User attribute data determination method, device, equipment, system, medium and product

Abstract

The invention discloses a user attribute data determination method, device, equipment, system, medium and product. The invention relates to the technical field of data processing. The method comprises the following steps: acquiring a project proof material of a declaration project uploaded by a target user; according to the project keyword corresponding to the declaration project and the to-be-verified item, performing text verification on the text certification material to obtain a text verification result; performing user portrait verification on each item activity picture by adopting the user portrait picture to obtain a portrait verification result; according to a project auditing model corresponding to the declaration project, performing activity scene verification on the project activity picture to obtain a scene verification result; determining a picture verification result according to the portrait verification result and the scene verification result; and if the text verification result and the picture verification result both pass verification, determining target credit attribute data of the target user according to the current credit attribute data and the declaration project credit attribute data of the target user.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of data processing technology, and in particular, to a method, device, equipment, system, medium and product for determining user attribute data. Background Art

[0002] At present, many local units usually adopt manual operations in the process of reviewing, calculating and managing user credit attribute data. That is, users declare activity projects offline, and relevant reviewers manually review the declared projects and manually calculate the user credit attribute data.

[0003] The review efficiency of existing user declaration activity projects is low, which leads to low efficiency in the process of determining user credit attribute data; and the accuracy and reliability of manually determined user credit attribute data are low. Summary of the invention

[0004] Embodiments of the present invention provide a method, device, equipment, system, medium and product for determining user attribute data, so as to solve the problems of low efficiency, accuracy and reliability in the process of determining user credit attribute data in the prior art.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining user attribute data, including:

[0006] Obtaining the project certification materials of the application project uploaded by the target user; the project certification materials include text certification materials and picture certification materials; the picture certification materials include a user portrait picture and at least one project activity picture;

[0007] According to the project keywords and items to be verified corresponding to the declared project, the textual certification materials are verified to obtain textual verification results;

[0008] Using the user portrait pictures to perform user portrait verification on each of the project activity pictures to obtain a portrait verification result; and,

[0009] According to the project review model corresponding to the declared project, the activity scene verification is performed on the activity pictures of each project to obtain the scene verification result;

[0010] Determine the picture verification result according to the portrait verification result and the scene verification result;

[0011] If both the text verification result and the image verification result are verified to be passed, obtaining the current credit attribute data of the target user;

[0012] Determine the declared project credit attribute data corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit attribute data of the target user and the declared project credit attribute data.

[0013] In a second aspect, an embodiment of the present invention further provides a device for determining user attribute data, the device comprising:

[0014] A certification material acquisition module is used to acquire the project certification materials of the application project uploaded by the target user; the project certification materials include text certification materials and picture certification materials; the picture certification materials include a user portrait picture and at least one project activity picture;

[0015] A text verification module, used to perform text verification on the textual certification materials according to the project keywords and items to be verified corresponding to the declared project, and obtain text verification results;

[0016] A portrait verification module, used to use the user portrait picture to perform user portrait verification on each of the project activity pictures to obtain a portrait verification result; and

[0017] A scene verification module is used to verify the activity scene of each project activity picture according to the project review model corresponding to the application project, and obtain the scene verification result;

[0018] A picture verification module, used to determine the picture verification result according to the portrait verification result and the scene verification result;

[0019] A current credit score determination module, configured to obtain the current credit attribute data of the target user if both the text verification result and the image verification result are verified to be passed;

[0020] The target credit data determination module is used to determine the declared project credit attribute data corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit score of the target user and the declared project credit attribute data.

[0021] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for determining user attribute data as described in any one of the embodiments of the present invention is implemented.

[0022] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for determining user attribute data as described in any one of the embodiments of the present invention.

[0023] In a fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements a method for determining user attribute data as described in any one of the embodiments of the present invention.

[0024] The embodiment of the present invention obtains the project certification materials of the declared project uploaded by the target user, and performs text verification on the text certification materials according to the project keywords corresponding to the declared project to obtain the text verification results, and performs image verification on the image certification materials according to the project review model and user portrait corresponding to the declared project to obtain the image verification results, thereby realizing the automated review of the project certification materials, reducing the occupation of human resources, and improving the efficiency, accuracy and reliability of the active project review. If both the text verification results and the image verification results are verified, the current credit attribute data of the target user is obtained, the declared project credit attribute data corresponding to the declared project is determined, and the target credit attribute data of the target user is determined according to the current credit attribute data of the target user and the declared project credit attribute data, thereby realizing the automated determination of the target credit attribute data of the target user, and improving the determination efficiency, accuracy and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 1 ;

[0027] Figure 2 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 2 ;

[0028] Figure 3 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 3 ;

[0029] Figure 4 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 4 ;

[0030] Figure 5 A schematic diagram of the structure of a device for determining user attribute data provided by an embodiment of the present invention;

[0031] Figure 6A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0032] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0033] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0034] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be noted that in the embodiments of the present invention, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solution of the present invention, but it does not mean that the inventor has or must use the relevant content of the solution. The acquisition, storage, use, processing, etc. of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0035] At present, resident users in many regional units, such as village-level units, can enrich their own credit attribute data by participating in various activity projects certified in their jurisdiction. For example, improve their own credit rating or credit score. In this process, it is usually necessary to review the activity projects reported by resident users, so as to further determine and calculate the user's credit attribute data. In the prior art, the review method for the activity projects reported by users is often manually reviewed by one or more project reviewers, and the user's credit attribute data is manually determined or calculated based on the review results. The existing review method consumes a lot of human resources, has low review efficiency, and because the review and calculation process is easily affected by subjective factors, the accuracy and reliability of the review and calculation are relatively low.

[0036] The present invention provides a method, device, equipment, system, medium and product for determining user attribute data. By obtaining the project certification materials of the declared project uploaded by the target user, and performing text verification on the text certification materials according to the project keywords corresponding to the declared project, and performing image verification on the image certification materials according to the project review model corresponding to the declared project, the image verification result is obtained, thereby realizing the automated review of the user's declared project, and in the review process, different review means or review methods are used based on different types of certification materials, thereby improving the accuracy and reliability of the review results. If the review is passed, the current credit attribute data of the target user is obtained, the declared project credit attribute data corresponding to the declared project is determined, and the target credit attribute data of the target user is determined according to the current credit attribute data of the target user and the declared project credit attribute data, thereby realizing the automated determination of the user's target credit attribute data, while improving the efficiency of determining the attribute data, and improving the accuracy and reliability of determining the attribute data.

[0037] Figure 1 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 1 This embodiment can be applied to the case of automatically reviewing project materials for activity projects declared by users and automatically determining the credit attribute data of users. This method can be executed by a user attribute data determination device, which can be implemented in the form of hardware and / or software. Figure 1 As shown, the method comprises the following steps:

[0038] S110. Obtain project certification materials uploaded by the target user for the declared project; the project certification materials include text certification materials and picture certification materials; the picture certification materials include a user portrait picture and at least one project activity picture;.

[0039] Among them, the target users can be resident users who have the need to apply for active projects. The target users can apply through the project application system and upload project certification materials on the corresponding operation interface.

[0040] Among them, the declared project is the activity project declared by the target user; the activity project is the activity behavior recognized by the relevant departments or relevant regions. For example, the activity project may include voluntary blood donation activities, disaster relief donation activities organized by the village collective, village voluntary activities, village organization meetings, activities to help the elderly, activities to help children, and auxiliary activities in cooperation with relevant law enforcement departments.

[0041] Among them, the project certification materials can be the certification documents provided by the target users to prove the authenticity and validity of their participation in the activity project, which can specifically include text certification materials and picture certification materials.

[0042] The textual proof materials may include the target user's identity proof information to prove the legality, validity and authenticity of his / her identity, and may also include project description information of the target user describing the activity project in which he / she participates to prove the authenticity of his / her participation in the activity.

[0043] Picture proof materials may include user portrait pictures of the target user to prove the legitimacy, validity and authenticity of his or her identity, as well as project activity pictures obtained by shooting or collecting images of the activity site in which the target user participates to prove the authenticity and validity of his or her participation in the activity.

[0044] Among them, the user portrait picture is a picture that contains a user portrait image. Before performing user portrait verification on each project activity picture, the user portrait picture can be verified first. Specifically, it can be verified whether the user portrait picture contains a portrait image. And whether the user portrait picture meets the preset portrait picture requirements. The preset portrait picture requirements are pre-set by relevant technical personnel based on actual needs. For example, the preset portrait picture requirements can be a white or blue background area, a size of 300×500 pixels, and a resolution not exceeding 300dpi (dots per inch).

[0045] Exemplarily, when a user's application request for an active project is detected in the project application system, the project certification materials of the application project uploaded by the target user performing the operation are obtained. In the operation interface for application of active projects, a selection list of active projects that can be applied for selection by users can be provided. Users can select one or more application projects from the selection list of active projects according to actual application needs, and upload the project certification materials corresponding to the selected application projects.

[0046] S120. According to the project keywords and items to be verified corresponding to the declared project, text verification is performed on the textual certification materials to obtain text verification results.

