Object Recognition Method, Device, Computer-Readable Storage Medium, and Electronic Device

Abstract

The present invention discloses a method, apparatus, computer-readable storage medium, and electronic device for identifying an object. It relates to the field of artificial intelligence. The method includes: obtaining the behavior information of a target object within a first preset time range, where the behavior information at least includes the geographical location information of at least one scene where the target object has stayed; determining a first feature vector based on the behavior information, a first deposit time, and the deposit frequency of the target object, where the first deposit time represents the deposit time of the target object's last deposit, and the first feature vector is used to characterize the deposit characteristics of the target object; identifying the object type of the target object based on the first feature vector, where the object type represents whether the target object is an object whose requested deposit time is greater than a preset time. The present invention solves the technical problem of low identification accuracy of the related method for identifying customer types in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular, to a method, apparatus, computer-readable storage medium, and electronic device for identifying an object. Background Art

[0002] Under the current international economic situation, relevant financial institutions (such as banks, etc.) need to sell more long-term deposits to increase financial reserves. The long-term deposits of customers are very important to banks, and understanding customer characteristics is the key for banks to increase product sales. Therefore, relevant personnel have begun to use statistical strategies to identify potential users of bank long-term deposits.

[0003] In recent years, scientists have found that through variable data analysis, feature selection, and machine learning techniques, it is possible to analyze customer characteristics and variables that can affect customer decisions to identify different types of consumers, thereby determining customer types, such as whether a customer will make a long-term deposit. However, in the related methods of the prior art, there is a problem of low recognition accuracy in the process of identifying customer types.

[0004] In view of the above problems, no effective solution has been proposed yet. Summary of the Invention

[0005] Embodiments of the present invention provide a method, apparatus, computer-readable storage medium, and electronic device for identifying an object, so as to at least solve the technical problem of low recognition accuracy of related methods for identifying customer types in the prior art.

[0006] According to an aspect of an embodiment of the present invention, a method for identifying an object is provided, including: obtaining behavior information of a target object within a first preset time range, where the behavior information at least includes geographical location information of at least one scene where the target object has stayed; determining a first feature vector based on the behavior information, a first deposit time, and the deposit frequency of the target object, where the first deposit time represents the deposit time of the target object's last deposit, and the first feature vector is used to represent the deposit characteristics of the target object; identifying the object type of the target object based on the first feature vector, where the object type represents whether the target object is an object whose requested deposit time is greater than a preset time.

[0007] Further, the method for identifying an object further includes: sorting a plurality of geographical location information based on the stay order to obtain a target sequence, where the behavior information further includes the stay order corresponding to at least one scene where the target object has stayed; determining the transition probability between the positions corresponding to any two geographical location information among the plurality of geographical location information of the target object based on the target sequence to obtain a plurality of target transition probabilities; determining the first feature vector based on the plurality of target transition probabilities, the first deposit time, and the deposit frequency of the target object.

[0008] Further, the object recognition method further includes: generating a first target transition probability matrix based on multiple target transition probabilities; performing dimensionality reduction processing on the first target transition probability matrix to obtain a second target transition probability matrix; and determining a first feature vector based on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object.

[0009] Further, the object recognition method further includes: performing normalization processing on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object to obtain a first feature vector.

[0010] Further, the object recognition method further includes: before recognizing the object type of the target object based on the first feature vector, constructing a training set based on the second feature vector and the object types of historical objects, where the second feature vector is used to characterize the deposit characteristics of historical objects; training a preset model based on the training set to obtain a target preset model, where the target preset model is used to recognize the object type of the target object based on the first feature vector.

[0011] Further, the object recognition method further includes: before constructing a training set based on the second feature vector and the object types of historical objects, obtaining the historical behavior information of historical objects within a second preset time range; and determining a second feature vector based on the historical behavior information, the second deposit time, and the deposit frequency of historical objects, where the second deposit time represents the deposit time of the last deposit of historical objects.

