and the second prediction probability value to obtain a second numerical value;

step S503, obtaining the stability index according to the product of the first numerical value and the second numerical value.

In the above embodiments, it can be understood that the first predicted probability value is an actual occupancy and the second predicted probability value is an expected occupancy.

And adding the first prediction probability value and the second prediction probability value to obtain a first value, wherein the first value is as follows:

first value = sum ((first predicted probability value-second predicted probability value);

performing logarithmic calculation according to the first predicted probability value and the second predicted probability value to obtain a second numerical value, wherein the first numerical value is as follows:

second value = ln (first predicted probability value/second predicted probability value);

finally, the stability indicator is obtained according to the product of the first value and the second value, and therefore the formula for obtaining the stability indicator is:

psi = sum ((first predicted probability value-second predicted probability value) × ln (first predicted probability value/second predicted probability value)).

In some embodiments, the second output result comprises a first predicted probability value, the second output result comprises a second predicted probability value, and the first predicted probability value and the second predicted probability value are each plural.

Referring to fig. 6, in some embodiments, step S404 may include steps S601 to S603:

step S601, dividing a plurality of score segments according to the first prediction probability value and the second prediction probability value;

step S602, respectively calculating a first prediction probability value and a second prediction probability value, and sub-stability indexes under corresponding score segments;

and step S603, accumulating according to the sub-stability indexes under each fractional segment to obtain the stability index.

For example, in the embodiment of the present application, the first prediction probability value and the second prediction probability value are divided into a plurality of score segments, and a final total stability index is calculated according to the sub-stability indexes under different score segments, where the calculation of the sub-stability indexes may refer to the calculation processes in step S501 to step S503, and details are not repeated here.

For example, in the embodiment of the present application, a value obtained by multiplying the probability value predicted by the position prediction model by 100 is divided into 10 segments, so as to be counted with the list, and the probability value may not be multiplied by 100 on the premise of meeting the requirements of the embodiment of the present application. Subsequently, the embodiment of the present application distributes the first prediction probability values under the corresponding fractional segments, and it can be understood that the sum of the first prediction probability values under the respective fractional segments is equal to 1. Similarly, the embodiment of the present application further distributes the second prediction probability values under the corresponding fractional segments, and similarly, the sum of the second prediction probability values under each fractional segment is equal to 1. So far, the comparison condition of the experimental sample and the prediction sample under different prediction probability values can be known in the embodiment of the application.

After the sub-stability indexes under each fractional segment are obtained through calculation, the stability indexes are obtained through accumulation according to the sub-stability indexes under each fractional segment in the embodiment of the application. It should be noted that each sub-stability index may be directly accumulated to obtain a stability index, and each sub-stability index may also be multiplied by a corresponding coefficient according to the difference of the fraction sections to perform weighting to obtain a stability index.

In the embodiment of the present application, the first prediction probability value is described as a%, the second prediction probability value is described as B%, and once you can obtain a table of sub-stability indexes under each score segment, as shown in table 3:

TABLE 3

And accumulating the sub-stability indexes of the last column corresponding to each fractional segment in the table 3 to obtain a final stability index.

When judging whether the model meets the requirements or not according to the stability index, determining whether to continuously adjust the parameters of the position prediction model according to the size of the stability index in the embodiment of the application. For example, when the stability index is less than 0.1, it indicates that the model is more stable on the test sample; when the stability index is between 0.1 and 0.25, the stability of the model on the test sample is general and can be combined with the practical situation to be available or not; and when the stability index is larger than 0.25, the model is unstable on the test sample and needs to be retrained, and the parameters of the position prediction model are continuously adjusted.

It will be appreciated that the stability described above can be combined with the design of test samples for several different time windows. The time window is the corresponding time of the sample access, if the training sample access time window is 2022 years 1 to 6 months, the test sample access time window is 2022 years 7 months.

Referring to fig. 7, in some embodiments, step S105 may include steps S701 to S703:

step S701, respectively carrying out feature construction according to a plurality of third position information to obtain a plurality of target features;

step S702, respectively inputting a plurality of target characteristics into the position prediction model after parameter adjustment to obtain a plurality of target prediction positions and corresponding target prediction probability values;

step S703, determining a position prediction result in the plurality of predicted positions according to the magnitude of each target prediction probability value.