[0047] It should be noted that, due to the different activity scenarios of different activity projects, different activity projects correspond to their own activity keywords. For example, the activity keywords for the activity project of helping the elderly may include: "elderly", "help", "community activities", "health assessment", "life skills", "psychological support", "communication and interaction" and "convenient services", etc. For another example, the activity keywords for donation activities may include: "charity", "public welfare", "donor", "rescuer", "quality of life", "transparency", "gratitude" and "beneficiary", etc. The project keywords are the activity keywords corresponding to the activity projects declared by the user.

[0048] It should be noted that different activity projects require different text proof materials to be uploaded, including different items of identity proof information to be uploaded. For example, for the activity project of helping the elderly, the identity proof information that needs to be uploaded includes: the user's valid identity proof information item, the valid identity proof information item of the assisted elderly, and the valid identity proof information item of the organizer of the activity. For the voluntary activities organized by the village, the identity proof information that needs to be uploaded includes: the user's valid identity proof information item and the village organization's valid identity proof information item.

[0049] The items to be verified may be the uploaded items required for the activity projects declared by the target user. In the process of verifying the identity information in the text proof materials, it is necessary to review, check or verify from the two dimensions of completeness and consistency.

[0050] Exemplarily, based on the project keywords and the items to be verified, the project description information in the textual supporting materials can be subjected to text verification. Specifically, the similarity between the textual vocabulary corresponding to the project keywords and the project description information corresponding to the items to be verified can be determined, thereby verifying whether the project description information is the description information corresponding to the declared project, and verifying whether the project description information is standardized and reasonable.

[0051] Optionally, it is also possible to verify whether the text information of the text evidence material complies with the format specifications of the evidence material, for example, whether the text font, paragraph spacing, number of words and text type of the project evidence material meet the preset standard format. Specifically, it is possible to determine the text format of the text evidence material uploaded by the user, and compare the text format with the standard format. If the comparison is consistent, the text evidence material is verified to be passed; if the comparison is inconsistent, the text evidence material is not verified to be passed.

[0052] S130, using the user portrait picture to perform user portrait verification on each project activity picture to obtain a portrait verification result.

[0053] When the user portrait image meets the preset portrait image requirements, the user portrait image is used to verify the user portrait of each project activity image to obtain a portrait verification result. Specifically, the user portrait image can be used to verify whether the user portrait exists in each project activity image, thereby verifying whether the target user participates in each project activity image uploaded by the target user.

[0054] S140. According to the project review model corresponding to the application project, the activity scene of each project activity picture is verified to obtain the scene verification result.

[0055] Among them, the project review model is used to verify whether the project activity pictures in the picture proof materials corresponding to the declared project are accurate and valid. Due to the different scenarios of different activity projects, different activity projects correspond to different activity review models. The project review model is the activity review model corresponding to the activity project declared by the target user.

[0056] The activity scene verification of the project activity picture can specifically be to determine whether the project activity picture meets the scene requirements. Different activity projects correspond to different activity scenes. Therefore, different activity projects correspond to different activity review models. The project review model is the activity review model corresponding to the activity project reported by the target user.

[0057] The activity audit model can be obtained by pre-training a pre-built network model based on historical picture evidence materials of the corresponding activity project in the historical period by relevant technical personnel. The network model can be an existing model with a known structure or a model built by relevant technical personnel themselves.

[0058] For example, each project activity picture can be input into the project review model, and the project review model performs scenario verification and outputs scenario verification results corresponding to each project activity picture. The project review model can be obtained by relevant technical personnel through pre-training of a pre-built network model based on project activity pictures in historical periods and combined with activity scenarios.

[0059] S150: Determine the image verification result according to the portrait verification result and the scene verification result.

[0060] If both the portrait verification result and the scene verification result are verified, it can be determined that the picture verification of the picture proof material has passed; if any of the portrait verification result and the scene verification result fails, it can be determined that the picture verification of the picture proof material has failed.

[0061] It should be noted that in this embodiment, the execution order between S130-S150 and S120 can be S130-S150 and S120 executed simultaneously; or S130-S150 can be executed first and then S120, or S120 can be executed first and then S130-S150; or S120 can be executed simultaneously during the execution of S130-S150, and this embodiment does not limit this.

[0062] S160: If both the text verification result and the image verification result are verified to be passed, the current credit attribute data of the target user is obtained.

[0063] The credit attribute data may be a user's credit score. A higher credit score indicates a higher user's credit rating; a lower credit score indicates a lower user's credit rating. The credit attribute data may be updated based on the user's activity project declaration. The current credit attribute data may be the target user's credit attribute data in the current time period.

[0064] Exemplarily, if both the text verification result and the image verification result are verified to be passed, the current credit attribute data of the target user is obtained; if any of the text verification result and the image verification result fails, it means that the review of the target user's project certification materials has failed, and the target credit attribute data of the target user remains the current credit attribute data.

[0065] S170: Determine the credit attribute data of the declared project corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit attribute data of the target user and the credit attribute data of the declared project.

[0066] Among them, due to the different participation, difficulty and social impact of different activity projects, different activity projects correspond to different activity project credit attribute data, that is, different activity projects correspond to different activity project credit scores. The credit attribute data of the declared project is the credit attribute data of the activity project corresponding to the activity project declared by the target user.

[0067] For example, the target credit score of the target user can be determined based on the current credit score of the target user and the credit score of the active project of the declared project. Specifically, the current credit score and the credit score of the active project can be summed to obtain the target credit score of the target user, that is, the target credit attribute data. For example, if the current credit score of the target user is 80 points and the credit score of the active project of the declared project is 3 points, the target credit score of the target user is 83 points.

[0068] The embodiment of the present invention obtains the project certification materials of the declared project uploaded by the target user, and performs text verification on the text certification materials according to the project keywords and the items to be verified corresponding to the declared project to obtain the text verification results, and performs image verification on the image certification materials according to the project review model corresponding to the declared project to obtain the image verification results, thereby realizing the automated review of the project certification materials, reducing the occupation of human resources, and improving the efficiency, accuracy and reliability of the review of the active project. If both the text verification result and the image verification result are verified, the current credit attribute data of the target user is obtained, the credit attribute data of the declared project corresponding to the declared project is determined, and the target credit attribute data of the target user is determined according to the current credit attribute data of the target user and the credit attribute data of the declared project, thereby realizing the automated determination of the target credit attribute data of the target user, and improving the determination efficiency, accuracy and reliability. According to the portrait verification result and the scene verification result, the picture verification result is determined, thereby realizing the automated verification of the picture certification materials, and improving the verification efficiency of the picture certification materials; the picture material verification is performed from the two verification dimensions of user portrait and activity scene, thereby improving the verification accuracy of the picture certification materials, thereby improving the verification accuracy of the project certification materials of the declared project.

[0069] Figure 2 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 2 The embodiment of the present invention optimizes and improves the review method of textual evidence materials based on the above embodiment, such as Figure 2 As shown, the method for determining user attribute data includes the following steps:

[0070] S210, obtaining the project certification materials of the application project uploaded by the target user; the project certification materials include text certification materials and picture certification materials; the picture certification materials include user portrait pictures and at least one project activity picture;

[0071] S220. According to the project keywords corresponding to the applied project, the project description information in the textual supporting materials is verified to obtain the description information verification result.

[0072] It should be noted that the project description information in the textual proof materials submitted by the target user usually includes multiple activity keywords in the scenario of the activity project in which the target user participates. Therefore, based on the activity scenarios of the corresponding activity projects and the historical project description information submitted by users in the historical period, activity keywords corresponding to different activity projects can be generated. Among them, the project keywords are the activity keywords of the activity projects declared by the target user. Therefore, the project description information can be verified by extracting keywords and comparing keywords in the project proof materials of the target user.

[0073] In an optional embodiment, the project description information in the textual evidence is verified according to the project keywords corresponding to the declared project, and the description information verification result is obtained, including:

[0074] Step a1: extract keywords from the project description information in the textual evidence material to obtain at least one description keyword.

[0075] Among them, the method of extracting keywords from the project description information can be a frequency-based method, a statistics or machine learning-based method, a deep learning-based method, or a set rule-based method.

[0076] The frequency-based keyword extraction method can be a method of extracting keywords by determining the TF-IDF (Term Frequency-Inverse Document Frequency) value, or a term frequency (TermFrequency) statistical method. The statistical or machine learning-based method can be, for example, a graph-based ranking algorithm (TextRank), which can specifically calculate the importance of words through the co-occurrence relationship between words, and extract keywords based on the importance of words. Alternatively, it can also be LSA (Latent Semantic Analysis), which can specifically convert the text into a semantic space through singular value decomposition, and extract the words with the largest variance in the semantic space as keywords.

