User occupation identification method and device, equipment, storage medium and program product

Abstract

The application discloses a user occupation identification method and device, equipment, a storage medium and a program product, relates to the computer technical field, and discloses a user occupation identification method, comprising: generating a personal trajectory vector set corresponding to a user terminal according to the dynamic link relationship between the user terminal and a base station, and generating the time coding and the position coding of the personal trajectory vector set;Pre-training is carried out on the base large model based on the time coding and the position coding, and the static features of the user and the static features of the space contained in the pre-trained base large model are extracted;According to the static features of the user and the static features of the space, the pre-constructed occupation classifier is trained to obtain the trained occupation classifier;The trained occupation classifier is used for user occupation identification.The application improves the accuracy of user occupation identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, and particularly relates to a user occupation identification method and device, equipment, storage medium and program product. BACKGROUND

[0002] In a traditional scheme, the means for determining the occupation of a user based on data includes: constructing a classification model to predict the occupation based on the application category used by the user, application installation records and application usage data, etc. In the above means, data collection is difficult, feature vector construction is complex, and the time period is long, resulting in low accuracy of the predicted occupation of the user. SUMMARY

[0003] The main purpose of the present application is to provide a user occupation identification method, device, equipment, storage medium and program product, aiming at solving the technical problem of low accuracy of the predicted occupation of the user.

[0004] To achieve the above purpose, the present application provides a user occupation identification method, which comprises:

[0005] Generating a personal trajectory vector set corresponding to the user terminal according to the dynamic link relationship between the user terminal and the base station, and generating time encoding and position encoding of the personal trajectory vector set;

[0006] Pre-training the base large model based on the time encoding and the position encoding, and extracting the static features of the user and the static features of the space contained in the pre-trained base large model;

[0007] Training the pre-constructed occupation classifier according to the static features of the user and the static features of the space to obtain a trained occupation classifier; the trained occupation classifier is used for user occupation identification.

[0008] In an embodiment, the step of generating a personal trajectory vector set corresponding to the user terminal according to the dynamic link relationship between the user terminal and the base station comprises:

[0009] Generating a ternary grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station, the ternary grid group comprising a terminal number, a grid identifier of a grid area where the base station is located and a time of the user terminal in the grid identifier;

[0010] Generating a personal trajectory vector set with a time stamp according to the ternary grid group.

[0011] In an embodiment, before the step of generating a ternary grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station, the method further comprises:

[0012] Obtaining a target area;

[0013] grid division is performed on the target area to form a grid set covering the target area, the grid set including at least one grid area, each grid area corresponding to a grid identifier.

[0014] In an embodiment, the step of generating a ternary grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station comprises:

[0015] generating a terminal number, a base station latitude and longitude, and a time when the user terminal is at the base station latitude and longitude according to the dynamic link relationship between the user terminal and the base station;

[0016] determining a grid identifier where the base station is located according to the base station latitude and longitude;

[0017] generating a ternary grid group of the user terminal according to the grid identifier where the base station is located, the terminal number, and the time.

[0018] In an embodiment, the step of training a pre-constructed occupation classifier according to the static characteristics of the user and the static characteristics of the space comprises:

[0019] obtaining a set of occupation true value samples, and constructing a set of training trajectory vectors according to the set of occupation true value samples;

[0020] generating time encoding and position encoding of the set of training trajectory vectors;

[0021] training a pre-constructed occupation classifier based on the time encoding and the position encoding of the set of training trajectory vectors, and the static characteristics of the user and the static characteristics of the space.

[0022] In an embodiment, the step of pre-training a base large model based on the time encoding and the position encoding comprises:

[0023] constructing a pre-training task, the pre-training task including predicting missing grid identifiers and corresponding link times in the set of personal trajectory vectors, the missing grid identifiers being a set proportion of dynamically masked grid identifiers;

[0024] inputting the time encoding and the position encoding into the base large model for pre-training based on the pre-training task.

