Abnormal object determination method and device, equipment, medium and product

Abstract

The invention discloses an abnormal object determination method and device, equipment, a medium and a product. The method comprises the following steps: acquiring multiple evaluation associated data corresponding to multiple to-be-evaluated resource supply objects associated with a target object: for the multiple to-be-evaluated resource supply objects, performing feature extraction on the corresponding evaluation associated data according to the data type of each evaluation associated data corresponding to the to-be-evaluated resource supply objects, obtaining an assessment association feature corresponding to the to-be-assessed resource supply object; and determining an abnormal resource supply object list corresponding to the target object according to an abnormal evaluation model obtained by pre-training and the evaluation association features corresponding to the plurality of to-be-evaluated resource supply objects. According to the technical scheme, intellectualization and automation of anomaly evaluation of the resource supply object are realized, and the anomaly evaluation efficiency and accuracy of the resource supply object are improved.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to an abnormal object determination method, device, equipment, medium and product. Background Art

[0002] Supply chain management has become an important part of the core competitiveness of enterprises. As a key link in the supply chain, the stability and reliability of resource supply objects directly affect the efficiency and stability of the entire supply chain. Therefore, assessing and managing the risks of resource supply objects is an important part of supply chain management.

[0003] In the related art, the evaluation and management of resource supply objects usually rely on manual experience, and the resource supply objects are evaluated manually. This object evaluation method consumes too much manpower and time costs, has low object evaluation efficiency, and cannot objectively evaluate resource supply objects. The accuracy of abnormal evaluation is low, which in turn affects the execution progress of subsequent projects. Summary of the invention

[0004] The present invention provides a method, device, equipment, medium and product for determining an abnormal object, so as to realize the intelligent and automatic abnormal evaluation of resource supply objects, and improve the abnormal evaluation efficiency and accuracy of resource supply objects.

[0005] According to one aspect of the present invention, a method for determining an abnormal object is provided, the method comprising:

[0006] Acquire multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with the target object: wherein the resource supply objects to be evaluated provide relevant resources for the target object; and the evaluation-related data are data required for performing abnormal evaluation on the resource supply objects to be evaluated;

[0007] For the plurality of resource supply objects to be evaluated, feature extraction is performed on the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain evaluation-related features corresponding to the resource supply object to be evaluated;

[0008] Determine a list of abnormal resource supply objects corresponding to the target object according to a pre-trained abnormal assessment model and assessment association features corresponding to a plurality of resource supply objects to be assessed;

[0009] The abnormal resource supply object list includes an object identifier corresponding to at least one abnormal resource supply object.

[0010] According to another aspect of the present invention, there is provided an abnormal object determination device, the device comprising:

[0011] A data acquisition module, used to acquire multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with a target object: wherein the resource supply objects to be evaluated provide relevant resources for the target object; and the evaluation-related data are data required for performing abnormal evaluation on the resource supply objects to be evaluated;

[0012] A feature extraction module, for extracting features of the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain evaluation-related features corresponding to the resource supply object to be evaluated;

[0013] An abnormal assessment module is used to determine a list of abnormal resource supply objects corresponding to the target object based on a pre-trained abnormal assessment model and assessment association features corresponding to multiple resource supply objects to be assessed; wherein the abnormal resource supply object list includes an object identifier corresponding to at least one abnormal resource supply object.

[0014] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the abnormal object determination method described in any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the abnormal object determination method described in any embodiment of the present invention when executed.

[0019] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the abnormal object determination method according to any embodiment of the present invention is implemented.

[0020] The technical solution of the embodiment of the present invention obtains multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with the target object. Further, for the multiple resource supply objects to be evaluated, feature extraction is performed on the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain the evaluation-related features corresponding to the resource supply objects to be evaluated. Further, according to the pre-trained abnormal evaluation model and the evaluation-related features corresponding to the multiple resource supply objects to be evaluated, a list of abnormal resource supply objects corresponding to the target object is determined, which solves the problems in the related art that the object evaluation method is too labor-intensive and time-consuming, the object evaluation efficiency is low, and the abnormal evaluation accuracy is low, realizes the intelligent and automated abnormal evaluation of resource supply objects, improves the abnormal evaluation efficiency and accuracy of resource supply objects, and realizes the effect of timely and precise management of resource supply objects.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 is a flowchart of a method for determining an abnormal object provided according to Embodiment 1 of the present invention;

[0024] Figure 2 is a schematic diagram of an architecture of an abnormal object determination system provided according to Embodiment 1 of the present invention;

[0025] Figure 3 is a flowchart of a method for determining an abnormal object provided according to Embodiment 2 of the present invention;

[0026] Figure 4 is a schematic diagram of the structure of an abnormal object determination device provided according to Embodiment 3 of the present invention;

[0027] Figure 5 It is a structural schematic diagram of an electronic device for implementing the abnormal object determination method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0029] 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 are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, 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 "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] Embodiment 1

[0031] Figure 1 This is a flowchart of an abnormal object determination method provided by the first embodiment of the present invention. This embodiment is applicable to the case of performing abnormal evaluation on resource supply objects. The method can be executed by an abnormal object determination device. The abnormal object determination device can be implemented in the form of hardware and / or software. The abnormal object determination device can be configured in a terminal and / or a server. Figure 1 As shown, the method includes:

[0032] S110: Acquire multiple pieces of evaluation-related data corresponding to multiple to-be-evaluated resource supply objects associated with the target object.