[0012] According to another aspect of the embodiments of the present invention, there is also provided an object recognition device, including: an acquisition module, configured to acquire the behavior information of a target object within a first preset time range, where the behavior information at least includes the geographical location information of at least one scene where the target object has stayed; a determination module, configured to determine a first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object, where the first deposit time represents the deposit time of the last deposit of the target object, and the first feature vector is used to characterize the deposit characteristics of the target object; and an identification module, configured to identify the object type of the target object based on the first feature vector, where the object type represents whether the target object is an object whose requested deposit time is greater than a preset time.

[0013] According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, in which a computer program is stored, where the computer program is configured to execute the above-mentioned object recognition method when running.

[0014] According to another aspect of the embodiments of the present invention, an electronic device is further provided. The electronic device includes one or more processors; a memory for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement a program for running, where the program is configured to execute the above object recognition method when running.

[0015] According to another aspect of the embodiments of the present invention, a computer program product is further provided, including a computer program / instructions, which when executed by a processor, implement the above object recognition method.

[0016] In the embodiments of the present invention, a method for identifying the object type of a target object is adopted based on the behavior information of the target object, the deposit time of the last deposit, and the deposit frequency. By obtaining the behavior information of the target object within a first preset time range, and then based on the behavior information, the first deposit time, and the deposit frequency of the target object, a first feature vector is determined. Thus, based on the first feature vector, the object type of the target object is identified. Among them, the behavior information at least includes the geographical location information of at least one scene where the target object has stayed. The first deposit time represents the deposit time of the target object's last deposit. The first feature vector is used to represent the deposit characteristics of the target object. The object type represents whether the target object is an object whose requested deposit time is greater than the preset time.

[0017] In the above process, by determining the first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object, the richness of the first feature vector corresponding to the target object is achieved, so that the deposit characteristics of the target object can be better reflected. Furthermore, it promotes the object type of the target object identified based on the first feature vector to be more accurate, avoiding the problem of low recognition accuracy for objects with low consumption amount and low purchase frequency in the prior art when identifying the object type only based on the consumption behavior of the target object.

[0018] It can be seen that the solution provided by this application achieves the purpose of identifying the object type of the target object based on the behavior information of the target object, the deposit time of the last deposit, and the deposit frequency, thus achieving the technical effect of improving the recognition accuracy, and further solving the technical problem of low recognition accuracy of the related methods for identifying customer types in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 is a schematic diagram of an optional object recognition method according to an embodiment of the present invention;

[0021] Figure 2 is a flowchart of an alternative object recognition method according to an embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of an alternative object recognition device according to an embodiment of the present invention;

[0023] Figure 4 is a schematic diagram of an alternative electronic device according to an embodiment of the present invention. Detailed implementation manners

[0024] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should be noted that the relevant information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in the present disclosure are all information and data authorized by the user or fully authorized by all parties. For example, an interface is provided between the present system and relevant users or institutions. Before obtaining relevant information, a request for acquisition needs to be sent to the aforementioned users or institutions through the interface, and after receiving the consent information feedback by the aforementioned users or institutions, the relevant information can be obtained.

[0027] Embodiment 1

[0028] According to an embodiment of the present invention, an embodiment of a method for identifying an object is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0029] Figure 1 is a schematic diagram of an optional method for identifying an object according to an embodiment of the present invention, as Figure 1 shown, the method includes the following steps:

[0030] Step S101, obtain the behavior information of the target object within a first preset time range, where the behavior information at least includes the geographical location information of at least one scene where the target object has stayed.

[0031] In an optional embodiment, the behavior information of the target object within a first preset time range can be obtained through devices such as electronic devices, servers, application systems, etc. In this application, the behavior information of the target object within a first preset time range is obtained through an object recognition system. Among them, the target object is an object that has been registered and deposited in a financial institution (such as a bank, etc.).

[0032] Optionally, the object recognition system can obtain the spatio-temporal behavior data (i.e., behavior information) of the target object within the past month, or within the past year, or within other preset time ranges. Among them, the behavior information at least includes the geographical location information of at least one POI (point of interest) where the target object has stayed, and the POI corresponds to the geographical location one by one. A POI can be a restaurant, hotel, beach, bank, cinema, park, etc. The POI is equivalent to the aforementioned scene. The behavior information of the target object has a certain correlation with whether the target object is likely to be a long-term deposit user. For example, when it is determined based on the behavior information that the target object often visits high-consumption places, the target object is more likely to become a long-term deposit user.