[0077] The deep learning-based approach can be, for example, the use of pre-trained word embedding models, such as the GloVe (global log bilinear regression model) model or the Word2Vec (word embeddings) model, etc., to capture the semantic information between words and identify keywords; or, learn the contextual representation of the text through a pre-trained Transformer model, and further extract keywords. The rule-based approach can be part-of-speech filtering or phrase extraction, which can specifically filter out words with specific parts of speech such as nouns and verbs as keywords based on the part-of-speech tagging results; or, extract phrases with specific grammatical structures or frequent co-occurrences as key phrases through grammatical rules or statistical rules. Grammatical rules and statistical rules can be pre-set by relevant technical personnel according to actual needs.

[0078] Exemplarily, based on any one of the above keyword extraction methods, keyword extraction can be performed on the project description information in the textual proof material to obtain at least one description keyword. Alternatively, based on the above multiple keyword extraction methods, keyword extraction can be performed on the project description information in the textual proof material to obtain at least one description keyword corresponding to different keyword extraction methods. Keywords are merged for each description keyword corresponding to each keyword extraction method to obtain at least one identical description keyword.

[0079] Specifically, if the keywords extracted by keyword extraction method A include: keyword a, keyword b, keyword c, keyword d and keyword e; the keywords extracted by keyword extraction method B include: keyword a, keyword b, keyword c, keyword d and keyword f; the keywords extracted by keyword extraction method C include: keyword a, keyword b, keyword c, keyword d and keyword g, and it can be determined that the same keywords extracted by the three keyword extraction methods are: keyword a, keyword b, keyword c, keyword d, then the same keywords: keyword a, keyword b, keyword c, keyword d are used as the final description keywords.

[0080] Step a2: perform consistency comparison between each description keyword and the project keyword to determine the first target keyword that meets the consistency comparison condition.

[0081] The consistency comparison condition may be preset by relevant technical personnel. For example, the consistency comparison condition may be that the description keyword to be compared is the same as the project keyword, that is, the same keyword.

[0082] Exemplarily, each description keyword is compared with the project keyword for consistency. For example, if the description keywords are keyword a, keyword b and keyword c, and the project keywords are keyword m, keyword n ​​and keyword z, keyword a is compared with keyword m, keyword n ​​and keyword z for consistency, and it is determined whether keyword a meets the consistency comparison conditions with keyword m, keyword n ​​and keyword z. If keyword a meets the consistency comparison conditions with any of the keywords m, keyword n ​​and keyword z, keyword a is determined as the first target keyword. The determination method of keyword b and keyword c is the same, and specifically, they are compared with keyword m, keyword n ​​and keyword z for consistency, and it is determined whether the consistency comparison conditions are met. This embodiment will not be repeated. Therefore, the first target keyword is the keyword that is exactly the same as the project keyword.

[0083] Step a3: Determine the proportion of the first keyword according to the number of first keywords of the first target keyword and the number of description keywords of the description keyword.

[0084] The first keyword ratio may be the ratio between the number of first keywords and the number of description keywords, which is used to evaluate the consistency between description keywords and project keywords. If the first keyword ratio is larger, the consistency between description keywords and project keywords is greater; if the first keyword ratio is smaller, the consistency between description keywords and project keywords is smaller.

[0085] Exemplarily, if the number of first keywords of the first target keyword is 10 and the number of description keywords of the description keyword is 20, the proportion of the first keyword is 50%.

[0086] Step a4: If the proportion of the first keyword does not meet the preset proportion judgment condition, then compare the similarity of other keywords with the project keyword to determine the second target keyword that meets the similarity comparison condition; the other keywords are the description keywords in each description keyword except the first target keyword.

[0087] Among them, the proportion judgment condition can be pre-set by relevant technical personnel based on actual experience values ​​or experimental values. For example, the proportion judgment condition can be that the proportion of the first keyword is greater than a preset proportion threshold. The proportion threshold can be set manually according to actual conditions, such as the proportion threshold can be set to 90%.

[0088] Exemplarily, if the proportion of the first keyword meets a preset proportion judgment condition, it may indicate that the text verification of the project description information of the text proof material has passed.

[0089] If the proportion of the first keyword meets the preset proportion judgment condition, it means that the number of description keywords that completely match the project keyword in the project description information is insufficient. It is understandable that there are usually two keywords in the comparison process. Although the two keywords are not exactly the same or consistent, the meanings represented by the two keywords are the same. For example, the meanings expressed by the keyword "help" and the keyword "assistance" are close or similar. Therefore, the project description information can be further verified by further determining similar keywords.

[0090] Specifically, a similarity comparison is performed between other keywords and the project keywords, wherein the other keywords are description keywords other than the first target keyword in each description keyword. The similarity comparison method between keywords can be a word vector model comparison method, a similarity calculation method based on vocabulary and grammar, or a similarity calculation method based on semantics or context.

[0091] Among them, the word vector model comparison method can be, for example, a cosine similarity comparison method; specifically, the cosine angle between two word vectors can be calculated to measure their similarity. Alternatively, the word vector model comparison method can also be a Euclidean distance comparison method; specifically, in the word vector space, the smaller the Euclidean distance between two word vectors, the more similar they are. Alternatively, the vector model comparison method can also be a Manhattan distance comparison method; specifically, the sum of the absolute values ​​of the coordinates of two word vectors can be calculated, which can also be used to measure the similarity between keywords.

[0092] The similarity calculation method based on vocabulary and grammar may be the edit distance, which may specifically be the minimum number of operations required to convert one keyword into another keyword, such as inserting, deleting, and replacing characters, etc. Alternatively, the similarity calculation method based on vocabulary and grammar may also be the Jaccard Similarity method, etc.

[0093] Among them, the similarity calculation method based on semantics or context can be a semantic similarity assessment (SemanticSimilarity Measures), which can specifically be to use natural language processing technology and semantic models, such as the WordNet model or the BERT (Bidirectional Encoder Representations from Transformers) model, to analyze the semantic meaning of keywords, thereby determining the similarity between keywords.

[0094] Exemplarily, the above-mentioned keyword similarity comparison method is used to compare the similarity of other keywords with the project keyword to determine the second target keyword that meets the similarity comparison condition. The similarity comparison condition can be preset by relevant technical personnel based on actual experience values ​​or experimental values. For example, the similarity comparison condition can be that the similarity between the other keywords and the project keyword is greater than a preset similarity threshold. The similarity threshold can be preset according to actual scenario requirements; for example, the similarity threshold can be set to 50%.

[0095] It should be noted that the number of other keywords and project keywords may be multiple. In the case of multiple keywords, the multiple other keywords are compared with the multiple project keywords for similarity, which can be the same as the consistency comparison between the description keyword and the project keyword. Specifically, if there are other keywords, namely keyword a, keyword b and keyword c, and the project keywords are keyword m, keyword n ​​and keyword z, then keyword a is compared with keyword m, keyword n ​​and keyword z for similarity, and it is determined whether keyword a meets the similarity comparison condition with keyword m, keyword n ​​and keyword z. If keyword a meets the similarity comparison condition with any one of keyword m, keyword n ​​and keyword z, keyword a is determined as the second target keyword. The determination method of keyword b and keyword c is the same, and specifically, the similarity is compared with keyword m, keyword n ​​and keyword z, and it is determined whether the similarity comparison condition is met. This embodiment will not be repeated.

[0096] Step a5: Determine the proportion of the second keyword according to the number of second keywords of the second target keyword and the number of description keywords of the description keyword.

[0097] The second keyword proportion can be the ratio between the number of second keywords and the number of description keywords, which is used to evaluate the similarity between other keywords and project keywords. The larger the second keyword proportion, the higher the similarity between other keywords and project keywords; the smaller the second keyword proportion, the lower the similarity between other keywords and project keywords.

[0098] Exemplarily, if the number of second keywords of the second target keyword is 5 and the number of description keywords of the description keyword is 20, the proportion of the second keyword is 25%.

[0099] Step a6: Based on the proportion of the second keyword, text verification is performed on the project description information in the textual proof material to obtain a description information verification result.

[0100] Exemplarily, if the proportion of the second keyword is greater than the preset proportion threshold, it may indicate that the text verification of the project description information in the text proof material has passed; if the proportion of the second keyword is not greater than the preset proportion threshold, it may indicate that the text verification of the project description information in the text proof material has failed.

[0101] The above technical solution compares the consistency of each description keyword obtained by keyword extraction of the project description information with the project keyword, determines the first target keyword that meets the consistency comparison condition, and determines the first keyword ratio between the first target keyword and the description keyword. If the first keyword ratio does not meet the preset ratio judgment condition, then determine the second target keyword that meets the similarity comparison condition, and determine the second keyword ratio between the second target keyword and the description keyword. According to the second keyword ratio, the project description information is verified textually. In the text verification process, the project keywords in the application scenarios of different activity projects are combined, and verification is performed from the two verification dimensions of keyword consistency and similarity, thereby improving the verification accuracy of the project description information. In addition, the two judgment processes of keyword consistency and similarity are performed in a progressive manner, so that the verification results can be quickly obtained in the scenario of highly consistent keyword matching, thereby improving the verification efficiency.

[0102] S230. According to the items to be verified corresponding to the declared project, the identity information in the textual certification materials is verified to obtain the identity information verification result.