[0033] The target object may be an object for which the resource supply object needs to be evaluated. Optionally, the target object may be an organization or an individual. An organization is a social entity composed of several individuals or groups with common goals and certain boundaries. The resource supply object to be evaluated may be a resource supply object for which an abnormal evaluation is to be performed. The resource supply object to be evaluated provides relevant resources for the target object. The resource supply object may be an organization or an individual that provides the required resources for an enterprise, and these resources are indispensable for the enterprise to carry out normal production, services and other activities. The evaluation-related data is the data required for the abnormal evaluation of the resource supply object to be evaluated. The evaluation-related data may be used to characterize the basic situation of the resource supply object to be evaluated and the object development situation in the historical time interval from the historical moment to the current moment. Optionally, the evaluation-related data includes object basic information, object development data and object assessment and evaluation information. Among them, the object basic information may include object identification (including object name and object code, etc.), object establishment date, object type, object business scope and object business period, etc. The object development situation data may be used to characterize the object development situation of the resource supply object to be evaluated in the historical time interval from the historical moment to the current moment. The object development data may include the number of available personnel, the number of objects served, the object business status data, and the object business risk data, etc. The object assessment information may include the assessment score and the assessment description information, etc. The assessment-related data may be data represented in various forms, and optionally, data represented in the form of key-value pairs.

[0034] As an optional implementation of this embodiment, multiple resource supply objects to be evaluated can be predetermined. Furthermore, for multiple resource supply objects to be evaluated, information associated with the resource supply object to be evaluated can be queried in a relevant database based on the object name of the resource supply object to be evaluated. Furthermore, the information obtained from the query can be saved in the form of a key-value pair, and the saved information can be used as evaluation-related data corresponding to the resource supply object to be evaluated. Exemplarily, assuming that the number of disposable personnel corresponding to the resource supply object A to be evaluated is 120 obtained from the relevant database, the number of disposable personnel (employee_num) can be used as the key field, and the specific value 120 can be used as the value field corresponding to the key field.

[0035] S120 . For a plurality of resource supply objects to be evaluated, extract features of corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain evaluation-related features corresponding to the resource supply object to be evaluated.

[0036] Among them, the data type refers to an abstract representation used to define the properties and structure of data in storage units such as variables and parameters. Different data types correspond to different data processing methods. Optionally, the data types include sequence numerical type, non-sequence numerical type and text type. The sequence numerical type can be understood as a data type in which the value in the data changes with time. Exemplarily, the evaluation-related data of the sequence numerical type can include the number of disposable personnel, the number of objects served, the object business status data and the object business risk data. The non-sequence numerical type can be understood as a data type in which the value in the data does not change with time. Exemplarily, the evaluation-related data of the non-sequence numerical type can include assessment scores. The text type can be understood as a data type represented in text form. Exemplarily, the evaluation-related data of the text type can include evaluation description information. The evaluation-related feature can be understood as the key feature associated with the abnormal object evaluation contained in the evaluation-related data. In this embodiment, the evaluation-related feature can include the data feature corresponding to each evaluation-related data.

[0037] In this embodiment, in order to effectively reduce the data dimension while retaining the associated features contained in the information to improve the efficiency and accuracy of abnormal object assessment, when obtaining multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated, the data type of each evaluation-related data corresponding to the resource supply object to be evaluated can be determined for each of the multiple resource supply objects to be evaluated. Furthermore, for the multiple evaluation-related data, feature extraction can be performed on the evaluation-related data based on the data type of the evaluation-related data to obtain data features corresponding to the evaluation-related data. Furthermore, the obtained data features corresponding to the multiple evaluation-related data can be used as evaluation-related features corresponding to the resource supply object to be evaluated.

[0038] As an optional implementation of this embodiment, for multiple resource supply objects to be evaluated, the feature extraction method corresponding to the corresponding evaluation-related data can be determined according to the data type corresponding to each evaluation-related data corresponding to the resource supply object to be evaluated. Further, for multiple evaluation-related data, feature extraction can be performed on the evaluation-related data according to the feature extraction method corresponding to the evaluation-related data, and data features corresponding to the evaluation-related data can be obtained. Further, the obtained data features corresponding to the multiple evaluation-related data can be used as the evaluation-related features corresponding to the resource supply object to be evaluated.

[0039] S130 , determining a list of abnormal resource supply objects corresponding to the target object according to the pre-trained abnormal assessment model and the assessment association features corresponding to the plurality of resource supply objects to be assessed.

[0040] Among them, the abnormal assessment model can be a model that takes the evaluation-related features corresponding to the resource supply object to be evaluated as an input object to perform abnormal assessment on the resource supply object to be evaluated based on the evaluation-related features. The abnormal assessment model can be a machine learning model, in which a machine learning algorithm for abnormality detection can be deployed. Optionally, the machine learning algorithm deployed in the abnormal assessment model includes an isolation forest algorithm. The abnormal assessment model can be obtained by training the machine learning model through the historical evaluation-related features corresponding to the sample resource supply object. It can be understood that the isolation forest algorithm is an abnormality detection algorithm based on a tree structure, which is an abnormality detection method for isolating abnormal points, thereby distinguishing them from normal points and judging their abnormalities. Different from the traditional abnormality detection algorithm, the isolation forest algorithm adopts a bottom-up method, that is, first constructing subtrees and then merging them into a forest (i.e., a set of isolated trees), so that abnormal points existing in the data can be efficiently detected. The abnormal resource supply object list can be understood as a list of abnormal resource supply objects that characterize the abnormal resource supply objects existing in the resource supply object to be evaluated. The abnormal resource supply object list can include an object identifier corresponding to at least one abnormal resource supply object. The object identifier can be information used to identify the resource supply object. Optionally, the object identifier includes an object name, an object code and / or other information for distinguishing a resource supply object, etc. An abnormal resource supply object can be understood as a resource supply object whose provided resources cannot meet normal standards and preset requirements due to some reason.