[0033] Specifically, the object recognition system can set a residence time threshold, and when the time that the target object stays in a certain scene is greater than the residence time threshold, determine that the scene is a POI where the target object has stayed, and obtain the geographical location information corresponding to the scene, where the geographical location information can include the longitude and latitude information of the location where the scene is located.

[0034] It should be noted that by obtaining the behavior information of the target object, it is convenient to subsequently identify the object type.

[0035] Step S102: Determine a first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object. Herein, the first deposit time represents the deposit time of the target object's last deposit, and the first feature vector is used to represent the deposit characteristics of the target object.

[0036] In step S102, the object recognition system may first obtain the time of the target object's last deposit, e.g., May 3, 2021, and the deposit frequency of the target object. Herein, the deposit frequency of the target object may be determined based on the deposit behavior of the target object within a preset time range, or may be determined based on the deposit behavior of the target object during the period from the time when the target object generates the first deposit record to the current time.

[0037] Furthermore, after determining the first deposit time and the deposit frequency of the target object, the object recognition system may determine the first feature vector in combination with the aforementioned behavior information. Herein, the object recognition system may process the behavior information and determine the first feature vector based on the processed behavior information, the first deposit time, and the deposit frequency of the target object. The processing performed by the object recognition system on the behavior information may be: determining the POIs that the target object often visits, the POIs that the target object often consumes, or the occurrence frequency corresponding to each POI, the occurrence time corresponding to each POI, the transition probability between any POIs of the target object, etc.

[0038] It should be noted that determining the first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object enriches the first feature vector corresponding to the target object, thereby better reflecting the deposit characteristics of the target object, and further enabling the object type of the target object recognized based on the first feature vector to be more accurate, avoiding the problem of low recognition accuracy for objects with low consumption amounts and low purchase frequencies in the prior art when identifying the object type only based on the consumption behavior of the target object.

[0039] Step S103: Identify the object type of the target object based on the first feature vector. Herein, the object type represents whether the target object is an object whose requested deposit time is greater than the preset time.

[0040] Optionally, the object recognition system may use a pre-trained target preset model to identify the object type of the target object based on the first feature vector. Herein, the target preset model may be a deep forest model or other machine learning models. Optionally, the object recognition system may also use other methods to identify the object type of the target object.

[0041] In step S103, the object recognition system may use the first feature vector X of the target object testInput the pre-trained deep forest model, and then obtain the corresponding output result Y test , where the output result is used to characterize the possibility that the target object is a non-long-term deposit customer, that is, if Y test approaches 0, it is determined that the target object is a potential customer for long-term deposits, that is, it is determined that the target object is an object whose requested deposit time is greater than the preset time and may make long-term deposits. If Y test approaches 1, it is determined that the target object is a customer who does not make long-term deposits, that is, it is determined that the target object is an object whose requested deposit time is less than or equal to the preset time, and the possibility of making long-term deposits is relatively low.

[0042] It should be noted that by identifying the object type of the target object based on the first feature vector, the effective identification of the object type of the target object is realized.

[0043] Based on the solution defined in the above steps S101 to S103, it can be known that in the embodiment of the present invention, a method for identifying the object type of a target object based on the behavior information of the target object, the deposit time of the last deposit, and the deposit frequency is adopted. By obtaining the behavior information of the target object within the first preset time range, and then based on the behavior information, the first deposit time, and the deposit frequency of the target object, the first feature vector is determined. Thus, based on the first feature vector, the object type of the target object is identified, where the behavior information at least includes the geographical location information of at least one scene where the target object has stayed, the first deposit time represents the deposit time of the target object's last deposit, the first feature vector is used to characterize the deposit characteristics of the target object, and the object type represents whether the target object is an object whose requested deposit time is greater than the preset time.