[0103] In an optional embodiment, according to the to-be-verified items corresponding to the declared project, the identity information in the textual certification materials is verified to obtain the identity information verification result, including:

[0104] Step b1: perform integrity verification on the identity verification items in the identity verification information according to the items to be verified corresponding to the declared items, and obtain a text integrity verification result.

[0105] Among them, the identity proof item can be at least one proof text uploaded by the target user for the declared project to verify his or her own identity information; the integrity verification can specifically verify whether the number of identity proof items is the same as the number of items to be verified.

[0106] For example, if the items to be verified corresponding to the declared project include item A, item B and item C, the identity proof items in the identity proof information may include identity proof item A and identity proof item B. In the process of verifying the integrity of the identity proof information, since the items to be verified are not the same in number as the identity proof items, after comparison, the identity proof item C corresponding to the item C to be verified is missing, and therefore, the integrity verification of the identity proof items fails.

[0107] Step b2: According to the data type corresponding to the item to be verified, a consistency check is performed on the proof text corresponding to the identity proof item to obtain a text consistency verification result.

[0108] It should be noted that different items to be verified correspond to different data types of the data to be verified. For example, for item A to be verified, the corresponding data type is a string type; for item B to be verified, the corresponding data type is a text type.

[0109] Exemplarily, determine the first data type of the item to be verified corresponding to the identity proof item, and determine the second data type of the proof text corresponding to the identity proof item, and determine whether the first data type and the second data type are the same type. If so, the text consistency verification passes; if not, the text consistency verification fails.

[0110] Step b3: Determine the identity information verification result based on the text integrity verification result and the text consistency verification result.

[0111] If both the text integrity verification result and the text consistency verification result are passed, it can be determined that the text verification of the identity information of the text proof material is passed; if any of the text integrity verification result and the text consistency verification result is failed, it can be determined that the text verification of the identity information of the text proof material is failed.

[0112] The above technical solution verifies the project description information in the textual proof materials according to the project keywords corresponding to the declared project, and verifies the identity information in the textual proof materials according to the items to be verified corresponding to the declared project, obtains the identity information verification result, and determines the text verification result according to the description information verification result and the identity information verification result, thereby realizing the automatic verification of the identity information. Based on different verification methods, the verification is performed separately from the two dimensions of data type consistency and verification item integrity, which improves the verification accuracy of the identity information, thereby improving the verification accuracy of the textual proof materials, and further improving the verification accuracy of the project proof materials.

[0113] S240: Determine a text verification result according to the description information verification result and the identity information verification result.

[0114] If both the description information verification result and the identity information verification result are verified, it can be determined that the text verification of the text proof material has passed; if any of the description information verification result and the identity information verification result fails, it can be determined that the text verification of the text proof material has failed.

[0115] S250: Use the user portrait pictures to perform user portrait verification on the activity pictures of each project to obtain portrait verification results.

[0116] S260. According to the project review model corresponding to the application project, the activity scene of each project activity picture is verified to obtain the scene verification result.

[0117] S270: Determine the image verification result according to the portrait verification result and the scene verification result.

[0118] S280: If both the text verification result and the image verification result are verified to be passed, the current credit attribute data of the target user is obtained.

[0119] S290: Determine the credit attribute data of the declared project corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit attribute data of the target user and the credit attribute data of the declared project.

[0120] It should be noted that in this embodiment, the execution order between S220-S240 and S250-S270 can be S220-S240 and S250-S270 are executed simultaneously; it can also be S220-S240 is executed first, and then S250-S270 is executed, or S250-S270 is executed first, and then S220-S240 is executed; it can also be that S250-S270 is executed simultaneously during the execution of S220-S240, and this embodiment does not limit this.

[0121] The technical solution of this embodiment realizes automatic verification of textual proof materials and improves the verification efficiency of textual proof materials by performing textual verification on the project description information in the textual proof materials according to the project keywords corresponding to the declared projects, and performing textual verification on the identity proof information in the textual proof materials according to the items to be verified corresponding to the declared projects, and determining the text verification result according to the description information verification result and the identity information verification result; verifies the project description information and the identity proof information respectively based on different verification methods, fully considers the specific verification methods under different verification dimensions, improves the verification accuracy of textual proof materials, and thus improves the verification accuracy of project proof materials of declared projects.

[0122] Figure 3 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 3 The embodiment of the present invention optimizes and improves the review method of the picture evidence material based on the above embodiment, such as Figure 3 As shown, the method for determining user attribute data includes the following steps:

[0123] S310. Obtain the project certification materials of the application project uploaded by the target user; the project certification materials include text certification materials and picture certification materials; the picture certification materials include a user portrait picture and at least one project activity picture.

[0124] S320. According to the project keywords and items to be verified corresponding to the declared project, text verification is performed on the textual proof materials to obtain text verification results.

[0125] S330, obtaining the number of pictures required for the project corresponding to the application project.

[0126] It should be noted that the number of pictures required for different activity items to be uploaded to prove the authenticity of the user's participation in the activity can be the same or different, and can be pre-set by relevant technical personnel according to actual needs, and the mapping relationship between the activity items and the number of pictures required for the activity can be pre-stored. For example, a mapping relationship table between the activity items and the number of pictures required for the activity can be generated and stored. The mapping relationship table stores at least one activity item and its corresponding number of pictures required for the activity.

[0127] The number of images required for the project may be the number of images required for the activity corresponding to the activity project declared by the target user.

[0128] Exemplarily, a mapping relationship table between activity projects and the number of pictures required for the activities is obtained, and based on the mapping relationship table, the number of pictures required for the activities corresponding to the declared projects is searched, and the number of pictures required for the activities obtained by the search is determined as the number of pictures required for the project.

[0129] S340. Verify the number of pictures of the project activities according to the number of pictures required for the project, and obtain a verification result of the number of pictures.

[0130] Determine the number of uploaded images of at least one project activity image uploaded by the target user, and judge whether the number of uploaded images is not less than the number of images required for the project. If so, the image quantity verification of the project activity images passes; if not, the image quantity verification of the project activity images fails.

[0131] Optionally, if the verification of the number of pictures of the project activity pictures fails, that is, the number of pictures of the project activity pictures uploaded by the target user is insufficient, a picture supplement prompt message may be sent to the terminal to which the target user belongs, so that the target user can supplement the project activity pictures in time when receiving the picture supplement prompt message. If the project activity pictures supplemented by the user are received within the preset picture supplement time period, the number of pictures of the project activity pictures supplemented by the user may be verified again. If the second verification still fails, it means that the verification of the picture certification materials of the declared project has failed.

[0132] S350: If the result of the picture quantity verification is passed, user portrait verification is performed on each project activity picture using the user portrait picture to obtain a portrait verification result.

[0133] In an optional embodiment, user portrait pictures are used to perform user portrait verification on each project activity picture to obtain a portrait verification result, including:

[0134] Step c1: performing portrait recognition on the project activity picture to obtain at least one portrait area image corresponding to the project activity picture.

[0135] For any project activity picture, a pre-selected or pre-trained portrait recognition model for portrait recognition is used to perform portrait recognition on the project activity picture to obtain at least one portrait area image in the project activity picture. The portrait area image may be an area image corresponding to the area where the portrait is located in the project activity picture.

[0136] Among them, the portrait recognition model can be a VGG (Visual Geometry Group) model, a ResNet (Residual Network) model or a FaceNet (Face Network, a network for facial feature recognition) model, etc.

[0137] Optionally, in addition to the above-mentioned method of using a portrait recognition model for recognition, the portrait recognition method may also use methods such as image segmentation or facial key point recognition algorithms. Specifically, it may be a combination of local features and global features, using local features to describe facial key points and global features to describe the entire face. Alternatively, the image may be segmented into different areas or objects to distinguish between the portrait and the background, thereby more accurately identifying the position and outline of the portrait.

[0138] Step c2: perform image similarity comparison between each portrait area image and the user portrait image in the user portrait picture, and obtain similarity comparison results corresponding to each portrait area image in the project activity picture.

[0139] It should be noted that in an activity project, in addition to the target user, there may be other people related to the activity project participating in the activity project. Therefore, in the obtained project activity picture, in addition to the user portrait image of the target user, there may also be multiple participant portrait images of other activity participants. Therefore, the number of portrait area images obtained by performing portrait recognition on the project activity picture may be one or more. In the process of using user portrait images for similarity comparison, it is necessary to perform similarity comparison with the recognized portrait area portraits one by one.

[0140] Exemplarily, the user portrait image is compared with each portrait region image for image similarity, and similarity comparison results between each portrait region image and the user portrait image are obtained. The image similarity comparison method may be a similarity comparison method based on image content, a similarity comparison method based on image features, and a similarity comparison method based on image structure.

[0141] Among them, the similarity comparison method based on image content can be a structural similarity (SSIM) comparison method, which can comprehensively consider image brightness, contrast and structural information during the similarity comparison of two images. Alternatively, it can also be a Peak Signal-to-Noise Ratio (PSNR) comparison method, which quantifies the similarity between two images by calculating the mean square error between them. Alternatively, it can also be a difference method, which calculates the pixel difference or histogram difference between two images.