[0041] It should be noted that before applying the anomaly assessment model provided in this embodiment, the pre-constructed machine learning model can be trained in a supervised or unsupervised manner. Before training the machine learning model, multiple training sample data can be constructed to train the model based on the training sample data. Optionally, the training process of the anomaly assessment model can be: obtaining multiple training sample data; wherein the training sample data includes historical evaluation association features corresponding to the sample resource supply object at a historical moment; training the pre-constructed machine learning model according to the multiple training sample data to obtain the anomaly assessment model.

[0042] Among them, the sample resource supply object can be a historical resource supply object associated with the target object, that is, a resource supply object that has supplied relevant resources to the target object at a historical moment; or, it can also be a historical resource supply object associated with other objects, that is, a resource supply object that has supplied relevant resources to other objects at a historical moment, etc. The historical evaluation association feature can be determined based on the data features of the historical evaluation association data corresponding to the sample resource supply object at a historical moment. The machine learning model can be a model that deploys a machine learning algorithm and the model parameters are initial values ​​or default values. Optionally, the machine learning algorithm deployed in the machine learning model includes an isolation forest algorithm. It should be noted that the advantage of using a machine learning model that deploys an isolation forest algorithm as an abnormal evaluation model is that the machine learning model of the isolation forest algorithm is a lightweight model, which can accurately determine abnormal resource supply objects based on the application of fewer computing resources, and the model has high real-time performance and higher feasibility.

[0043] As an optional implementation of the first embodiment of the present invention, multiple historical evaluation associated data corresponding to multiple sample resource supply objects may be obtained. Furthermore, for multiple sample resource supply objects, feature extraction may be performed on the corresponding historical evaluation associated data according to the data type corresponding to each item of historical evaluation associated data corresponding to the sample resource supply object, so as to obtain the historical evaluation associated features corresponding to the sample resource supply object, and use the historical evaluation associated features corresponding to the sample resource supply object as training sample data. Furthermore, multiple training sample data may be obtained. Furthermore, a pre-built machine learning model that deploys an isolation forest algorithm may be trained in an unsupervised manner based on multiple training sample data. Furthermore, the trained machine learning model may be used as an abnormality assessment model.

[0044] In this embodiment, when the evaluation-related features corresponding to multiple resource supply objects to be evaluated are obtained, the evaluation-related features corresponding to the multiple resource supply objects to be evaluated can be input into the pre-trained abnormal evaluation model to process the evaluation-related features based on the abnormal evaluation model, and obtain the abnormal evaluation results corresponding to each resource supply object to be evaluated. Further, the list of abnormal resource supply objects corresponding to the target object can be determined according to the abnormal evaluation results corresponding to the multiple resource supply objects to be evaluated. The advantage of such a setting is that the resource supply objects with abnormal conditions can be discovered in time, which is convenient for timely analysis and processing of abnormal resource supply objects, and is conducive to the precise and timely management of resource supply objects. It should also be noted that the advantage of using the evaluation-related features corresponding to the resource supply objects to be evaluated for abnormal evaluation is that, on the basis of fully extracting and mining the features contained in the evaluation-related data, the abnormal object evaluation is performed based on the mined features, which improves the reliability of the abnormal evaluation of the resource supply objects and improves the accuracy of the abnormal evaluation.

[0045] Optionally, a list of abnormal resource supply objects corresponding to the target object is determined based on a pre-trained abnormal assessment model and assessment-related features corresponding to multiple resource supply objects to be assessed, including: for multiple resource supply objects to be assessed, inputting the assessment-related features corresponding to the resource supply objects to be assessed into the pre-trained abnormal assessment model to obtain object abnormality scores corresponding to the resource supply objects to be assessed; sorting the multiple resource supply objects to be assessed in descending order of the object abnormality scores, and selecting a preset number of resource supply objects to be assessed with top rankings as abnormal resource supply objects; and constructing a list of abnormal resource supply objects based on the multiple abnormal resource supply objects.

[0046] Among them, the object anomaly score can be understood as the score obtained after evaluating the degree of abnormality of the resource supply object to be evaluated based on the abnormal evaluation model. Exemplarily, the scoring interval of the object anomaly score can be 0-100 points. Among them, the object anomaly score is 0 points, indicating that the degree of abnormality of the resource supply object to be evaluated is the lowest; the object anomaly score is 100 points, indicating that the degree of abnormality of the resource supply object to be evaluated is the highest. The preset number can be used to limit the number of selected objects. The preset number can be determined based on the number of resource supply objects to be evaluated. Optionally, the preset number can be one percent, two percent or three percent of the total number of resource supply objects to be evaluated. Exemplarily, assuming that the preset number is one percent of the total number of resource supply objects to be evaluated, when the number of resource supply objects to be evaluated is 100, the preset number can be 1; when the number of resource supply objects to be evaluated is 200, the preset number can be 2.