[0025] In addition, to achieve the above-mentioned purpose, the present application also provides a user occupation identification device, the user occupation identification device comprising:

[0026] a generation module configured to generate a set of personal trajectory vectors corresponding to the user terminal according to the dynamic link relationship between the user terminal and the base station, and generate time encoding and position encoding of the set of personal trajectory vectors;

[0027] extracting a static feature of a user and a static feature of a space contained in the pre-trained base large model based on the time encoding and the position encoding;

[0028] training a pre-constructed occupation classifier according to the static feature of the user and the static feature of the space to obtain a trained occupation classifier; the trained occupation classifier is used for user occupation identification.

[0029] In addition, to achieve the above object, the present application further provides a user occupation identification device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the user occupation identification method as described above.

[0030] In addition, to achieve the above object, the present application further provides a storage medium, which is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the user occupation identification method as described above.

[0031] In addition, to achieve the above object, the present application further provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps of the user occupation identification method as described above.

[0032] The one or more technical solutions provided by the present application have at least the following technical effects:

[0033] According to the dynamic link relationship between the user terminal and the base station, a personal trajectory vector set corresponding to the user terminal is generated, and a time encoding and a position encoding of the personal trajectory vector set are generated; a base large model is pre-trained based on the time encoding and the position encoding, and a static feature of a user and a static feature of a space contained in the pre-trained base large model are extracted; a pre-constructed occupation classifier is trained according to the static feature of the user and the static feature of the space to obtain a trained occupation classifier; the trained occupation classifier is used for user occupation identification. The personal trajectory vector set is generated through the dynamic link relationship, so as to reduce the calculation load of the large model and improve the comprehensiveness of personal trajectory determination, so as to facilitate subsequent user occupation identification. The static features are obtained by pre-training the base large model based on the time encoding and the position encoding, so as to construct the classifier. The classifier constructed by the static features in the dynamic link relationship between the user and the base station is used for identifying the occupation of the user, thereby improving the accuracy of user occupation identification. BRIEF DESCRIPTION OF DRAWINGS

[0034] The accompanying drawings, which are incorporated into and form a part of the specification, illustrate preferred embodiments of the present application and, together with the description, serve to explain the principles of the application.

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings required by the embodiments or the prior art description will be briefly introduced. Obviously, for those skilled in the art, based on the drawings, other drawings can be obtained without creative labor.

[0036] Figure 1 The flowchart provided by the user occupation identification method embodiment one of the present application;

[0037] Figure 2 The flowchart provided by the user occupation identification method embodiment two of the present application;

[0038] Figure 3 The module structure diagram of the user occupation identification device of the present application embodiment;

[0039] Figure 4 The device structure diagram of the hardware running environment involved in the user occupation identification method in the present application embodiment.

[0040] The purpose implementation, functional characteristics and advantages of the present application will be further explained with reference to the accompanying drawings combined with the embodiments. DETAILED DESCRIPTION

[0041] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application, and are not used to limit the present application.

[0042] In order to better understand the technical solutions of the present application, the following will be described in detail combined with the drawings in the specification and specific embodiments.

[0043] The main solution of the present application embodiment is: according to the dynamic link relationship between the user terminal and the base station, the personal trajectory vector set corresponding to the user terminal is generated, and the time coding and position coding of the personal trajectory vector set are generated; based on the time coding and position coding, the base station large model is pre-trained, and the static features of the user and the static features of the space contained in the pre-trained base station large model are extracted; according to the static features of the user and the static features of the space, the pre-constructed occupation classifier is trained to obtain the trained occupation classifier; the trained occupation classifier is used for user occupation identification.

[0044] In the present embodiment, for the convenience of description, the following describes the user occupation identification device as the execution subject.

[0045] The prior art determines the occupation of a user based on data, including constructing a classification model to predict the occupation based on the application category used by the user, application installation records, application usage data, and the like. The above means have difficulties in data collection, complex feature vector construction, and long time period, resulting in low accuracy of the final predicted occupation of the user.

[0046] The present application provides a solution that generates a personal trajectory vector set through dynamic link relationships to reduce the computational load of large models and improve the comprehensiveness of personal trajectory determination to facilitate subsequent user occupation identification. The base large model is pre-trained based on time encoding and location encoding to obtain static features, and a classifier is constructed based on the static features exhibited in the dynamic link relationship between the user and the base station to identify the occupation of the user, thereby improving the accuracy of user occupation identification.