[0142] Among them, the similarity comparison method based on image features can be a histogram comparison method, which can specifically compare the color histograms of two images to determine the similarity in color distribution. Alternatively, it can also be a local feature descriptor comparison method, such as SIFT (Scale-Invariant Feature Transform) and SURF (Speeded Up Robust Features), etc., which can specifically compare the key points and descriptors of the two images. Alternatively, it can also be a deep learning feature comparison method, such as a VGG model or a GAN (Generative Adversarial Network) model to extract image feature representations and calculate the similarity between images.

[0143] Among them, the similarity comparison method based on image structure can be shape matching, which can be specifically shape matching through contour detection or edge detection; or it can also be template matching, matching the template of one image with different positions of another image to determine the similarity between the two images.

[0144] Exemplarily, any one of the above-mentioned image similarity comparison methods can be used to determine the similarity comparison results between each portrait area image and the user portrait image. Alternatively, the above-mentioned multiple image similarity comparison methods can be used to determine the similarity comparison results between each portrait area image and the user portrait image under various image similarity comparison methods, and the final similarity comparison result is determined based on the preset similarity threshold according to the similarity comparison results corresponding to the various image similarity comparison methods. Among them, the number of selected image similarity comparison methods is preferably an odd number.

[0145] Specifically, if the selected image similarity comparison methods are method A, method B, method C, method D and method E. The result determined by method A is that the similarity between image m and image n is 90%; the result determined by method B is that the similarity between image m and image n is 85%; the result determined by method C is that the similarity between image m and image n is 40%; the result determined by method D is that the similarity between image m and image n is 55%; the result determined by method E is that the similarity between image m and image n is 85%. If the preset similarity threshold is 80%, two images that are not lower than the preset similarity threshold are considered similar, and two images that are lower than the preset similarity threshold are considered dissimilar. The above methods that meet the preset similarity threshold judgment condition are method A, method B and method E, and the methods that do not meet the preset similarity threshold judgment condition are method C and method D. The number of methods that meet the conditions is greater than the number of methods that do not meet the conditions. Therefore, it can be determined that image m and image n are similar, and the similarity is 86.67%.

[0146] Step c3: Based on the similarity comparison results corresponding to the portrait area images in the project activity pictures, user portrait verification is performed on the project activity pictures to obtain portrait verification results.

[0147] If any of the similarity comparison results corresponding to the portrait area images is greater than the preset similarity threshold, the user portrait verification of the project activity picture is passed; if no similarity comparison result is greater than the preset similarity threshold among the similarity comparison results corresponding to the portrait area images, the user portrait verification of the project activity picture is not passed. The similarity threshold can be preset by relevant technical personnel according to actual needs, for example, it can be set to 90%.

[0148] The above technical scheme obtains the similarity comparison results corresponding to each portrait area image in the project activity picture by comparing the image similarity between each portrait area image and the user portrait image in the user portrait picture, and verifies the user portrait of the project activity picture based on the similarity comparison results corresponding to each portrait area image in the project activity picture, thereby realizing automated verification of the user portrait of the project activity picture. In addition, during the verification process, the user portrait image is compared with the similarity of each portrait area portrait in the activity picture, thereby improving the accuracy of user portrait verification for the project activity picture.

[0149] S360. Use the project review model corresponding to the application project to verify the activity scene of each project activity picture and obtain the scene verification result.

[0150] In an optional embodiment, the model training method of the project review model corresponding to the declared project is as follows:

[0151] Step d1: obtain historical project activity pictures of the declared project in different activity scenes and shooting angles during the historical time period, and use the historical project activity pictures as positive sample data.

[0152] Among them, different activity scenes may refer to the scenes in which the declared project is located. For example, for the activity project of helping the elderly, different activity scenes may be in a home scene and in an outdoor scene, etc. Among them, the project review model is the activity review model corresponding to the activity project declared by the target user.

[0153] Step d2: Generate negative sample data corresponding to the positive sample data.

[0154] The negative sample data may have the same image size, approximately the same sample number, and the same preprocessing process as the positive sample data. For example, the negative sample data and the positive sample data may be processed using the same preprocessing method, such as image resizing, color space conversion, and standardization.

[0155] The label of the positive sample data indicates that the image contains a specific target or feature, and the label of the negative sample data indicates that the image does not contain the target or feature. Negative sample data can be randomly generated by relevant technical personnel, or historical project activity pictures corresponding to other activity projects can be used as negative sample data.

[0156] Step d3: Use positive sample data and negative sample data to train the pre-built network model until the preset model training end conditions are met, and obtain the model of the project review model corresponding to the application project.

[0157] The network model may be pre-built by relevant technical personnel, or may be a network model of an existing structure, such as CNN (Convolutional Neural Network).

[0158] Exemplarily, the positive sample data and the negative sample data are input into the pre-built network model, and for each preset training batch, the model performs forward propagation to calculate the prediction results based on the input image. The loss function is calculated, and the parameters of the model are updated according to the gradient calculated by the loss function for back propagation until the entire training data set is traversed or the preset model training end condition is reached, for example, the maximum number of iterations is reached or the loss converges, and a trained activity audit model is obtained.

[0159] The above-mentioned technical scheme obtains historical project activity pictures of the declared project in different activity scenes and different shooting angles within the historical time period, and uses the historical project activity pictures as positive sample data, and uses the positive sample data and negative sample data to perform model training on the pre-constructed network model until the preset model training end conditions are met, thereby obtaining the model of the project review model corresponding to the declared project, and realizing the accurate training of the activity review models corresponding to different activity projects. The model training is combined with different scenes and different angles, which further improves the accuracy of model training, thereby improving the verification accuracy of activity scene verification.

[0160] It can be understood that the execution order of user portrait verification and activity scene verification for each project activity picture can be: user portrait verification first, then activity scene verification; it can also be: activity scene verification first, then user portrait verification; or user portrait verification and activity scene verification can be performed at the same time. This embodiment does not impose any limitation on the execution order of the two verification processes.

[0161] S370: Determine the image verification result according to the portrait verification result and the scene verification result.

[0162] S380: If both the text verification result and the image verification result are verified to be passed, the current credit attribute data of the target user is obtained.

[0163] S390: Determine the credit attribute data of the declared project corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit attribute data of the target user and the credit attribute data of the declared project.

[0164] The technical solution of this embodiment verifies the number of pictures of project activity pictures according to the number of pictures required for the project to obtain the picture number verification result, uses user portrait pictures to verify the user portrait of each project activity picture to obtain the portrait verification result, and uses the project review model corresponding to the declared project to verify the activity scene of each project activity picture to obtain the scene verification result, and determines the picture verification result based on the portrait verification result and the scene verification result, thereby realizing the automated verification of picture proof materials and improving the verification efficiency of picture proof materials; verifies picture materials from the picture quantity verification dimension, improves the verification accuracy of picture proof materials, and thus improves the verification accuracy of project proof materials of declared projects.

[0165] Figure 4 The process of the method for determining user attribute data provided by the embodiment of the present invention is as follows Figure 4 The embodiment of the present invention optimizes and improves the method of determining the target credit attribute data of the target user based on the above embodiment, such as Figure 4As shown, the method for determining user attribute data includes the following steps:

[0166] S410, obtaining the project certification materials of the application project uploaded by the target user; the project certification materials include text certification materials and picture certification materials; the picture certification materials include user portrait pictures and at least one project activity picture;

[0167] S420. Perform text verification on the textual certification materials according to the project keywords and items to be verified corresponding to the declared project, and obtain text verification results.

[0168] S430, using the user portrait picture to perform user portrait verification on each project activity picture to obtain a portrait verification result.

[0169] S440. According to the project review model corresponding to the application project, the activity scene of each project activity picture is verified to obtain the scene verification result.

[0170] S450: Determine the image verification result according to the portrait verification result and the scene verification result.

[0171] S460: If both the text verification result and the image verification result are verified to be passed, the current credit attribute data of the target user is obtained.

[0172] S470, obtaining an activity project credit attribute data relationship table; the activity project credit attribute data relationship table includes activity project credit attribute data corresponding to at least one activity project.

[0173] The activity project credit attribute data may specifically be the activity project credit score. Different activity projects have different activity project credit scores, which may be pre-set by relevant technical personnel based on factors such as the frequency of activity participation, the difficulty of the activity process, or the degree of activity impact. For example, the activity project credit score corresponding to activity project A is 3 points, and the activity project credit score corresponding to activity project B is 2 points, etc.

[0174] The activity project credit attribute data relationship table stores activity project credit attribute data corresponding to at least one activity project, and the relationship table can be continuously updated as the activity project is updated and pre-stored in the corresponding database.

[0175] S480. Determine the credit attribute data of the declared project corresponding to the declared project according to the active project credit attribute data relationship table.

[0176] The declared project credit attribute data may be the activity project credit attribute data corresponding to the activity project declared by the target user.