[0047] As an optional implementation of this embodiment, for multiple resource supply objects to be evaluated, the evaluation-related features corresponding to the resource supply objects to be evaluated can be input into the pre-trained abnormal evaluation model. Then, the evaluation-related features can be processed based on the abnormal evaluation model, and the object abnormality score corresponding to the resource supply object to be evaluated can be output. Further, in the case of obtaining the object abnormality scores corresponding to multiple resource supply objects to be evaluated, the multiple resource supply objects to be evaluated can be sorted in order from high to low according to the object abnormality scores. Further, a preset number of resource supply objects to be evaluated with high rankings can be selected from the sorted multiple resource supply objects to be evaluated, and at least one selected resource supply object to be evaluated can be used as an abnormal resource supply object. Further, the object identifier corresponding to at least one abnormal resource supply object can be displayed in the form of a table, and a list of abnormal resource supply objects corresponding to the target object can be obtained.

[0048] It should be noted that in order to improve the model performance of the anomaly assessment model during the model application process, the anomaly assessment model can improve the anomaly detection accuracy of the anomaly assessment model for abnormal resource supply objects. In the process of applying the anomaly assessment model, the anomaly assessment model can also be optimized. Optionally, the method further includes: updating the model parameters of the anomaly assessment model based on the assessment association features corresponding to the multiple resource supply objects to be assessed.

[0049] As an optional implementation of this embodiment, when an abnormal assessment model is obtained, the assessment-related features corresponding to the resource supply object to be assessed can be processed based on the abnormal assessment model to determine the abnormal resource supply object. After each preset time period, the assessment-related features corresponding to the multiple resource supply objects to be assessed that have been assessed can be obtained, and the model parameters of the abnormal assessment model can be corrected according to the multiple assessment-related features that have been obtained. Thus, the model performance can be continuously optimized during the model application process of the abnormal assessment model. Among them, the preset time period can be any time period, optionally, 1 month, 3 months or 6 months, etc.

[0050] In this embodiment, in order to enable the user to understand the abnormal resource supply objects associated with the target object more clearly and intuitively, after determining the list of abnormal resource supply objects corresponding to the target object, it also includes: displaying the list of abnormal resource supply objects; in response to an identification selection operation for an object identifier, determining the abnormal resource supply object corresponding to the selected object identifier, obtaining object association information corresponding to the abnormal resource supply object, and displaying the object association information.

[0051] The identification selection operation can be understood as an operation that triggers the selection of any identification. Optionally, the identification selection trigger operation includes but is not limited to at least one of a click operation, a long press operation, a touch operation, and a slide operation. Object association information can be understood as information that can characterize the basic situation and detailed situation of the resource supply object.

[0052] As an optional implementation of this embodiment, when a list of abnormal resource supply objects corresponding to the target object is obtained, the obtained list of abnormal resource supply objects can be displayed in the target interface, and the abnormal resource supply object list can include at least one object identifier corresponding to the abnormal resource supply object. Further, when an identifier selection operation for an object identifier is received, the abnormal resource supply object corresponding to the selected object identifier can be determined. Further, object association data corresponding to the abnormal resource supply object can be retrieved from the database, and the retrieved object association information can be displayed in the target interface.

[0053] For example, Figure 2 Schematic diagram of an abnormal object determination system provided by an embodiment of the present invention. The abnormal object determination system can be applied to any target organization that needs to evaluate resource supply objects. Figure 2 As shown, the abnormal object determination system includes a data storage layer, a data collection layer, a data processing layer, an abnormal evaluation layer and a front-end display layer. Among them, the data storage layer is used to store the data collected by the data collection layer and the list of resource supply objects maintained in the target organization, and store the public data of the resource supply objects collected by the data collection layer and the internal data about the resource supply objects included in the target organization in the local database in the data storage layer; the list of resource supply objects maintained in the target organization is stored in the file system in the data storage layer. The data collection layer is used to collect multiple evaluation-related data corresponding to the resource supply object. The multiple evaluation-related data include public data and internal data. Based on the data collection layer, the public data of the resource supply object from the relevant database can be collected, and the collected public data includes basic information, transaction data, business status data, etc. corresponding to the resource supply object. And, based on the data collection layer, the assessment score and evaluation description information corresponding to the resource supply object can be obtained from the internal database of the target organization. The data processing layer can be used to obtain multiple evaluation-related data corresponding to the resource supply object to be evaluated and to extract features of the acquired evaluation-related data to obtain evaluation-related features corresponding to the resource supply object to be evaluated. The abnormal evaluation layer can be used to evaluate abnormal resource supply objects based on the evaluation association features corresponding to the resource supply objects to be evaluated, and to update the model parameters of the isolation forest model used for abnormal evaluation. The front-end display layer can be used to display the abnormal resource supply object list and the detailed information of the abnormal resource supply objects.