[0044] It is easy to notice that in the above process, by determining the first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object, the richness of the first feature vector corresponding to the target object is realized, so that the deposit characteristics of the target object can be better reflected, and further, the object type of the target object identified based on the first feature vector is more accurate, avoiding the problem of low recognition accuracy for objects with low consumption amount and low purchase frequency in the prior art when identifying the object type only based on the consumption behavior of the target object.

[0045] It can be seen that the solution provided by the present application achieves the purpose of identifying the object type of the target object based on the behavior information of the target object, the deposit time of the last deposit, and the deposit frequency, thereby realizing the technical effect of improving the recognition accuracy, and further solving the technical problem of low recognition accuracy of the related methods for identifying customer types in the prior art.

[0046] In an alternative embodiment, in the process of determining the first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object, the object recognition system may sort multiple geographical location information based on the stay order to obtain a target sequence, and then determine the transition probability between the positions corresponding to any two geographical location information of the target object among the multiple geographical location information based on the target sequence, to obtain multiple target transition probabilities, so as to determine the first feature vector based on the multiple target transition probabilities, the first deposit time, and the deposit frequency of the target object. Among them, the behavior information further includes the stay order corresponding to at least one scene where the target object has stayed.

[0047] Optionally, as Figure 2 shown, based on the stay order and behavior information of the target object, the POI sequence corresponding to the target object (i.e., the aforementioned target sequence) can be obtained. For example, user test ={P a (lon a , lat a ), P b (lon b , lat b ), … PN(lonN, latN), where useri represents the POI sequence of the target object test, and Pa, Pb, … PN represent multiple geographical location information corresponding to the target object, such as the geographical location information of scenes such as restaurants, hotels, beaches, banks, cinemas, parks, etc. lon N represents the longitude of position P N , and lat N represents the latitude of position P N .

[0048] Furthermore, based on the POI sequence, the probability that the target object transfers from the x-th type of POI (the position corresponding to the geographical location information) to the y-th type of POI can be determined. For example, the POI sequence corresponding to the target object is {P 1 , P 2 , P 4 , P 7 , P 1 , P 2 , P 5 , P 6 , P 2 , P 1}. There are ten POIs in this sequence and 9 conversions. The conversion from P 1 to P 2 occurs 2 times. Therefore, I 12 =2 / 9, where I 12 represents the transition probability from the 1st POI to the 2nd POI.

[0049] Further, based on the foregoing method for calculating the transition probability, the transition probability between each geographical location information and other geographical location information in all geographical location information can be calculated, so that a plurality of target transition probabilities can be obtained, and a first feature vector can be determined based on the plurality of target transition probabilities, the first deposit time, and the deposit frequency of the target object.

[0050] It should be noted that by calculating the target transition probability, the full utilization of the spatio-temporal behavior data of the target object is realized, and further, the feature vector corresponding to the target object is expanded.

[0051] In an alternative embodiment, in the process of determining the first feature vector based on the plurality of target transition probabilities, the first deposit time, and the deposit frequency of the target object, the object recognition system may generate a first target transition probability matrix based on the plurality of target transition probabilities, and then perform dimensionality reduction processing on the first target transition probability matrix to obtain a second target transition probability matrix, so as to determine the first feature vector based on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object.

[0052] Optionally, after calculating the transition probability between each geographical location information and other geographical location information in all geographical location information, it may also be determined that the transition probability between each geographical location information and itself is 0, e.g., I 11 = 0, so that the sum of the number of target transition probabilities and the number of transition probabilities between each geographical location information and itself is equal to the square of the number of POI types in the POI sequence. For example, in the foregoing POI sequence {P 1 , P 2 , P 4 , P 7 , P 1 , P 2 , P 5 , P 6 , P 2 , P 1}, if the number of POI types in the POI sequence is 7, then the target transition probability plus the transition probability between each geographical location information and itself has 49.

[0053] Further, as Figure 2 shown, the object recognition system may generate an N-order square matrix G (i.e., the foregoing first target transition probability matrix) based on the target transition probabilities corresponding to N types of POIs, as follows:

[0054]

[0055] Among them, G test represents the first target transition probability matrix corresponding to the target object test.