For example, when the position is predicted according to the position prediction model after the parameters are adjusted, different prediction results may be obtained according to different third position information, and then the final position prediction result is determined according to the prediction probability values of the results. Specifically, in the embodiment of the present application, feature construction may be performed according to a plurality of third location information, so as to obtain a plurality of target features, a construction process of the target features is similar to the above sample feature construction process, and is not described herein again, and then the plurality of target features are input into the location prediction model after the parameters are adjusted, so as to obtain a plurality of target prediction locations and corresponding target prediction probability values, it may be understood that each target prediction location has a corresponding target prediction probability value, and finally, a location prediction result is determined in the plurality of prediction locations according to the size of each target prediction probability value.

It can be understood that the target predicted position is a third target position corresponding to the target predicted probability value output by the position prediction model, and the position prediction model in the embodiment of the present application functions to predict and calculate the predicted probability of the input position information, and select one position information with the highest probability as a final position prediction result, that is, the corresponding geographic position with the highest probability value is used as the predicted geographic position of the target object at the time point.

For example, when there are a plurality of target objects, each of which has a different id, the third location information corresponding to each target object at the historical time is shown in table four:

serial number	Target object (id)	Current time of day	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

Watch four

Inputting the target characteristics obtained according to the third position information into the position prediction model after the parameters are adjusted to obtain different target prediction positions and corresponding target prediction probability values, as shown in table five:

serial number	Target object (id)	Current time of day	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

Watch five

Finally, in the embodiment of the present application, the corresponding location information with the maximum target prediction probability value in table five is selected as the location prediction result, and the final location prediction result is shown in table six:

serial number	Target object (id)	The current time	Location prediction results	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

Watch six

Referring to fig. 8, an embodiment of the present application further provides a position prediction system, which can implement the position prediction method, where the position prediction system includes:

a sample object obtaining module 801, configured to obtain first position information of a sample object at a plurality of historical time instants and second position information of the sample object at a current time instant;

a target variable obtaining module 802, configured to mark according to a correspondence between the first location information and the second location information to obtain mark information, and determine a target variable according to the mark information;

a parameter adjusting module 803, configured to perform feature construction according to the first location information to obtain a sample feature, input the sample feature and the target variable into a preset location prediction model, and adjust a parameter of the location prediction model according to an output result of the location prediction model;

a target object obtaining module 804, configured to obtain third location information of the target object at multiple historical moments;

and 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 parameter adjustment to obtain a position prediction result.

For example, the location prediction system in the embodiment of the present application may perform the location prediction method in the above embodiment, where the location prediction system may be a terminal device, for example, a smart phone, a personal computer, or a server, or a system composed of multiple devices, and is not limited herein.

In the process of executing the position prediction method, the position prediction system in the embodiment of the application may acquire first position information of the sample object at a plurality of historical times, acquire second position information of the sample object at the current time, and use the position information as a sample, so as to perform model training later.

The first position information and the second position information of the user can be obtained by the position prediction system after being authorized by 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 operation process of the position prediction system; the vehicle may be an automobile, a motorcycle, an electric bicycle, or the like, for example, the automobile may 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, and the user may also input the first position information and the second position information on the automobile and send the first position information and the second position information to the position prediction system by the automobile; the mobile device may record the first location information and obtain the second location information under the authorization of the user, and send the first location information and the second location information to the location prediction system, or the user may input the first location information and the second location information on the mobile device, and send the first location information and the second location information to the location prediction system by the mobile device.

For example, the first location information is information characterizing a location, the first location information may be a coordinate location, which represents latitude and longitude information of the sample object, and the first location information may be multiple, which characterizes multiple coordinate locations at different historical times; the second position information is information representing a position of a place, and the second position information may be a coordinate position representing latitude and longitude information of the sample object. It is to be understood that the first location information and the second location information are only location information at different times.

For example, in the embodiment of the present application, the marking may be performed according to a corresponding relationship between the first location information and the second location information, so as to obtain marked marking information, and the target variable may be determined according to the marking information. It will be appreciated that the target variables summarize information that it is desirable to predict the result from the perspective of the algorithm used to construct the location prediction model, and thus the target variables are the results predicted by the machine learning algorithm.

For example, in the embodiment of the present application, the marking may be performed according to a similarity relationship between the first location information and the second location information, and the corresponding marking information is made according to whether the first location information and the second location information are similar, where the marking information may be a field for indicating the similarity relationship between the first location information and the second location information. In some embodiments, in the embodiments of the present application, the first position information and the second position information are labeled as 1, and the different position information is labeled as 0, so as to obtain a target sample, and perform feature conversion based on the target sample, so as to obtain a target variable, or the first position information after the labeling information is labeled may be directly used as the target variable, which is not limited herein.

In the training process, feature construction is performed according to the first position information to obtain a sample feature, the sample feature is a vector feature and can be used for making output data of the model, and according to the embodiment, a target variable is constructed in advance, so that 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.