[0054] The technical solution of the embodiment of the present invention obtains multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with the target object. Further, for the multiple resource supply objects to be evaluated, feature extraction is performed on the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain the evaluation-related features corresponding to the resource supply objects to be evaluated. Further, according to the pre-trained abnormal evaluation model and the evaluation-related features corresponding to the multiple resource supply objects to be evaluated, a list of abnormal resource supply objects corresponding to the target object is determined, which solves the problems in the related art that the object evaluation method is too labor-intensive and time-consuming, the object evaluation efficiency is low, and the abnormal evaluation accuracy is low, realizes the intelligent and automated abnormal evaluation of resource supply objects, improves the abnormal evaluation efficiency and accuracy of resource supply objects, and realizes the effect of timely and precise management of resource supply objects.

[0055] Embodiment 2

[0056] Figure 3 It is a flow chart of an abnormal object determination method provided in Example 2 of the present invention. On the basis of the foregoing embodiment, the method for determining the evaluation-related features is further refined. The feature extraction method corresponding to the evaluation-related data is determined according to the data type of the evaluation-related data. Further, feature extraction is performed on the corresponding evaluation-related data according to the feature extraction method to obtain data features corresponding to the evaluation-related data. Further, based on the data features corresponding to the multiple evaluation-related data, the evaluation-related features corresponding to the resource supply object to be evaluated are determined. Then, an abnormal evaluation is performed on the resource supply object to be evaluated based on the evaluation-related features. The specific implementation method can refer to the technical solution of this embodiment. Among them, the technical terms that are the same or similar to those in the above-mentioned embodiments are not repeated here.

[0057] like Figure 3 As shown, the method includes:

[0058] S210: Acquire multiple pieces of evaluation-related data corresponding to multiple to-be-evaluated resource supply objects associated with the target object.

[0059] S220 . For a plurality of resource supply objects to be evaluated, determine a feature extraction method corresponding to the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated.

[0060] Among them, the feature extraction method can be understood as the method used to extract features from relevant data. For evaluation-related data of different data types, the data features required for abnormal evaluation are different. Therefore, different feature extraction methods can be used to extract features from relevant evaluation-related data to extract data features required for abnormal evaluation of resource supply objects to be evaluated. Optionally, feature extraction methods include similarity determination methods, vectorization processing methods, and direct extraction methods. The similarity determination method can be a method for determining the similarity between currently acquired evaluation-related data and historical evaluation-related data, and the determined similarity can be used to characterize the degree of change of the currently acquired evaluation-related data relative to the historical evaluation-related data. The vectorization processing method can be a method for performing semantic analysis on evaluation-related data to extract word vectors from the data. The direct extraction method can be a method of directly using the acquired evaluation-related data as corresponding data features.

[0061] In this embodiment, there may be multiple ways to determine the feature extraction method corresponding to the evaluation association data according to the data type of the evaluation association data. Optionally, the method may be determined according to preset type-method mapping information.

[0062] As an optional implementation of this embodiment, the data type corresponding to the evaluation-related data can be predetermined, and the feature extraction method corresponding to each data type can be determined. Further, type-method mapping information can be constructed based on each data type and its corresponding feature extraction method. Further, for multiple resource supply objects to be evaluated, the data type of each evaluation-related data corresponding to the resource supply object to be evaluated can be determined respectively. Further, for each evaluation-related data, the feature extraction method corresponding to the evaluation-related data can be determined based on the pre-constructed type-method mapping information and the data type of the evaluation-related data.

[0063] Exemplarily, the data type includes a sequence numerical type. For each evaluation-related data, when it is determined that the data type of the evaluation-related data is a sequence numerical type, the feature extraction method corresponding to the sequence numerical type can be determined as a similarity determination method. Furthermore, feature extraction can be performed on the evaluation-related data under the data type according to the similarity determination method to obtain corresponding data features.

[0064] Exemplarily, the data type includes a non-sequential numerical type. For each evaluation-related data, when it is determined that the data type of the evaluation-related data is a non-sequential numerical type, it can be determined that the feature extraction method corresponding to the non-sequential numerical type is a direct extraction method. Then, feature extraction is performed on the evaluation-related data under the data type according to the direct extraction method to obtain corresponding data features.

[0065] Exemplarily, the data type includes a text type. For each evaluation-related data, when it is determined that the data type of the evaluation-related data is a text type, the feature extraction method corresponding to the text type can be determined to be a vectorization processing method. Furthermore, feature extraction can be performed on the evaluation-related data under the data type according to the vectorization processing method to obtain corresponding data features.

[0066] S230 . For each item of evaluation-related data, extract features of the evaluation-related data based on a feature extraction method corresponding to the evaluation-related data to obtain data features corresponding to the evaluation-related data.

[0067] The data features may be understood as valid features included in the corresponding evaluation-related data and associated with the abnormal evaluation. Optionally, the data features include data similarity, text vectors and / or evaluation-related data corresponding to the evaluation-related data.

[0068] In this embodiment, while respectively determining the feature extraction method for each evaluation-related data corresponding to the resource supply object to be evaluated, for each evaluation-related data, feature extraction can be performed on the evaluation-related data according to the feature extraction method corresponding to the evaluation-related data to obtain data features corresponding to the evaluation-related data.

[0069] Optionally, the data type includes a sequence numerical type; the feature extraction method includes a similarity determination method; feature extraction is performed on the evaluation-related data based on the feature extraction method corresponding to the evaluation-related data to obtain data features corresponding to the evaluation-related data, including: obtaining historical evaluation-related data corresponding to the evaluation-related data; processing the historical evaluation-related data and the evaluation-related data according to a preset similarity determination method to obtain data similarity corresponding to the evaluation-related data, and using the data similarity as the data features corresponding to the evaluation-related data.