[0056] Furthermore, as Figure 2 shown, the object recognition system can reduce the first target transition probability matrix to one dimension to obtain the second target transition probability matrix g test =(I 11 , I 12 , I 13 , …, I 1N , I 21 , I 22 , I 23 , …, I 2N , I 31 , I 32 , I 33 , …, I 3N , I N1 , I N2 , I N3 , …, I NN ). Thus, the first eigenvector X can be determined based on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object test =(I 11 , I 12 , I 13 , …, I 1N , I 21 , I 22 , I 23 , …, I 2N , I 31 , I 32 , I 33 , …, I 3N , I N1 , I N2 , I N3 , …, I NN , r test , f test ), where r test represents the first deposit time corresponding to the target object test, and f test represents the deposit frequency of the target object test. In the first eigenvector X test , there are a total of N×N + 2 parameters, representing N×N + 2 features of the target object.

[0057] It should be noted that by determining the second target transition probability matrix based on multiple target transition probabilities, the first eigenvector formed by each target object can maintain the same structure, thereby improving the accuracy of object type recognition of the target object.

[0058] In an alternative embodiment, in the process of determining the first feature vector based on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object, the object recognition system may normalize the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object to obtain the first feature vector.

[0059] Optionally, as Figure 2 shown, the object recognition system may normalize the previously obtained g test , r test and f test all to determine the final first feature vector.

[0060] It should be noted that by normalizing the data, the obtained first feature vector is more convenient for analysis, thereby improving the recognition efficiency.

[0061] In an alternative embodiment, before identifying the object type of the target object based on the first feature vector, the object recognition system may construct a training set based on the second feature vector and the object types of historical objects, and then train a preset model based on the training set to obtain a target preset model, where the second feature vector is used to characterize the deposit features of historical objects, and the target preset model is used to identify the object type of the target object based on the first feature vector.

[0062] Optionally, the object recognition system may construct a training set based on the instances of objects of the object types that have been determined currently (i.e., the aforementioned historical objects) where X i represents the second feature vector corresponding to historical object i, y i characterizes the object type of historical object i, and y i ∈ {0, 1}, when y i = 1, it represents that historical object i is a non-long-term deposit user, and when y i = 0, it represents that historical object i is a long-term deposit user.

[0063] Furthermore, taking the training set as the input data of the deep forest (i.e., the preset model), training the deep forest, so as to obtain the trained deep forest, that is, the target preset model. After that, as Figure 2 shown, it is possible to identify the object type of the target object based on the trained deep forest to determine whether the target object is a potential long-term deposit customer of the bank. Among them, the deep forest is the superposition of different types of forests, representing learning using a cascaded forest structure, and the methods used in the deep forest can be logistic regression, random forest, and decision tree.

[0064] It should be noted that by training a model based on data from historical objects and identifying user types based on the trained model, it is easy to improve recognition accuracy, thereby more accurately identifying long-term deposit customers and, at the same time, improving recognition efficiency.

[0065] In an optional embodiment, before constructing a training set based on the second feature vector and the object type of the historical object, the object recognition system may obtain historical behavior information of the historical object within a second preset time range, thereby determining the second feature vector based on the historical behavior information, the second deposit time, and the deposit frequency of the historical object, wherein the second deposit time represents the deposit time of the last deposit of the historical object.

[0066] Optionally, the method for determining the second feature vector based on the historical behavior information, the second deposit time, and the deposit frequency of the historical object is the same as the method for determining the first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object, so it is not repeated here. The second feature vector of the determined historical object i can be expressed as X i =(I 11 ,I 12 ,I 13 ,…,I 1N ,I 21 ,I 22 ,I 23 ,…,I 2N ,I 31 ,I 32 ,I 33 ,…,I 3N ,I N1 ,I N2 ,I N3 ,…,I NN ,r i ,f i ).

[0067] It should be noted that since the geographic location of each target object's behavior is associated with the POI and the order of stay, the difference in the POI sequence of each target object is large enough, so that the second feature vector determined based on the historical behavior information, the second deposit time and the deposit frequency of the historical object has diversity, which facilitates the training of the preset model and further improves the recognition accuracy of the trained preset model.