[0047] It should be noted that the execution subject of the present embodiment can be a computing service device with data processing, network communication, and program running functions, such as a tablet computer, a personal computer, a mobile phone, or an electronic device capable of realizing the above functions, a user occupation identification device, etc. The present embodiment and the following embodiments will be described below with the user occupation identification device as an example.

[0048] Based on this, the present embodiment provides a user occupation identification method, which is described below with reference to Figure 1 , Figure 1 The flowchart of the first embodiment of the user occupation identification method of the present application is shown in the figure.

[0049] In the present embodiment, the user occupation identification method includes steps S10-S30:

[0050] Step S10, according to the dynamic link relationship between the user terminal and the base station, a personal trajectory vector set corresponding to the user terminal is generated, and a time encoding and a location encoding of the personal trajectory vector set are generated.

[0051] The present embodiment can be applied to the planning and landing fields of smart cities, advertising marketing, infrastructure, and commercial supporting, etc. In the present embodiment, the dynamic link relationship between the user terminal and the base station is information obtained after the user terminal is authorized.

[0052] It should be noted that the dynamic link relationship indicates that the position and signal strength of the user terminal change over time, and the connection between the user terminal and the base station is dynamic. The dynamic link relationship includes parameters such as mobile phone number, base station latitude and longitude, and link time of the user terminal at the current latitude and longitude. The personal trajectory vector set is a vector set corresponding to the user terminal, and the vectors in the set describe the spatial position, activity pattern, and movement path of the user.

[0053] Optionally, the signal strength, location data, and base station information in the dynamic connection relationship are acquired, and the dynamic connection relationship is processed, including feature extraction, location mapping, vector construction, to generate a personal trajectory vector set corresponding to the user terminal. Optionally, the vectors in the personal trajectory vector set can include the user's current location, speed, moving direction, time information, etc.

[0054] It should be noted that the signal strength is the signal strength between the user terminal and multiple base stations, which is used to estimate the distance and construct the user's moving trajectory in combination with the timestamp. The location data is the actual location of the user terminal obtained in combination with GPS (Global Positioning System) data or other location tracking technologies. The base station information includes the location information and coverage range of the base station.

[0055] Feature extraction is to extract features such as moving speed, stay time, and visited base station type from the signal strength and location data. Location mapping is to map the location data into a high-dimensional space and consider the user's movement and stay between different base stations. Vector construction is to construct the user's spatial vector based on the extracted features.

[0056] The personal trajectory vector set can also be generated through time series modeling, clustering and classification, and personalized features. Among them, time series modeling is to model the user's moving pattern using time series data analysis techniques such as time series neural network and recurrent neural network, to generate time series spatial vectors. Clustering and classification is to cluster and analyze the generated vectors, to classify the user's moving pattern into different categories, or to classify the vectors to identify the user's common activity area or behavior pattern. Personalized features are to generate personalized spatial vector sets according to the user's historical moving data and behavior. For example, a user may frequently go to a specific location during a specific time period, and this information can be incorporated into the vector.

[0057] Optionally, the personal trajectory vector set is used for location prediction to predict the user's future location or moving path. Optionally, the personal trajectory vector set is used to analyze the user's behavior pattern to provide personalized service recommendation or optimize network resource allocation. Optionally, the personal trajectory vector set is used for anomaly detection to detect abnormal patterns in the user's behavior, such as abnormal moving speed or location deviation.

[0058] For example, the location data and signal strength of the user terminal at different time points are as follows:

[0059] Time Base station 1 signal strength Base station 2 signal strength GPS coordinates T1 -70 dBm -65 dBm (40.748,-73.985) T2 -68 dBm -67 dBm (40.749,-73.986) T3 -72 dBm -66 dBm (40.750,-73.987)

[0060] From these data, the user's moving speed, frequently visited area, signal strength change, etc. can be calculated, and the following personal trajectory vector can be generated:

[0061] Vector 1: contains the user's location and signal strength features at time T1;

[0062] Vector 2: contains the user's location, moving speed, and signal strength change at time T2;

[0063] Vector 3: contains the user's location, moving direction, and stay time at time T3.

[0064] Wherein, each vector can be further used to analyze the user's behavior patterns, predict future locations, etc.