[0070] Wherein, the historical evaluation associated data is the evaluation associated data corresponding to the resource supply object to be evaluated at any historical moment. Exemplarily, assuming that the evaluation associated data to be feature extracted is the number of disposable personnel of the resource supply object to be evaluated in December this year, the historical evaluation associated data corresponding to the evaluation associated data can be the number of disposable personnel corresponding to the resource supply object to be evaluated at any historical moment last year. The similarity determination method can include any method that can determine the similarity between at least two types of data, optionally including calculating the Jaccard similarity coefficient, calculating the cosine similarity, calculating the Manhattan distance or calculating the Pearson correlation coefficient, etc. Wherein, the data similarity can be used to measure the similarity and / or difference between related data. In this embodiment, the data similarity can be used to characterize the degree of data change of the evaluation associated data relative to the historical evaluation associated data. Exemplarily, the data similarity corresponding to the number of disposable personnel can be used to characterize the degree of change of the number of disposable personnel currently acquired relative to the number of disposable personnel corresponding to the historical moment.

[0071] As an optional implementation manner of the present embodiment, for evaluation-related data whose data type is a sequence numerical type, historical evaluation-related data corresponding to the evaluation-related data can be obtained from a relevant database. Furthermore, the similarity coefficient between the evaluation-related data and the historical evaluation-related data can be determined according to a preset similarity determination method, and the similarity coefficient can be used as the data similarity corresponding to the evaluation-related data. Furthermore, the determined data similarity can be used as a data feature corresponding to the evaluation-related data. Exemplarily, continuing to refer to the above example, the Jaccard similarity coefficient between the currently acquired number of disposable personnel and the corresponding number of disposable personnel at a historical moment can be determined, and the determined similarity coefficient can be used as a data feature corresponding to the number of disposable personnel.

[0072] Optionally, the data type includes a text type; the feature extraction method includes a vectorization processing method; feature extraction is performed on the evaluation-related data based on the feature extraction method corresponding to the evaluation-related data to obtain data features corresponding to the evaluation-related data, including: processing the evaluation-related data according to a preset vectorization processing method to obtain a text vector corresponding to the evaluation-related data, and using the text vector as the data feature corresponding to the evaluation-related data.

[0073] Among them, the vectorization processing method can be any method that can convert unstructured data into structured numerical data. Optionally, the vectorization processing method includes but is not limited to word embedding, bag of words model, vector encoding method, etc. The text vector can be a numerical vector representation of text data. Optionally, the text vector is a word vector associated with the evaluation data. It can be understood that the text vector is numerical data, which is convenient for computers to process and analyze it; and the text vector can capture the semantic relationship between texts and improve the performance of natural language processing tasks.

[0074] As an optional implementation of the present embodiment, for evaluation-related data whose data type is textual, the evaluation-related data can be vectorized according to a preset vectorization processing method, and a text vector corresponding to the evaluation-related data can be obtained. Furthermore, the obtained text vector can be used as a data feature corresponding to the evaluation-related data. Exemplarily, assuming that the textual evaluation-related data is evaluation description information, the word vector corresponding to the evaluation description information can be determined by word embedding. Furthermore, the obtained word vector can be used as a data feature corresponding to the evaluation description information.

[0075] Optionally, the data type includes a non-sequence numeric type; the feature extraction method includes a direct extraction method; and feature extraction is performed on the evaluation-related data based on the feature extraction method to obtain data features corresponding to the evaluation-related data, including: using the evaluation-related data as the data features corresponding to the evaluation-related data.

[0076] As an optional implementation of this embodiment, for evaluation-related data whose data type is non-sequential numerical type, the change of the evaluation-related data does not change with the change of time. Furthermore, the numerical value of the evaluation-related data can be directly used as the data feature corresponding to the evaluation-related data. Exemplarily, assuming that the non-sequential numerical evaluation-related data is an assessment score, the specific score of the assessment score of the resource supply object to be assessed can be used as the data feature corresponding to the assessment score.

[0077] S240: constructing an evaluation-related feature corresponding to the resource supply object to be evaluated based on data features corresponding to the plurality of evaluation-related data.

[0078] In this embodiment, when the data features corresponding to each evaluation-related data are obtained, a multidimensional dataset can be constructed according to the multiple data features, and the constructed multidimensional dataset is used as the evaluation-related features corresponding to the resource supply object to be evaluated.

[0079] S250: Determine a list of abnormal resource supply objects corresponding to the target object according to the pre-trained abnormal assessment model and the assessment association features corresponding to the plurality of resource supply objects to be assessed.

[0080] The technical solution of the embodiment of the present invention, for multiple resource supply objects to be evaluated, determines the feature extraction method corresponding to the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated; for each evaluation-related data, extracts the evaluation-related data based on the feature extraction method corresponding to the evaluation-related data to obtain the data features corresponding to the evaluation-related data; constructs the evaluation-related features corresponding to the resource supply object to be evaluated based on the data features corresponding to the multiple evaluation-related data, and achieves the effect of fully mining the features contained in the evaluation-related data to enhance the comprehensiveness of the evaluation-related features. Furthermore, on the basis of fully extracting and mining the features contained in the evaluation-related data, abnormal object evaluation is performed based on the mined features, which improves the reliability of abnormal evaluation of resource supply objects and improves the accuracy of abnormal evaluation.