[0177] Exemplarily, the credit attribute data of the declared project corresponding to the declared project can be determined according to the mapping relationship between each active project and the credit attribute data of the active project pre-stored in the credit attribute data relationship table of the active project. Specifically, if the credit attribute data relationship table of the active project is as shown in Table 1.

[0178] Table 1

[0179]

[0180] According to Table 1, if the declared project is active project B, the corresponding declared project credit score is 2 points; if the declared project is active project C, the corresponding declared project credit score is 5 points.

[0181] S490: Determine target credit attribute data of the target user based on the current credit attribute data of the target user and the declared project credit attribute data of the declared project.

[0182] The target credit attribute data is the target credit score of the target user. For example, if the current credit attribute data of the target user is a credit score of 30 points, and the credit attribute data of the declared project is a credit score of 3 points, then the target credit attribute data of the target user is a target credit score of 33 points.

[0183] The technical solution of this embodiment determines the declared project credit attribute data corresponding to the declared project based on the acquired active project credit attribute data relationship table, and determines the target credit attribute data of the target user based on the current credit attribute data of the target user and the declared project credit attribute data of the declared project, thereby realizing the automated determination of the target credit data, reducing human resources without manual calculation, improving the determination efficiency, and being unaffected by human subjective factors, thereby improving the determination accuracy of the target credit attribute data to a certain extent.

[0184] Furthermore, to meet the needs of actual activity scenarios, the activity project credit attribute data relationship table can be updated regularly. In an optional embodiment, the activity project credit attribute data relationship table is updated as follows:

[0185] Step e1: Obtain historical declared items of historical users within a preset historical time period.

[0186] The preset historical time period may be preset by relevant technical personnel based on actual experience or experimental values, for example, the preset historical time period may be 3 months, 6 months or 1 year, etc. The historical application items are activity items that have been applied for by historical users in the historical period.

[0187] Step e2: Determine the historical project application volume under different project types based on the project types of the historical application projects.

[0188] In a specific embodiment, based on the time sequence, the historical application projects obtained include: Project A, Project B, Project C, Project A, Project C, Project C, Project B, Project C and Project A. Project A, Project B and Project C are active projects of different project types. It can be seen that the historical project application amount of Project A is 3 times, the historical project application amount of Project B is 2 times, and the historical project application amount of Project C is 4 times.

[0189] Step e3: Update the activity project credit attribute data corresponding to each activity project in the activity project credit attribute data relationship table according to the historical project application volume.

[0190] It should be noted that for any type of activity project, if the number of applications for the activity project is large, it can be considered that the participation rate of the activity project is high. In order to balance the participation rate of different activity projects, rather than being limited to a specific activity project, the credit score corresponding to the activity project with a large number of participants can be set lower. In order to encourage more users to participate in other activity projects, the credit score of the activity project with a small number of participants or a more complex and difficult implementation process can be set higher, that is, credit attribute data. Therefore, the activity project credit attribute data corresponding to different activity projects can be updated based on the historical project application volume corresponding to different project types.

[0191] In an optional embodiment, based on the historical project application amount, the activity project credit attribute data corresponding to each activity project in the activity project credit attribute data relationship table is updated, including:

[0192] Step e31: sort the active projects according to the historical project application volume, and obtain the sorting sequence numbers corresponding to the active projects.

[0193] For example, different types of activity projects can be ranked according to the number of historical project applications, and the ranking numbers corresponding to the activity projects can be obtained. The smaller the number, the greater the number of historical project applications in the historical period, and the larger the number, the smaller the number of historical project applications in the historical period.

[0194] Specifically, if the number of historical project applications for active project A is 100, the number of historical project applications for active project B is 150, and the number of historical project applications for active project C is 120, by sorting active project A, active project B, and active project C from the most to the least, we can get the sorting number corresponding to active project A is 01, the sorting number corresponding to active project C is 02, and the sorting number corresponding to active project B is 03.

[0195] Step e32: Determine the adjustment range of the attribute data corresponding to each activity item according to each sorting sequence number.

[0196] It is understandable that the smaller the ranking number, the higher the participation in the activity project; the larger the ranking number, the lower the participation in the activity project. Therefore, a lower attribute data adjustment range can be set for a smaller ranking number, and a higher attribute data adjustment range can be set for a larger ranking number. Optionally, the attribute data of the top few with smaller ranking numbers may not be adjusted.

[0197] In a specific embodiment, if the sorting number of activity item A is 01, the sorting number of activity item B is 02, the sorting number of activity item C is 03, the sorting number of activity item D is 04, and the sorting number of activity item E is 05. Since the sorting numbers of activity items A and B are smaller, their attribute data are not adjusted. The adjustment range of the attribute data of activity item C can be set to 5%, the adjustment range of the attribute data of activity item D can be set to 7%, and the adjustment range of the attribute data of activity item E can be set to 10%.

[0198] Step e33: update the activity project credit attribute data corresponding to each activity project according to the adjustment range of each attribute data, and obtain updated activity project credit attribute data.

[0199] Specifically, if the activity project credit score corresponding to activity project A is 3 points and the attribute data adjustment range is 5%, then the activity project credit score of activity project A after adjustment is 3.15 points, which can be obtained by adding the original score to the original score multiplied by the adjustment range.

[0200] By determining the adjustment range of attribute data based on the historical project declaration volume, combining scenario factors, human factors and other subjective factors of different activity types, the adjustment of credit attribute data of activity projects of different activity projects is rationalized.

[0201] The above technical solution obtains the historical declared projects of historical users in a preset historical time period, determines the historical project declared quantities under different project types according to the project types of the historical declared projects, and updates the active project credit attribute data corresponding to each active project in the active project credit attribute data relationship table according to the historical project declared quantities, thereby realizing the dynamic update of the active project credit attribute data relationship table, thereby improving the accuracy of the declared project credit attribute data of the declared projects determined based on the active project credit attribute data relationship table in the subsequent process, and making the determined target credit attribute data of the target users more reasonable.

[0202] It is understandable that after the target user declares a project and obtains his own target credit attribute data, there may be other users participating in related activities who question or object to the target credit attribute data owned by the target user. For example, there may be reasons such as fraud in participating in the activity that affect the authenticity of the declared project. Therefore, when other related users initiate a request to object to the declaration of the target user, the target credit attribute data of the target user can be further traced in response to the related user's request to object to the declaration.

[0203] Furthermore, after determining the target credit attribute data of the target user, the method further includes:

[0204] Step f1: During the project publicity period of the application project, if an application objection request is received from a relevant user, the application objection request is parsed to obtain objection certification information.

[0205] It should be noted that each activity project declared by different users has its corresponding project publicity period. The project publicity periods of different activity projects can be the same or different, and can be pre-set by relevant technical personnel. For example, the project publicity period corresponding to activity project A is 30 days, and the project publicity period corresponding to activity project B is 20 days, etc. During the project publicity period, any relevant user can submit a declaration objection request to the project declared by the target user and upload the objection proof information to verify the authenticity and validity of the declared project.

[0206] Exemplarily, when receiving a declaration objection request initiated by a relevant user, it is verified whether the request acquisition time of the objection request is within the project publicity period of the declaration project. If so, the declaration objection request is further parsed to obtain the objection proof information; if not, the declaration objection request can be deemed as an invalid request.

[0207] Among them, the objection proof information may include text or pictures of the objection proof reasons, as well as valid user identity proof information of the relevant user, etc.

[0208] Step f2: Send the objection proof information to the information verification platform for information verification, and obtain the information verification result fed back by the information verification platform.

[0209] Among them, the information verification platform can be a platform for users to verify the authenticity of the information of the objection proof. The specific verification includes verifying the authenticity of the information of the user identity validity proof information of the relevant user, and verifying whether the text or picture of the objection proof reason is valid.

[0210] The information verification result may include whether the objection certificate is valid or invalid. If the authenticity of the information of the valid user identity certificate of the relevant user is verified, and the text or picture of the objection certificate is verified, the objection certificate is valid. If the authenticity of the information of the valid user identity certificate of the relevant user is not verified, or the text or picture of the objection certificate is not verified, the objection certificate is invalid.

[0211] Step f3: If the information verification result is that the objection proof is valid, the target credit attribute data of the target user is updated based on the preset credit attribute data penalty mechanism.

[0212] If the information verification result is that the objection proof is invalid, the objection invalidation verification result will be fed back to the relevant personnel. If the information verification result is that the objection proof is valid, the target credit attribute data of the target user will be updated based on the preset information attribute data penalty mechanism.

[0213] Among them, the information attribute data penalty mechanism can be pre-set by relevant technical personnel, and specifically can be deducted from the target credit attribute data, that is, the target credit score, and the deduction intensity can be set based on different activity projects. For example, the credit score deduction mechanism for activity project A is to deduct 10 points, and the credit score deduction mechanism for activity project A is to deduct 8 points, etc. It can be specifically set according to the actual activity scenario, and this embodiment does not limit this.