[0065] In a feasible embodiment, the spatial vectors in the set of personal trajectory vectors are encoded to obtain time encoding and location encoding.

[0066] The time encoding can be generated by timestamp conversion, in which the timestamp is converted into discrete time units such as hours, minutes, and seconds. For example, the time of day is converted to "08:00" or "14:30". Alternatively, the time can be encoded using sine and cosine functions to preserve the periodicity of the time. For example, the hour number is encoded using sin(2π*t / 24) and cos(2π*t / 24). The time encoding includes time sequence features, in which the time interval is calculated, such as the time spent by the user moving from one location to another, as a time sequence feature.

[0067] The location encoding can be location coordinate standardization, i.e. standardizing the location coordinates (such as latitude and longitude) to a fixed range (0 to 1) to make it suitable for machine learning models. The location encoding can be to divide the geographical area into grids or regions, and each grid is assigned a unique encoding. For example, the city is divided into several grids, and the location is represented by the grid number. The location encoding can be to use location embedding technology to map the location coordinates to a low-dimensional vector space. This method can capture the similarity between geographical locations.

[0068] Combining time encoding and location encoding can create a full-time personal trajectory vector that contains time and location information, which can better capture the user's behavior patterns and location changes.

[0069] Step S20, pre-training the base large model based on the time encoding and the location encoding, and extracting the static features of the user and the static features of the space contained in the pre-trained base large model.

[0070] Optionally, a transformer model architecture based on self-attention mechanism is selected as the base model, and the set of personal trajectory vectors is input for encoding to generate time encoding and location encoding.

[0071] Optionally, step S20 comprises: constructing a pre-training task, the pre-training task comprising predicting missing grid identifications and corresponding link times in the set of personal trajectory vectors, the missing grid identifications being a set proportion of dynamically masked grid identifications; inputting the time encoding and the position encoding into a base large model based on the pre-training task for pre-training.

[0072] It should be noted that after generating the time encoding and the position encoding, the time encoding and the position encoding are input into a base large model using a bidirectional attention mechanism, and a pre-training task of predicting a set proportion of dynamically masked grid IDs (Identification, identity) and corresponding link times in each unlabeled trajectory is used to pre-train the dynamic large model, and the model is iterated and converged to form a pre-training dynamic large model skeleton containing static and dynamic features of users and spaces. Optionally, the set proportion is 20%.

[0073] The professional attributes of the user determine the specific statistical rules or characteristics exhibited in the dynamic link relationship between the user's mobile phone and the base station. A pre-training dynamic learning large model based on the set of personal trajectory vectors is constructed, and based on the time encoding and the position encoding, long-sequence space observation learning with absolute time information is performed to obtain user and regional dynamic and static representations.

[0074] Static features are relatively stable features of a user over a period of time, and static features are not easily changed over time. Static features include regular patterns exhibited in the link relationship between the user and the base station, which can reflect the professional attributes of the user. For example, a person's occupation can affect their daily movement patterns, such as places they frequently visit, times they visit these places, etc.

[0075] For example, static features include: a home location, which is a place that a user frequently visits, such as a home address or a workplace, which is relatively stable over a period of time. Visit frequency, for example, the frequency with which a user visits a particular location. Visit time, the time pattern in which a user visits a particular location. Activity area, i.e. the main area in which a user is active, such as some people may be mainly active in business districts, while others may be more active in residential areas. Movement pattern, i.e. the pattern in which a user moves from one location to another, such as a daily commute route. Base station link pattern, i.e. the strength and duration of a user's link with a particular base station, reflecting the user's stay time in a certain area. Device usage habits, i.e. the habits of a user using a mobile phone or other mobile devices, such as the frequency of use of certain applications. Social network structure, i.e. the position of a user in a social network, such as contacts, the frequency and strength of contact with contacts. Consumption habits, such as the consumption patterns of a user at a particular location. Daily activity patterns of a user, such as wake-up time, bedtime, leisure activity time.

[0076] Step S30, training a pre-constructed occupation classifier according to the static features of the user and the static features of the space, to obtain a trained occupation classifier; the trained occupation classifier is used for user occupation identification.