[0081] Embodiment 3

[0082] Figure 4 Schematic diagram of the structure of an abnormal object determination device provided by Embodiment 3 of the present invention. Figure 4 As shown, the device includes: a data acquisition module 310, a feature extraction module 320 and an abnormality assessment module 330.

[0083] Among them, the data acquisition module 310 is used to obtain multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with the target object: wherein the evaluation-related data is the data required for performing an abnormal evaluation on the resource supply objects to be evaluated; the resource supply objects to be evaluated supply relevant resources for the target object; the feature extraction module 320 is used to extract features of the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply objects to be evaluated for the multiple resource supply objects to be evaluated, so as to obtain evaluation-related features corresponding to the resource supply objects to be evaluated; the abnormal evaluation module 330 is used to determine a list of abnormal resource supply objects corresponding to the target object based on a pre-trained abnormal evaluation model and the evaluation-related features corresponding to the multiple resource supply objects to be evaluated; wherein the list of abnormal resource supply objects includes an object identifier corresponding to at least one abnormal resource supply object.

[0084] The technical solution of the embodiment of the present invention obtains multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with the target object. Further, for the multiple resource supply objects to be evaluated, feature extraction is performed on the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain the evaluation-related features corresponding to the resource supply objects to be evaluated. Further, according to the pre-trained abnormal evaluation model and the evaluation-related features corresponding to the multiple resource supply objects to be evaluated, a list of abnormal resource supply objects corresponding to the target object is determined, which solves the problems in the related art that the object evaluation method is too labor-intensive and time-consuming, the object evaluation efficiency is low, and the abnormal evaluation accuracy is low, realizes the intelligent and automated abnormal evaluation of resource supply objects, improves the abnormal evaluation efficiency and accuracy of resource supply objects, and realizes the effect of timely and precise management of resource supply objects.

[0085] Optionally, the anomaly assessment module 330 includes: an anomaly score determination unit, an abnormal object determination unit, and a list construction unit.

[0086] An abnormality score determination unit is used to input the evaluation association features corresponding to the resource supply objects to be evaluated into a pre-trained abnormality evaluation model for the plurality of resource supply objects to be evaluated, so as to obtain object abnormality scores corresponding to the resource supply objects to be evaluated;

[0087] an abnormal object determination unit, configured to sort the plurality of resource supply objects to be evaluated in descending order of the object abnormality scores, and select a preset number of resource supply objects to be evaluated that are ranked top as abnormal resource supply objects;

[0088] A list construction unit is used to construct an abnormal resource supply object list corresponding to the target object based on at least one of the abnormal resource supply objects.

[0089] Optionally, the feature extraction module 320 includes: a feature extraction method determination unit, a data feature determination unit and a feature construction unit.

[0090] A feature extraction method determining unit, used to determine a feature extraction method corresponding to the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated;

[0091] A data feature determination unit, configured to extract features of each item of the evaluation-related data based on the feature extraction method corresponding to the evaluation-related data, to obtain data features corresponding to the evaluation-related data;

[0092] The feature construction unit is used to construct an evaluation-related feature corresponding to the resource supply object to be evaluated based on data features corresponding to the plurality of evaluation-related data.

[0093] Optionally, the data type includes a sequence numerical type; the feature extraction method includes a similarity determination method; and the data feature determination unit includes: a historical data acquisition subunit and a data feature determination subunit.

[0094] A historical data acquisition subunit is used to acquire historical evaluation-related data corresponding to the evaluation-related data; wherein the historical evaluation-related data is the evaluation-related data corresponding to the resource supply object to be evaluated at any historical moment;

[0095] The data feature determination subunit is used to process the historical evaluation associated data and the evaluation associated data according to a preset similarity determination method to obtain data similarity corresponding to the evaluation associated data, and use the data similarity as the data feature corresponding to the evaluation associated data.

[0096] Optionally, the data type includes a text type; the feature extraction method includes a vectorized processing method; and the data feature determination unit includes: a text vector determination subunit.

[0097] The text vector determination subunit is used to process the evaluation-related data according to a preset vectorization processing method to obtain a text vector corresponding to the evaluation-related data, and use the text vector as a data feature corresponding to the evaluation-related data.

[0098] Optionally, the device also includes: a training sample acquisition module and a model training module.

[0099] A training sample acquisition module is used to acquire a plurality of training sample data; wherein the training sample data includes historical evaluation association features corresponding to the sample resource supply object at historical moments;

[0100] The model training module is used to train a pre-built machine learning model according to the plurality of training sample data to obtain an anomaly assessment model.

[0101] Optionally, the device also includes: a model parameter updating module.

[0102] A model parameter updating module is used to update the model parameters of the abnormal assessment model based on the assessment association features corresponding to the plurality of resource supply objects to be assessed.

[0103] Optionally, the device further includes: a list display module and an object association information display module.

[0104] A list display module, configured to display the abnormal resource supply object list after determining the abnormal resource supply object list corresponding to the target object;

[0105] The object association information display module is used to respond to the identifier selection operation for the object identifier, determine the abnormal resource supply object corresponding to the selected object identifier, obtain the object association information corresponding to the abnormal resource supply object, and display the object association information.

[0106] The abnormal object determination device provided in the embodiment of the present invention can execute the abnormal object determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0107] Embodiment 4

[0108] Figure 5 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0109] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0110] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0111] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as an abnormal object determination method.