For example, after the training of the position prediction model is completed in the above steps, the embodiment of the application may apply the position prediction model, and in the process of applying, may obtain third position information of the target object at multiple historical times.

The third location information of the user may be obtained by the location prediction system after being authorized by the user, or the user may also input the third location information in the location prediction system, or the third location information may be data generated in the operation process of the location prediction system; the vehicle may be an automobile, a motorcycle, an electric bicycle, or the like, for example, the automobile may 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 may also input the third location information on the automobile and send the third location information to the location prediction system by the automobile; the mobile device may obtain the third location information under the authorization of the user, and send the third location information to the location prediction system, or the user may input the third location information on the mobile device, and the third location information is sent to the location prediction system by the mobile device.

For example, the third location information may be a coordinate location, which represents latitude and longitude information of the target object, and may be multiple, which represents multiple coordinate locations at different historical times.

It should be noted that the position prediction is different from the conventional supervised learning, the position to be predicted is the position occurring in a certain time, and the conventional supervised learning predicts whether to predict (binary classification) and the prediction data amount (regression) in general. The system provided by the embodiment of the application constructs samples and target variables, changes data into a common supervised learning mode, finally predicts whether the samples appear in historical positions, namely a two-classification problem, derives characteristics from the existing little information and supports model training. According to the embodiment of the application, some rules in the statistical method can be processed into features to be added into model training, so that information obtained by the statistical rules can be covered in an algorithm, and the final result is superior to that of the statistical method.

The specific implementation of the position prediction system is substantially the same as the specific implementation of the position prediction method, and is not described herein again. On the premise of meeting the requirements of the embodiment of the application, the position prediction system can also be provided with other functional modules so as to realize the position prediction method in the embodiment.

The embodiment of the application also provides electronic equipment, wherein the electronic equipment comprises a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the position prediction method. The electronic equipment can be any intelligent terminal including a tablet computer, a vehicle-mounted computer and the like.

Referring to fig. 9, fig. 9 illustrates a hardware structure of an electronic device according to another embodiment, where the electronic device includes:

the processor 901 may be implemented by a general-purpose CPU (central processing unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits, and is configured to execute a relevant program to implement the technical solution provided in the embodiment of the present application;

the memory 902 may 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). The memory 902 may store an operating system and other application programs, and when the technical solution provided by the embodiments of the present disclosure is implemented by software or firmware, the relevant program codes are stored in the memory 902 and called by the processor 901 to execute the location prediction method according to the embodiments of the present disclosure;

an input/output interface 903 for implementing information input and output;

a communication interface 904, configured to implement communication interaction between the device and another device, where communication may be implemented in a wired manner (e.g., USB, network cable, etc.), or in a wireless manner (e.g., mobile network, WIFI, bluetooth, etc.);

a bus 905 that transfers information between various components of the device (e.g., the processor 901, the memory 902, the input/output interface 903, and the communication interface 904);

wherein the processor 901, the memory 902, the input/output interface 903 and the communication interface 904 enable a communication connection within the device with each other through a bus 905.

An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the location prediction method.

The memory, as a non-transitory computer-readable storage medium, may be used to store non-transitory software programs as well as non-transitory computer-executable programs. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and these remote memories may be connected to the processor through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute limitations on the technical solutions provided in the embodiments of the present application, and it is obvious to those skilled in the art that the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems with the evolution of technologies and the emergence of new application scenarios.

It will be appreciated by those skilled in the art that the embodiments shown in the figures are not intended to limit the embodiments of the present application and may include more or fewer steps than those shown, or some of the steps may be combined, or different steps may be included.

The above described embodiments of the apparatus are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, may be located in one place, or may be distributed over a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.

One of ordinary skill in the art will appreciate that all or some of the steps of the methods, systems, functional modules/units in the devices disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof.

The terms "first," "second," "third," "fourth," and the like in the description of the application and the above-described figures, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

It should be understood that, in this application, "at least one" means one or more, "a plurality" means two or more. "and/or" is used to describe the association relationship of the associated object, indicating that there may be three relationships, for example, "a and/or B" may indicate: only A, only B and both A and B are present, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.

In the several embodiments provided in the present application, it should be understood that the disclosed system and method may be implemented in other ways. For example, the above-described system embodiments are merely illustrative, and for example, the division of the above-described units is only one type of logical functional division, and other divisions may be realized in practice, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, which are essential or part of the technical solutions contributing to the prior art, or all or part of the technical solutions, may be embodied in the form of a software product stored in a storage medium, which includes multiple instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods of the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing programs, such as a usb disk, a portable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The preferred embodiments of the present application have been described above with reference to the accompanying drawings, and the scope of the claims of the embodiments of the present application is not limited thereto. Any modifications, equivalents, and improvements that may occur to those skilled in the art without departing from the scope and spirit of the embodiments of the present application are intended to be within the scope of the claims of the embodiments of the present application.