[0068] It can be seen that in the present application, by proposing an STP-DF (spatial transition probability and Deep Forest algorithm) model, the spatio-temporal behavior data of deposit customers and the computing power of the deep forest are fully utilized, so that it has a higher recognition accuracy compared with existing related methods and can more accurately identify long-term deposit customers. At the same time, the running time of this method is shorter, and it can complete the identification of long-term deposit customers more quickly, helping the bank increase deposits. Thus, the purpose of identifying the object type of the target object based on the behavior information of the target object, the deposit time of the last deposit, and the deposit frequency is achieved, the technical effect of improving the recognition accuracy is realized, and furthermore, the technical problem of low recognition accuracy of related methods for identifying customer types in the prior art is solved.

[0069] Embodiment 2

[0070] According to an embodiment of the present invention, an embodiment of an object recognition device is provided, wherein Figure 3 is a schematic diagram of an optional object recognition device according to an embodiment of the present invention, as Figure 3 shown, the device includes:

[0071] An acquisition module 301, configured to acquire the behavior information of the target object within a first preset time range, where the behavior information at least includes the geographical location information of at least one scene where the target object has stayed;

[0072] A determination module 302, configured to determine a first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object, where the first deposit time represents the deposit time of the target object's last deposit, and the first feature vector is used to represent the deposit characteristics of the target object;

[0073] An identification module 303, configured to identify the object type of the target object based on the first feature vector, where the object type represents whether the target object is an object whose requested deposit time is greater than the preset time.

[0074] It should be noted that the above acquisition module 301, determination module 302, and identification module 303 correspond to steps S101 to S103 in the above embodiment. The examples and application scenarios implemented by the three modules and the corresponding steps are the same, but are not limited to the content disclosed in the above embodiment 1.

[0075] Optionally, the determination module further includes: a sorting module, configured to sort multiple pieces of geographical location information based on the staying order to obtain a target sequence, where the behavior information further includes the staying order corresponding to at least one scene where the target object has stayed; a first determination sub-module, configured to determine the transition probability between any two geographical location information corresponding to the target object among the multiple pieces of geographical location information based on the target sequence to obtain multiple target transition probabilities; a second determination sub-module, configured to determine a first feature vector based on the multiple target transition probabilities, the first deposit time, and the deposit frequency of the target object.

[0076] Optionally, the second determination sub-module further includes: a generation module, configured to generate a first target transition probability matrix based on the multiple target transition probabilities; a first processing module, configured to perform dimensionality reduction processing on the first target transition probability matrix to obtain a second target transition probability matrix; a third determination sub-module, configured to determine a first feature vector based on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object.

[0077] Optionally, the third determination sub-module further includes: a second processing module, configured to perform normalization processing on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object to obtain a first feature vector.

[0078] Optionally, the object recognition device further includes: a construction module, configured to construct a training set based on the second feature vector and the object type of the historical object, where the second feature vector is used to characterize the deposit feature of the historical object; a training module, configured to train a preset model based on the training set to obtain a target preset model, where the target preset model is used to identify the object type of the target object based on the first feature vector.

[0079] Optionally, the object recognition device further includes: an acquisition sub-module, configured to acquire the historical behavior information of the historical object within a second preset time range; a fourth determination sub-module, configured to determine a second feature vector based on the historical behavior information, the second deposit time, and the deposit frequency of the historical object, where the second deposit time represents the deposit time of the historical object's last deposit.

[0080] Embodiment 3

[0081] According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, in which a computer program is stored, where the computer program is configured to execute the above-mentioned object recognition method when running.

[0082] Embodiment 4

[0083] According to another aspect of the embodiments of the present invention, there is also provided an electronic device, where Figure 4It is a schematic diagram of an optional electronic device according to an embodiment of the present invention, as Figure 4 shown, the electronic device includes one or more processors; a memory for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement for running the program, wherein the program is set to execute the above-mentioned object recognition method when running.

[0084] Embodiment 5

[0085] According to another aspect of the embodiments of the present invention, there is also provided a computer program product, including computer programs / instructions, which implement the above-mentioned object recognition method when executed by a processor.