[0112] In some embodiments, the abnormal object determination method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the abnormal object determination method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the abnormal object determination method in any other appropriate manner (e.g., by means of firmware).

[0113] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0114] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0115] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 foregoing.

[0116] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0117] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0118] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0119] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0120] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining an abnormal object, characterized in that: include: Acquire multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with the target object: wherein the resource supply objects to be evaluated provide relevant resources for the target object; and the evaluation-related data are data required for performing abnormal evaluation on the resource supply objects to be evaluated; For the plurality of resource supply objects to be evaluated, feature extraction is performed on the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain evaluation-related features corresponding to the resource supply object to be evaluated; Determine a list of abnormal resource supply objects corresponding to the target object according to a pre-trained abnormal assessment model and assessment association features corresponding to a plurality of resource supply objects to be assessed; Among them, the abnormal assessment model is obtained by training the machine learning model through the historical assessment association features corresponding to the sample resource supply object; the abnormal resource supply object list includes an object identifier corresponding to at least one abnormal resource supply object.

2. The abnormal object determination method according to claim 1, characterized in that: The determining, based on the abnormal evaluation model obtained by pre-training and the evaluation association features corresponding to the plurality of resource supply objects to be evaluated, a list of abnormal resource supply objects corresponding to the target object comprises: For the plurality of resource supply objects to be evaluated, the evaluation-related features corresponding to the resource supply objects to be evaluated are input into a pre-trained abnormal evaluation model to obtain object abnormality scores corresponding to the resource supply objects to be evaluated; Sorting the plurality of resource supply objects to be evaluated in descending order of the object abnormality scores, and taking a preset number of resource supply objects to be evaluated that are ranked top as abnormal resource supply objects; An abnormal resource provisioning object list corresponding to the target object is constructed based on at least one of the abnormal resource provisioning objects.

3. The abnormal object determination method according to claim 1, characterized in that: The extracting features of the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated to obtain the evaluation-related features corresponding to the resource supply object to be evaluated includes: Determine a feature extraction method corresponding to the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated; For each item of the evaluation-related data, extracting features of the evaluation-related data based on the feature extraction method corresponding to the evaluation-related data, to obtain data features corresponding to the evaluation-related data; An evaluation-related feature corresponding to the resource supply object to be evaluated is constructed based on data features corresponding to the plurality of evaluation-related data.

4. The abnormal object determination method according to claim 3, characterized in that: The data type includes a sequence numerical type; the feature extraction method includes a similarity determination method; the feature extraction of the evaluation-related data based on the feature extraction method corresponding to the evaluation-related data to obtain data features corresponding to the evaluation-related data includes: Acquire historical evaluation-related data corresponding to the evaluation-related data; wherein the historical evaluation-related data is the evaluation-related data corresponding to the resource supply object to be evaluated at any historical moment; The historical evaluation associated data and the evaluation associated data are processed according to a preset similarity determination method to obtain data similarity corresponding to the evaluation associated data, and the data similarity is used as a data feature corresponding to the evaluation associated data.

5. The abnormal object determination method according to claim 3, characterized in that: The data type includes a text type; the feature extraction method includes a vectorized processing method; the feature extraction of the evaluation-related data based on the feature extraction method corresponding to the evaluation-related data to obtain data features corresponding to the evaluation-related data includes: The evaluation-related data are processed according to a preset vectorization processing method to obtain a text vector corresponding to the evaluation-related data, and the text vector is used as a data feature corresponding to the evaluation-related data.

6. The abnormal object determination method according to claim 1, characterized in that: Also includes: Acquire a plurality of training sample data; wherein the training sample data includes historical evaluation association features corresponding to sample resource supply objects at historical moments; The pre-built machine learning model is trained according to the plurality of training sample data to obtain an anomaly assessment model.

7. The abnormal object determination method according to claim 1, characterized in that: Also includes: The model parameters of the abnormality assessment model are updated based on the assessment-related features corresponding to the plurality of resource supply objects to be assessed.

8. The abnormal object determination method according to claim 1, characterized in that: After determining the abnormal resource supply object list corresponding to the target object, the method further includes: Displaying the abnormal resource supply object list; In response to an identification selection operation for the object identification, the abnormal resource supply object corresponding to the selected object identification is determined, object association information corresponding to the abnormal resource supply object is acquired, and the object association information is displayed.

9. An abnormal object determination device, characterized in that: include: A data acquisition module, used to acquire multiple evaluation-related data corresponding to multiple resource supply objects to be evaluated that are associated with a target object: wherein the resource supply objects to be evaluated provide relevant resources for the target object; and the evaluation-related data are data required for performing abnormal evaluation on the resource supply objects to be evaluated; A feature extraction module, for extracting features of the corresponding evaluation-related data according to the data type of each evaluation-related data corresponding to the resource supply object to be evaluated, so as to obtain evaluation-related features corresponding to the resource supply object to be evaluated; An abnormal assessment module is used to determine a list of abnormal resource supply objects corresponding to the target object based on a pre-trained abnormal assessment model and assessment association features corresponding to multiple resource supply objects to be assessed; wherein the abnormal resource supply object list includes an object identifier corresponding to at least one abnormal resource supply object.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the abnormal object determination method according to any one of claims 1 to 8.

Images