[0214] The above technical solution proposes a penalty mechanism for deducting points from the target credit attribute data of non-standard target users, which can effectively avoid the occurrence of malicious application for project declaration in exchange for credit points, thereby maintaining the fairness of the determination of the user's credit attribute data to a certain extent, and further can mobilize the enthusiasm of users to participate in activity projects to a certain extent.

[0215] Furthermore, after the time specified in the project publicity period of the declared project is reached, the target credit attribute data of the target user can be put into storage. Specifically, according to the upload request of the declared project of the target user, the partition key value can be uploaded in the request header to characterize the uniqueness of the user identifier. When performing the storage operation, the database table serial number can be calculated according to the partition key value, and the storage operation of the target credit attribute data can be performed based on the database table serial number. Among them, at least one database can be numbered in advance, and the database tables corresponding to each database can be numbered in advance. Different user identifiers correspond to corresponding databases. Specifically, a mapping relationship between the user identifier and the database number can be established, and when performing the storage operation, the database and database table to be written can be determined based on the user number.

[0216] Furthermore, the credit attribute data of each user stored in the database can be sorted to obtain a list of abnormal users and a list of normal users, and the abnormal user list and the normal user list can be made public so that relevant users can apply for activity projects after finding out their own list, so as to adjust the user's enthusiasm for applying for projects. By determining the abnormal user list and the normal user list, it is also convenient for subsequent data analysis operations on the user's credit attribute data.

[0217] Figure 5 Schematic diagram of the structure of the device for determining user attribute data provided by an embodiment of the present invention. Figure 5 As shown, the user attribute data determination device includes: a certification material acquisition module 501, a text verification module 502, a portrait verification module 503, a scene verification module 504, a picture verification module 505, a current credit score determination module 506 and a target credit data determination module 507;

[0218] Among them, the certification material acquisition module 501 is used to obtain the project certification materials of the application project uploaded by the target user; the project certification materials include text certification materials and picture certification materials; the picture certification materials include user portrait pictures and at least one project activity picture;

[0219] The text verification module 502 is used to perform text verification on the text certification materials according to the project keywords and items to be verified corresponding to the declared project, and obtain a text verification result;

[0220] A portrait verification module 503 is used to use the user portrait picture to perform user portrait verification on each of the project activity pictures to obtain a portrait verification result; and

[0221] The scene verification module 504 is used to verify the activity scene of each project activity picture according to the project review model corresponding to the application project, and obtain the scene verification result;

[0222] The picture verification module 505 is used to determine the picture verification result according to the portrait verification result and the scene verification result;

[0223] The current credit score determination module 506 is used to obtain the current credit attribute data of the target user if both the text verification result and the image verification result are verified to be passed;

[0224] The target credit data determination module 507 is used to determine the declared project credit attribute data corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit score of the target user and the declared project credit attribute data.

[0225] The embodiment of the present invention obtains the project certification materials of the declared project uploaded by the target user, and performs text verification on the text certification materials according to the project keywords and the items to be verified corresponding to the declared project to obtain the text verification results, and performs image verification on the image certification materials according to the project review model corresponding to the declared project to obtain the image verification results, thereby realizing the automated review of the project certification materials, reducing the occupation of human resources, and improving the efficiency, accuracy and reliability of the review of the active project. If both the text verification result and the image verification result are verified, the current credit attribute data of the target user is obtained, the credit attribute data of the declared project corresponding to the declared project is determined, and the target credit attribute data of the target user is determined according to the current credit attribute data of the target user and the credit attribute data of the declared project, thereby realizing the automated determination of the target credit attribute data of the target user, and improving the determination efficiency, accuracy and reliability. According to the portrait verification result and the scene verification result, the picture verification result is determined, thereby realizing the automated verification of the picture certification materials, and improving the verification efficiency of the picture certification materials; the picture material verification is performed from the two verification dimensions of user portrait and activity scene, thereby improving the verification accuracy of the picture certification materials, thereby improving the verification accuracy of the project certification materials of the declared project.

[0226] Furthermore, the text verification module 502 includes:

[0227] A description information verification unit, used to perform text verification on the project description information in the textual certification material according to the project keywords corresponding to the declared project, and obtain a description information verification result; and

[0228] An identity information verification unit, used to perform text verification on the identity information in the textual certification material according to the to-be-verified item corresponding to the declared item, and obtain an identity information verification result;

[0229] The text verification unit is used to determine the text verification result according to the description information verification result and the identity information verification result.

[0230] Furthermore, the description information verification unit includes:

[0231] A keyword extraction subunit, configured to extract keywords from the project description information in the textual evidence material to obtain at least one description keyword;

[0232] A consistency comparison subunit, used to compare the consistency of each of the description keywords with the project keywords, and determine a first target keyword that meets the consistency comparison condition;

[0233] A first proportion determination subunit is used to determine a first keyword proportion according to the first keyword quantity of the first target keyword and the description keyword quantity of the description keyword;

[0234] A similarity comparison subunit is used for comparing the similarity of other keywords with the project keyword to determine a second target keyword that meets the similarity comparison condition if the proportion of the first keyword does not meet the preset proportion judgment condition; the other keywords are description keywords other than the first target keyword in each of the description keywords;

[0235] A second proportion determination subunit is used to determine the proportion of the second keyword according to the number of second keywords of the second target keyword and the number of description keywords of the description keyword;

[0236] The text verification subunit is used to perform text verification on the project description information in the text proof material according to the proportion of the second keyword to obtain a description information verification result.

[0237] Furthermore, the identity information verification unit includes:

[0238] An integrity verification subunit, configured to perform integrity verification on the identity verification item in the identity verification information according to the item to be verified corresponding to the declared item, and obtain a text integrity verification result;

[0239] A consistency verification subunit, configured to perform consistency verification on the proof text corresponding to the identity proof item according to the data type corresponding to the item to be verified, and obtain a text consistency verification result;

[0240] The identity information verification subunit is used to determine the identity information verification result according to the text integrity verification result and the text consistency verification result.

[0241] Furthermore, the device further comprises:

[0242] A picture quantity acquisition module, used for obtaining the number of pictures required for the project corresponding to the declared project before performing user portrait verification on each project activity picture using the user portrait picture to obtain the portrait verification result;

[0243] A picture quantity verification module is used to verify the number of pictures of the project activity pictures according to the number of pictures required by the project, and obtain a picture quantity verification result;

[0244] The user portrait verification module is used to perform user portrait verification on each of the project activity pictures using the user portrait pictures if the picture quantity verification result is passed, so as to obtain a portrait verification result.

[0245] Furthermore, the user portrait verification module includes:

[0246] A portrait recognition unit, used to perform portrait recognition on the project activity picture to obtain at least one portrait area image corresponding to the project activity picture;

[0247] An image similarity comparison unit, used to compare the image similarity of each of the portrait area images with the user portrait image in the user portrait picture, and obtain similarity comparison results corresponding to each of the portrait area images in the project activity picture;

[0248] The user portrait verification unit is used to perform user portrait verification on the project activity picture according to the similarity comparison results respectively corresponding to the portrait area images in the project activity picture to obtain a portrait verification result.

[0249] Furthermore, the device further comprises: a model training module, which is used to perform model training on the project review model corresponding to the application project; the model training module comprises:

[0250] A historical picture acquisition unit, used to acquire historical project activity pictures of the declared project in different activity scenes and different shooting angles during a historical time period, and use the historical project activity pictures as positive sample data;

[0251] A negative sample generating unit, used to generate negative sample data corresponding to the positive sample data;

[0252] The model training unit is used to use the positive sample data and the negative sample data to perform model training on a pre-constructed network model until a preset model training end condition is met, thereby obtaining a model of a project review model corresponding to the declared project.

[0253] Furthermore, the target credit data determination module 505 includes:

[0254] A relationship table acquisition unit, used to acquire an activity project credit attribute data relationship table; the activity project credit attribute data relationship table includes activity project credit attribute data corresponding to at least one activity project;

[0255] An attribute data determination unit, used to determine the declared project credit attribute data corresponding to the declared project according to the active project credit attribute data relationship table;

[0256] The target credit data determination unit is used to determine the target credit attribute data of the target user according to the current credit attribute data of the target user and the declared project credit attribute data of the declared project.

[0257] Furthermore, the target credit data determination module 505 further includes: a relationship table updating unit, which is used to update the activity project credit attribute data relationship table; the relationship table updating unit includes:

[0258] A historical project acquisition subunit is used to acquire historical declared projects of historical users within a preset historical time period;

[0259] The historical application quantity determination subunit is used to determine the historical project application quantity under different project types according to the project types of the historical application projects;

[0260] The relationship table updating subunit is used to update the active project credit attribute data corresponding to each active project in the active project credit attribute data relationship table according to the historical project declaration amount.

[0261] Furthermore, the relationship table updating subunit is specifically used for:

[0262] According to the historical project application amount, the activity projects are sorted to obtain the sorting numbers corresponding to the activity projects;

[0263] Determine the adjustment range of the attribute data corresponding to each of the activity items according to each of the sorting numbers;

[0264] According to the adjustment range of each of the attribute data, the activity project credit attribute data corresponding to each of the activity projects is updated to obtain updated activity project credit attribute data.