[0077] Optionally, based on the long-period invariable characteristics of the user's occupation attribute, the static features of the user and the space contained in the hidden layer of the generated pre-training dynamic large model are selected to construct an occupation classifier, wherein the occupation classifier can be a binary classifier or a multi-classifier.

[0078] Optionally, the training of the classifier in step S30 includes: obtaining an occupation true value sample set, and constructing a training trajectory vector set according to the occupation true value sample set; generating time encoding and position encoding of the training trajectory vector set; training a pre-constructed occupation classifier based on the time encoding and position encoding of the training trajectory vector set, and the static features of the user and the static features of the space.

[0079] It should be noted that after the classifier is constructed, based on the long-period occupation true value sample data, the true value of the occupation engaged in for a longer period of time is selected as much as possible to construct a full-time period personal trajectory vector, and a large model classifier mode based on full-amount parameter fine-tuning of long-time span true value sample is trained.

[0080] Specifically, the true value of the occupation engaged in for a longer period of time is selected from the user occupation true value sample set, the personal trajectory vector set thereof is constructed based on the selected true value, and after time encoding and position encoding, the classifier is trained, and full-amount parameter fine-tuning of the pre-training dynamic large model skeleton is triggered to refresh the hidden user's static features, the static features of the space and the classifier parameters, so as to better capture the corresponding space-time features of different occupation type crowds.

[0081] Optionally, the personal trajectory vector set corresponding to the user terminal is input into the trained classifier, and the identification result of the classifier on the occupation of the user of the user terminal is obtained.

[0082] In the technical scheme of the embodiment, according to the dynamic link relationship between the user terminal and the base station, a personal trajectory vector set corresponding to the user terminal is generated, and time encoding and position encoding of the personal trajectory vector set are generated; the base station large model is pre-trained based on the time encoding and the position encoding, and static features of the user and static features of a space contained in the pre-trained base station large model are extracted; the pre-constructed occupation classifier is trained according to the static features of the user and the static features of the space, and a trained occupation classifier is obtained; and the trained occupation classifier is used for user occupation identification. The personal trajectory vector set is generated through the dynamic link relationship, so as to reduce the large model calculation load and improve the comprehensiveness of personal trajectory determination, so as to facilitate subsequent user occupation identification. The static features are obtained by pre-training the base station large model according to the time encoding and the position encoding, so as to construct the classifier. The classifier constructed by the static features shown in the dynamic link relationship between the user and the base station is used for identifying the occupation of the user, and the accuracy of the user occupation identification is improved.

[0083] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as the above-mentioned first embodiment can be referred to the above introduction, and will not be described in detail. On this basis, please refer to Figure 2 , step S10 includes:

[0084] Step S11, according to the dynamic link relationship between the user terminal and the base station, a ternary grid group of the user terminal is generated, the ternary grid group includes terminal number, grid identification of grid area where the base station is located and time of the user terminal in the grid identification;

[0085] Step S12, generating a personal trajectory vector set with time stamp according to the ternary grid group.

[0086] Optionally, before step S11, it further includes: acquiring a target area; performing grid division on the target area to form a grid set covering the target area, the grid set includes at least one grid area, and each grid area corresponds to a grid identification.

[0087] For the target area, the target area includes a city area, the target area is divided into a grid size of a preset size, for example, the preset size is 300m*300m, a grid set covering the target area is formed, and each grid in the grid set has a unique grid identification.

[0088] According to the dynamic link relationship between the terminal and the base station, the city is appropriately gridded, and a new ternary grid group is formed, and a personal trajectory vector set with time stamp data is formed.

[0089] Optionally, step S11 further comprises: generating a terminal number, a base station latitude and longitude, and a time of the user terminal at the base station latitude and longitude according to the dynamic link relationship between the user terminal and the base station; determining the grid identification where the base station is located according to the base station latitude and longitude; and generating a three-element grid group of the user terminal according to the grid identification where the base station is located, the terminal number, and the time. For example, based on the dynamic link relationship between the mobile phone and the base station, a three-element group is formed, which includes a mobile phone number, a base station latitude and longitude, and a time. According to the base station latitude and longitude, the grid where the base station is located is calculated, and the grid identification is obtained, which is converted to form a new three-element grid group, i.e., the three-element grid group includes a mobile phone number, a grid identification, and a time.