[0086] The above serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0087] In the above embodiments of the present invention, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0088] In the several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only illustrative. For example, the division of units can be a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of units or modules can be in electrical or other forms.

[0089] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] In addition, the functional units in the various embodiments of the present invention can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0091] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs.

[0092] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for identifying an object, characterized in that, it includes: Obtain the behavior information of the target object within a first preset time range, where the behavior information at least includes the geographical location information of at least one scene where the target object has stayed; Based on the behavior information, the first deposit time, and the deposit frequency of the target object, determine a first feature vector, where the first deposit time represents the deposit time of the target object's last deposit, and the first feature vector is used to characterize the deposit characteristics of the target object; Based on the first feature vector, identify the object type of the target object, where the object type represents whether the target object is an object whose requested deposit time is greater than the preset time; Based on the behavior information, the first deposit time, and the deposit frequency of the target object, determining the first feature vector includes: Based on the stay order, sort multiple geographical location information to obtain a target sequence, where the behavior information further includes the stay order corresponding to at least one scene where the target object has stayed; Based on the target sequence, determine the transition probability between the positions corresponding to any two geographical location information among the multiple geographical location information of the target object, and obtain multiple target transition probabilities; Based on the multiple target transition probabilities, the first deposit time, and the deposit frequency of the target object, determine the first feature vector.

2. The method according to claim 1, characterized in that, Based on the multiple target transition probabilities, the first deposit time, and the deposit frequency of the target object, determining the first feature vector includes: Generate a first target transition probability matrix based on the multiple target transition probabilities; Perform dimensionality reduction processing on the first target transition probability matrix to obtain a second target transition probability matrix; Based on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object, determine the first feature vector.

3. The method according to claim 2, characterized in that, Based on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object, determining the first feature vector includes: Perform normalization processing on the second target transition probability matrix, the first deposit time, and the deposit frequency of the target object to obtain the first feature vector.

4. The method according to claim 1, characterized in that, Before identifying the object type of the target object based on the first feature vector, the method further includes: Construct a training set based on the second feature vector and the object types of historical objects, where the second feature vector is used to characterize the deposit characteristics of the historical objects; Train a preset model based on the training set to obtain a target preset model, where the target preset model is used to identify the object type of the target object based on the first feature vector.

5. The method according to claim 4, characterized in that, Before constructing a training set based on the second feature vector and the object types of historical objects, the method further includes: Obtain the historical behavior information of historical objects within a second preset time range; Determine the second feature vector based on the historical behavior information, the second deposit time, and the deposit frequency of the historical object, where the second deposit time represents the deposit time of the last deposit of the historical object.

6. An identification device for an object, characterized in that it includes: An acquisition module for acquiring the behavior information of a target object within a first preset time range, where the behavior information at least includes the geographical location information of at least one scene where the target object has stayed; A determination module for determining a first feature vector based on the behavior information, the first deposit time, and the deposit frequency of the target object, where the first deposit time represents the deposit time of the last deposit of the target object, and the first feature vector is used to characterize the deposit characteristics of the target object; An identification module for identifying the object type of the target object based on the first feature vector, where the object type represents whether the target object is an object whose requested deposit time is greater than the preset time; The determination module further includes: A sorting module for sorting a plurality of geographical location information based on the stay order to obtain a target sequence, where the behavior information further includes the stay order corresponding to at least one scene where the target object has stayed; A first determination sub-module for determining the transition probability between the positions corresponding to any two geographical location information of the target object among the plurality of geographical location information based on the target sequence to obtain a plurality of target transition probabilities; A second determination sub-module for determining the first feature vector based on the plurality of target transition probabilities, the first deposit time, and the deposit frequency of the target object.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, where the computer program is set to execute the object identification method described in any one of claims 1 to 5 when running.

8. An electronic device, characterized in that The electronic device includes one or more processors; A memory for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement a program for running, where the program is set to execute the object identification method described in any one of claims 1 to 5 when running.

9. A computer program product, including a computer program / instructions, characterized in that The computer program / instructions implement the object identification method described in any one of claims 1 to 5 when executed by a processor.

Images