Claims

1. A method of location prediction, the method comprising:

acquiring first position information of a sample object at a plurality of historical moments and second position information of the sample object at the current moment;

marking according to the corresponding relation between the first position information and the second position information to obtain marking information, and determining a target variable according to the marking information;

performing feature construction according to the first position information to obtain sample features, inputting the sample features and the target variables into a preset position prediction model, and adjusting parameters of the position prediction model according to an output result of the position prediction model;

acquiring third position information of the target object at a plurality of historical moments;

and performing feature construction according to the third position information to obtain target features, and inputting the target features into the position prediction model after parameters are adjusted to obtain a position prediction result.

2. The method according to claim 1, wherein the labeling according to the correspondence between the first location information and the second location information to obtain labeled information, and determining a target variable according to the labeled information includes:

marking the first position information which is the same as the second position information to obtain first marking information;

marking the first position information different from the second position information to obtain second marking information;

the first position information including the first marker information is taken as a target variable, or the first position information including the second marker information is taken as the target variable.

3. The position prediction method according to claim 1, wherein there are a plurality of the sample objects; the performing feature construction according to the first position information to obtain a sample feature includes:

acquiring the number of positions represented by the first position information of each sample object and the first occurrence number of the sample object;

acquiring the number of first objects of the sample object on each piece of first position information and the second occurrence frequency of each piece of first position information;

and performing feature conversion according to the position number, the first occurrence number, the first object number and the second occurrence number to obtain sample features.

4. The method according to claim 1, wherein the step of inputting the sample characteristic and the target variable into a preset position prediction model, and the step of adjusting a parameter of the position prediction model according to an output result of the position prediction model comprises:

splitting the sample features and the target variables into training samples and test samples;

inputting the training sample 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 sample into the position prediction model after parameter adjustment to obtain a second output result, and inputting the test sample into the position prediction model after parameter adjustment to obtain a third output result;

and calculating to obtain a stability index according to the second output result and the third output result, and determining whether to continuously adjust the parameters of the position prediction model according to the stability index.

5. The location prediction method of claim 4, wherein the second output result comprises a first predicted probability value, and wherein the second output result comprises a second predicted probability value;

the calculating to obtain the stability index according to the second output result and the third output result includes:

adding and calculating according to the first prediction probability value and the second prediction probability value to obtain a first numerical value;

carrying out logarithmic calculation according to the first prediction probability value and the second prediction probability value to obtain a second numerical value;

and obtaining a stability index according to the product of the first numerical value and the second numerical value.

6. The method of claim 4, wherein the second output result comprises a first predicted probability value, the second output result comprises a second predicted probability value, and the first predicted probability value and the second predicted probability value are each plural;

dividing a plurality of score segments according to the first prediction probability value and the second prediction probability value;

respectively calculating the first prediction probability value and the second prediction probability value, and sub-stability indexes under the corresponding score segments;

and accumulating according to the sub-stability indexes under each fractional segment to obtain a stability index.

7. The method according to claim 1, wherein the performing a feature construction according to the third location information to obtain a target feature, and inputting the target feature into the location prediction model after adjusting parameters to obtain a location prediction result includes:

respectively carrying out feature construction according to the third position information to obtain a plurality of target features;

respectively inputting the target characteristics into the position prediction model after the parameters are adjusted to obtain a plurality of target prediction positions and corresponding target prediction probability values;

and determining a position prediction result in the plurality of prediction positions according to the magnitude of each target prediction probability value.

8. A position prediction system, characterized in that the system comprises:

the device comprises a sample object acquisition module, a data acquisition module and a data processing module, wherein the sample object acquisition module is used for 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;

the target variable acquisition module is used for marking according to the corresponding relation between the first position information and the second position information to obtain marking information, and determining a target variable according to the marking information;

the parameter adjusting module is used for carrying out 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;

the target object acquisition module is used for acquiring third position information of the target object at a plurality of historical moments;

and the position prediction module is used for carrying out feature construction according to the third position information to obtain target features, and inputting the target features into the position prediction model after parameters are adjusted to obtain a position prediction result.

9. An electronic device, characterized in that the electronic device comprises a memory and a processor, the memory storing a computer program which, when executed by the processor, implements the location prediction method of any one of claims 1 to 7.

10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the position prediction method of any one of claims 1 to 7.