[0090] A suitable sampling period and sampling frequency, such as a 1-month period and a 15-minute frequency, are selected to construct a set of personal trajectory vectors based on the three-element grid group, so as to form a data set suitable for pre-training a large model and reduce the computational load of the large model.

[0091] The massive data based on the dynamic link relationship between the user's mobile terminal and the base station contains the dynamic and static features of the user and the space, and contains specific statistical rules or features in the link relationship between the user and the base station determined by the user's professional attributes. Through the new three-element space-time grid group, a pre-trained dynamic large model is generated, and a suitable proportion of grid identifications and corresponding times that are dynamically hidden in each unlabeled trajectory are predicted as a pre-training task. The model is iterated and converged, and finally based on the pre-trained dynamic large model, a multi-classification model is formed through the means of driving the pre-trained dynamic learning large model parameter full adjustment by the professional true value, so as to realize the goal of one model serving multiple tasks, and alleviate the deficiency of traditional large models considering less explicit time elements, thereby solving or alleviating the deficiencies of traditional methods such as difficult data collection, complex feature vector construction, long time period, and high cost.

[0092] The transformer model architecture based on a self-attention mechanism is selected as a base model, a grid set with a proper size is set in a target area or city to form full coverage, each small grid has a unique identifier, a three-tuple is formed through a dynamic link relationship between a mobile phone and a base station, the latitude and longitude of the base station is converted into the corresponding grid identifier to form a new three-tuple grid set, a personal trajectory vector constituted by the three-tuple grid set is formed according to a proper sampling time, after all personal trajectory vector sets are time-encoded and position-encoded, a bidirectional attention mechanism is used to input the base model to generate a pre-trained dynamic large model, a pre-training task of predicting a proper proportion of grid identifiers and corresponding time in each unlabeled trajectory which are dynamically masked is performed, the model is iterated and converged, finally, the pre-trained dynamic large model is fine-tuned based on the static features of users and spaces contained in the hidden layer of the pre-trained dynamic large model, a long-period true value sample set is used to refresh the static features of users and spaces to better capture the corresponding features of different professional groups, so as to construct a two-classifier or multi-classifier to realize accurate prediction of the occupation of a person.

[0093] In the technical scheme of the embodiment, the three-tuple grid set is constructed through a dynamic link relationship to generate a personal trajectory vector set, the personal trajectory vector set is input into a large model to reduce the calculation load of the large model and improve the comprehensiveness of personal trajectory determination, so as to facilitate subsequent user occupation identification.

[0094] It should be noted that the above examples are only used for understanding the present application and do not constitute a limitation on the user occupation identification method of the present application, and more forms of simple transformation based on the technical concept are within the protection scope of the present application.

[0095] The present application also provides a user occupation identification device, please refer to Figure 3 , the user occupation identification device comprises:

[0096] The generation module 10 is configured to generate a personal trajectory vector set corresponding to a user terminal according to a dynamic link relationship between the user terminal and a base station, and generate time encoding and position encoding of the personal trajectory vector set;

[0097] The extraction module 20 is configured to pre-train a base large model based on the time encoding and the position encoding, and extract static features of users and static features of spaces contained in the pre-trained base large model;

[0098] The training module 30 is configured to train a pre-constructed occupation classifier according to the static features of the users and the static features of the spaces to obtain a trained occupation classifier; the trained occupation classifier is used for user occupation identification.

[0099] Optionally, the step of generating the set of personal trajectory vectors corresponding to the user terminal according to the dynamic link relationship between the user terminal and the base station comprises:

[0100] generating a tri-grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station, the tri-grid group comprising a terminal number, a grid identifier of a grid area where the base station is located, and a time when the user terminal is in the grid identifier;

[0101] generating a set of personal trajectory vectors with timestamps according to the tri-grid group.

[0102] Optionally, before the step of generating the tri-grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station, the method further comprises:

[0103] acquiring a target area;

[0104] dividing the target area into grids to form a grid set covering the target area, the grid set comprising at least one grid area, and each grid area corresponding to a grid identifier.