[0265] The user attribute data determination device provided in the embodiment of the present invention can execute the user attribute data determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0266] Figure 6 A schematic diagram of the structure of an electronic device 60 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.

[0267] like Figure 6As shown, the electronic device 60 includes at least one processor 61, and a memory connected to the at least one processor 61, such as a read-only memory (ROM) 62, a random access memory (RAM) 63, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 61 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 62 or the computer program loaded from the storage unit 68 to the random access memory (RAM) 63. In the RAM 63, various programs and data required for the operation of the electronic device 60 can also be stored. The processor 61, the ROM 62, and the RAM 63 are connected to each other via a bus 64. An input / output (I / O) interface 65 is also connected to the bus 64.

[0268] A number of components in the electronic device 60 are connected to the I / O interface 65, including: an input unit 66, such as a keyboard, a mouse, etc.; an output unit 67, such as various types of displays, speakers, etc.; a storage unit 68, such as a disk, an optical disk, etc.; and a communication unit 69, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 69 allows the electronic device 60 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0269] The processor 61 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 61 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, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 61 performs the various methods and processes described above, such as a method for determining user attribute data.

[0270] In some embodiments, the user attribute data determination method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 68. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 60 via the ROM 62 and / or the communication unit 69. When the computer program is loaded into the RAM 63 and executed by the processor 61, one or more steps of the user attribute data determination method described above may be performed. Alternatively, in other embodiments, the processor 61 may be configured to perform the user attribute data determination method in any other appropriate manner (e.g., by means of firmware).

[0271] 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.

[0272] 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.

[0273] 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.

[0274] 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).

[0275] 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.

[0276] 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.

[0277] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the user attribute data determination method provided by any embodiment of the present invention.

[0278] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0279] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining user attribute data, characterized in that: include: Obtain the project certification materials uploaded by the target user; The project supporting materials include text supporting materials and picture supporting materials; The picture evidence materials include a user portrait picture and at least one project activity picture; According to the project keywords and items to be verified corresponding to the declared project, the textual certification materials are verified to obtain textual verification results; Using the user portrait pictures to perform user portrait verification on each of the project activity pictures to obtain a portrait verification result; and, According to the project review model corresponding to the declared project, the activity scene verification is performed on the activity pictures of each project to obtain the scene verification result; Determine the image verification result according to the portrait verification result and the scene verification result; If both the text verification result and the image verification result are verified to be passed, obtaining the current credit attribute data of the target user; Determine the declared project credit attribute data corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit attribute data of the target user and the declared project credit attribute data.

2. The method according to claim 1, characterized in that The text verification of the textual certification materials is performed according to the project keywords and items to be verified corresponding to the declared project to obtain textual verification results, including: According to the project keywords corresponding to the declared project, the project description information in the textual certification materials is verified to obtain the description information verification result; and According to the items to be verified corresponding to the declared items, the identity information in the textual certification materials is verified to obtain the identity information verification result; A text verification result is determined according to the description information verification result and the identity information verification result.

3. The method according to claim 2, characterized in that The project description information in the textual certification material is verified according to the project keywords corresponding to the declared project to obtain the description information verification result, including: Extracting keywords from the project description information in the textual evidence material to obtain at least one description keyword; Comparing the consistency of each of the description keywords with the project keywords to determine a first target keyword that meets the consistency comparison condition; Determine a first keyword proportion according to the first keyword quantity of the first target keyword and the description keyword quantity of the description keyword; If the proportion of the first keyword does not meet the preset proportion judgment condition, then compare the similarity between other keywords and the project keyword to determine a second target keyword that meets the similarity comparison condition; the other keywords are description keywords other than the first target keyword in each of the description keywords; Determine a proportion of the second keyword according to the number of second keywords of the second target keyword and the number of description keywords of the description keyword; According to the proportion of the second keyword, text verification is performed on the project description information in the text certification material to obtain a description information verification result.

4. The method according to claim 2, characterized in that: The text verification of the identity information in the textual certification material is performed according to the items to be verified corresponding to the declared items to obtain the identity information verification result, including: According to the items to be verified corresponding to the declared items, the identity verification items in the identity verification information are verified for integrity to obtain a text integrity verification result; According to the data type corresponding to the item to be verified, a consistency check is performed on the proof text corresponding to the identity proof item to obtain a text consistency verification result; An identity information verification result is determined based on the text integrity verification result and the text consistency verification result.

5. The method according to claim 1, characterized in that Before the user portrait verification is performed on each of the project activity pictures using the user portrait pictures to obtain the portrait verification result, the method includes: Obtain the number of pictures required for the project corresponding to the declared project; According to the number of pictures required for the project, verify the number of pictures of the project activity pictures to obtain the picture number verification result; If the result of the picture quantity verification is passed, the user portrait verification of each of the project activity pictures is performed using the user portrait picture to obtain a portrait verification result.

6. The method according to claim 1, characterized in that The user portrait pictures are used to perform user portrait verification on each of the project activity pictures to obtain a portrait verification result, including: Performing portrait recognition on the project activity picture to obtain at least one portrait area image corresponding to the project activity picture; Comparing the image similarity of each of the portrait area images with the user portrait image in the user portrait picture, and obtaining the similarity comparison results corresponding to each of the portrait area images in the project activity picture; According to the similarity comparison results respectively corresponding to the portrait area images in the project activity picture, the project activity picture is subjected to user portrait verification to obtain a portrait verification result.

7. The method according to claim 1, characterized in that The model training method of the project review model corresponding to the application project is as follows: Obtaining historical project activity pictures of the declared project in different activity scenes and shooting angles during the historical time period, and using the historical project activity pictures as positive sample data; Generate negative sample data corresponding to the positive sample data; The positive sample data and the negative sample data are used to perform model training on a pre-constructed network model until a preset model training end condition is met, thereby obtaining a project review model corresponding to the declared project.

8. The method according to claim 1, characterized in that The step of determining the declared project credit attribute data corresponding to the declared project, and determining the target credit attribute data of the target user according to the current credit attribute data of the target user and the declared project credit attribute data, includes: Acquire an activity project credit attribute data relationship table; the activity project credit attribute data relationship table includes activity project credit attribute data corresponding to at least one activity project; Determine the credit attribute data of the declared project corresponding to the declared project according to the active project credit attribute data relationship table; The target credit attribute data of the target user is determined according to the current credit attribute data of the target user and the declared project credit attribute data of the declared project.

9. The method according to claim 8, characterized in that The update method of the activity project credit attribute data relationship table is as follows: Obtain historical declared items of historical users within a preset historical time period; According to the project types of historically applied projects, determine the application volume of historical projects under different project types; According to the historical project declaration amount, the activity project credit attribute data corresponding to each activity project in the activity project credit attribute data relationship table is updated.

10. The method according to claim 9, characterized in that The updating of the activity project credit attribute data corresponding to each activity project in the activity project credit attribute data relationship table according to the historical project application amount includes: According to the historical project application amount, the activity projects are sorted to obtain the sorting numbers corresponding to the activity projects; Determine the adjustment range of the attribute data corresponding to each of the activity items according to each of the sorting numbers; According to the adjustment range of each of the attribute data, the activity project credit attribute data corresponding to each of the activity projects is updated to obtain updated activity project credit attribute data.

11. The method according to claim 1, characterized in that: After determining the target credit attribute data of the target user, the method further includes: During the project publicity period of the application project, if an application objection request initiated by a relevant user is received, the application objection request is parsed to obtain objection certification information; Sending the objection proof information to the information verification platform for information verification, and obtaining the information verification result fed back by the information verification platform; If the information verification result is that the objection proof is valid, the target credit attribute data of the target user is updated based on a preset credit attribute data penalty mechanism.

12. A device for determining user attribute data, characterized in that: include: The certification material acquisition module is used to obtain the project certification materials of the declared project uploaded by the target user; The project supporting materials include text supporting materials and picture supporting materials; The picture evidence materials include a user portrait picture and at least one project activity picture; A text verification module, used to perform text verification on the textual certification materials according to the project keywords and items to be verified corresponding to the declared project, and obtain text verification results; A portrait verification module, used to use the user portrait picture to perform user portrait verification on each of the project activity pictures to obtain a portrait verification result; and A scene verification module is used to verify the activity scene of each project activity picture according to the project review model corresponding to the application project, and obtain the scene verification result; An image verification module, used to determine an image verification result according to the portrait verification result and the scene verification result; A current credit score determination module, configured to obtain the current credit attribute data of the target user if both the text verification result and the image verification result are verified to be passed; The target credit data determination module is used to determine the declared project credit attribute data corresponding to the declared project, and determine the target credit attribute data of the target user according to the current credit score of the target user and the declared project credit attribute data.

13. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the user attribute data determination method as described in any one of claims 1 to 11 when executing the computer program.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for determining user attribute data as described in any one of claims 1-11 is implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the method for determining user attribute data as described in any one of claims 1-11.

Images