[0105] Optionally, the step of generating the tri-grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station comprises:

[0106] generating a terminal number, a base station latitude and longitude, and a time when the user terminal is in the base station latitude and longitude according to the dynamic link relationship between the user terminal and the base station;

[0107] determining a grid identifier where the base station is located according to the base station latitude and longitude;

[0108] generating a tri-grid group of the user terminal according to the grid identifier where the base station is located, the terminal number, and the time.

[0109] Optionally, the step of training the pre-constructed occupation classifier according to the static features of the user and the static features of the space comprises:

[0110] acquiring a set of occupation true value samples, and constructing a set of training trajectory vectors according to the set of occupation true value samples;

[0111] generating time encoding and position encoding of the set of training trajectory vectors;

[0112] training the pre-constructed occupation classifier based on the time encoding and the position encoding of the set of training trajectory vectors, and the static features of the user and the static features of the space.

[0113] Optionally, the step of pre-training the base model based on the time encoding and the position encoding comprises:

[0114] constructing a pre-training task, the pre-training task comprising predicting missing grid identifications and corresponding link times in the set of personal trajectory vectors, the missing grid identifications being a set proportion of the dynamically masked grid identifications;

[0115] inputting the time encoding and the position encoding into a base large model for pre-training based on the pre-training task.

[0116] The user occupation identification device provided by the present application adopts the user occupation identification method in the above embodiments, and can solve the technical problem of low accuracy of predicted user occupation. Compared with the prior art, the user occupation identification device provided by the present application has the same beneficial effects as the user occupation identification method provided by the above embodiments, and other technical features in the user occupation identification device are the same as the features disclosed in the above embodiments, which will not be repeated here.

[0117] The present application provides a user occupation identification device, which comprises at least one processor and a memory in communication with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the user occupation identification method in Embodiment I.

[0118] Reference will now be made to the drawings, in which Figure 4 which shows a structural diagram of a user occupation identification device suitable for implementing the embodiments of the present application. The user occupation identification device in the embodiments of the present application can include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and the like, and fixed terminals such as digital TVs (Television), desktop computers, and the like. Figure 4 The user occupation identification device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0119] As Figure 4As shown, the user occupation identification device can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. Various programs and data required for operation of the user occupation identification device are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other by a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the user occupation identification device to communicate wirelessly or by wire with other devices to exchange data. Although the user occupation identification device is shown as having various systems, it should be understood that all of the shown systems are not required to be implemented or possessed. More or fewer systems can alternatively be implemented or possessed.

[0120] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of embodiments of the present disclosure are performed.

[0121] The user occupation identification device provided by the present disclosure adopts the user occupation identification method in the above-mentioned embodiments, and can solve the technical problem of low accuracy of predicted user occupation. Compared with the prior art, the user occupation identification device provided by the present disclosure has the same beneficial effects as the user occupation identification method provided by the above-mentioned embodiments, and other technical features in the user occupation identification device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0122] It should be understood that various parts of the present application can be realized with hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0123] The above description is merely that of a specific implementation of the present application, and the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, and all such changes or replacements should be encompassed within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0124] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e., computer programs) for performing the user occupation identification method in the above embodiments.

[0125] The computer readable storage medium provided by the present application may, for example, be a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electric connection with one or more conductive wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted by any suitable medium, including but not limited to an electric wire, an optical cable, an RF (Radio Frequency), etc., or any suitable combination of the above.

[0126] The above computer readable storage medium can be contained in the user occupation identification device; or can exist separately without being assembled into the user occupation identification device.

[0127] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the user occupation identification device, the user occupation identification device: generates a personal trajectory vector set through a dynamic link relationship, reduces the large model calculation load, improves the comprehensiveness of personal trajectory determination, and facilitates subsequent user occupation identification, pre-trains a base large model according to time coding and position coding to obtain static features, constructs a classifier, and identifies the occupation of the user through the classifier constructed by the static features shown in the dynamic link relationship between the user and the base station, and improves the accuracy of user occupation identification.

[0128] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0129] The flow diagrams and the block diagrams in the drawings are illustrations of possible architectures, functions, and operations for systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flow diagrams, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.

[0130] The modules described in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the names of the modules do not constitute a limitation on the modules themselves.

[0131] The computer readable storage medium provided in the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the user occupation identification method described above, and can solve the technical problem of low accuracy of predicted user occupation. Compared with the prior art, the computer readable storage medium provided in the present application has the same beneficial effects as the user occupation identification method provided in the above embodiments, and will not be described here.

[0132] The present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the user occupation identification method as described above.

[0133] The computer program product provided in the present application can solve the technical problem of low accuracy of predicted user occupation. Compared with the prior art, the computer program product provided in the present application has the same beneficial effects as the user occupation identification method provided in the above embodiments, and will not be described here.

[0134] The above only describes some embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structural transformation made by using the content of the present application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A user occupation identifying method, characterized by, The method comprises: According to the dynamic link relationship between the user terminal and the base station, a three-element grid group of the user terminal is generated, which includes the terminal number, the grid identification of the grid area where the base station is located, and the time of the user terminal in the grid identification; According to the three-element grid group, a personal trajectory vector set with a time stamp is generated, and a time code and a position code of the personal trajectory vector set are generated, and the features described in the personal trajectory vector set include the spatial position, the activity mode and the moving path of the user; Based on the time code and the position code, a base large model is pre-trained, and the static features of the user and the static features of the space contained in the pre-trained base large model are extracted; According to the static features of the user and the static features of the space, a pre-constructed occupation classifier is trained to obtain a trained occupation classifier; the trained occupation classifier is used for user occupation identification.

2. The method of claim 1, wherein, Before the step of generating a three-element grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station, the method further comprises: Obtain a target area; Grid division is performed on the target area to form a grid set covering the target area, and the grid set includes at least one grid area, and each grid area corresponds to a grid identification.

3. The method of claim 2, wherein, The step of generating a three-element grid group of the user terminal according to the dynamic link relationship between the user terminal and the base station comprises: According to the dynamic link relationship between the user terminal and the base station, the terminal number, the base station longitude and latitude, and the time of the user terminal in the base station longitude and latitude are generated; According to the base station longitude and latitude, the grid identification where the base station is located is determined; According to the grid identification where the base station is located, the terminal number and the time, a three-element grid group of the user terminal is generated.

4. The method of claim 1, wherein, The step of training a pre-constructed occupation classifier according to the static features of the user and the static features of the space comprises: Obtain a set of occupation true value samples, and construct a training trajectory vector set according to the set of occupation true value samples; Generate the time code and the position code of the training trajectory vector set; Based on the time code and the position code of the training trajectory vector set, and the static features of the user and the static features of the space, the pre-constructed occupation classifier is trained.

5. The method of claim 1, wherein, The step of pre-training a base large model based on the time code and the position code comprises: Construct a pre-training task, which includes predicting the missing grid identification and the corresponding link time in the personal trajectory vector set, and the missing grid identification is a grid identification that is dynamically masked by a set proportion; Based on the pre-training task, the time code and the position code are input into the base large model for pre-training.

6. A user occupation identifying apparatus characterized by comprising: The device comprises: The generating module is configured to generate a ternary grid group of the user terminal according to a dynamic link relationship between the user terminal and the base station, the ternary grid group including a terminal number, a grid identifier of a grid area where the base station is located, and a time of the user terminal in the grid identifier; generate a personal trajectory vector set with a time stamp according to the ternary grid group, and generate time encoding and position encoding of the personal trajectory vector set, the personal trajectory vector set describing features including a spatial position, an activity pattern, and a moving path of the user; The extracting module is configured to pre-train a base large model based on the time encoding and the position encoding, and extract static features of the user and static features of a space included in the pre-trained base large model; The training module is configured to train a pre-constructed occupation classifier according to the static features of the user and the static features of the space, to obtain a trained occupation classifier; and the trained occupation classifier is configured to perform user occupation identification.

7. A user occupation identifying apparatus characterized by comprising: The device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the user occupation identification method according to any one of claims 1 to 5.

8. A storage medium, characterized by The storage medium is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the user occupation identification method according to any one of claims 1 to 5.

9. A computer program product, characterised in that, The computer program product includes a computer program, and the computer program is executed by the processor to implement the steps of the user occupation identification method according to any one of claims 1 to 5.

Images