ID Mapping method, device and system based on similarity calculation, and storage medium

Abstract

The invention is suitable for the technical field of internet advertisements, and provides an ID Mapping method and device based on similarity calculation, a storage medium and a system.The method comprises the steps that a plurality of user data are extracted from a plurality of service systems to a data middle station, and a user domain is obtained; obtaining user characteristics of each piece of user data in the user domain; based on the user features, performing similarity calculation to obtain a similarity calculation result; and obtaining the similarity calculation result, and judging that the user features with the similarity calculation result exceeding a preset threshold value belong to the user features of the same source. Compared with the prior art, the method does not depend on a large amount of real-time calculation, similarity comparison is carried out on the collected data in the user platform, which data originates from the same user can be rapidly judged based on a comparison result, and higher user matching precision is obtained on the premise that the calculation amount is not remarkably increased.

Description

Technical Field

[0001] The present invention belongs to the technical field of Internet advertising, and in particular relates to an ID Mapping method device, storage medium and system based on similarity calculation. Background Art

[0002] With the development of Internet technology, compared with traditional advertising, modern advertising is not only diverse in type, but also has richer channels for dissemination. Users will use different terminals to access advertisements. How to identify the same user as a user ID from different advertisements, different channels and different devices according to certain rules is the key to achieving accurate advertising delivery. This process is called ID-Mapping. When users access relevant advertisements through terminals, identification information such as device model, cookieid, mac, imei and imsi will be generated accordingly. Due to different data sources, some scenarios have relevant identifications, while some scenarios do not. At this time, it is necessary to generate a unique user ID through ID-Mapping among the complex IDs.

[0003] At present, the commonly used solution for ID-Mapping is to use graph computing to find the association between various ID identifiers, so as to identify them as the same user. Its core idea is the shortest path planning algorithm in graph theory, which specifically describes the ID identifier as a point, forms "edges" between points through actual business processes, finds data relationships from points and edges, forms connectivity, and finally forms an ID mapping field to obtain a unique user ID.

[0004] In the existing technology, when ID mapping is performed in multiple business systems, the system will face challenges in data consistency and data synchronization. If data synchronization is not timely, the user that should have been identified as the same user will be identified as two users, which will have a great impact on subsequent advertising. As the amount of data increases, such algorithms become complex and the amount of calculation is likely to reach a bottleneck, which may degrade from the original real-time calculation to T+1 timeliness. Therefore, it is necessary to improve the existing ID mapping method. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide an ID Mapping method based on similarity calculation, aiming to solve the problem that high-precision user matching in existing ID Mapping methods often requires a high amount of calculation, and it is difficult to effectively balance the amount of calculation and the matching accuracy.

[0006] The embodiment of the present application is implemented by providing an ID Mapping method based on similarity calculation, the method comprising: Extract some user data from some business systems to the data center to obtain the user domain; Obtaining user characteristics of each piece of user data in the user domain; Based on the user characteristics, similarity calculation is performed to obtain a similarity calculation result; The similarity calculation result is obtained, and the user features whose similarity calculation results exceed a preset threshold are determined to be user features belonging to the same source.

[0007] Preferably, a method of extracting some user data from some business systems to the data center to obtain the user domain is: Synchronize data from multiple business systems to obtain a number of user data, wherein the user data includes user attribute data and behavior data; Cleaning and regularizing the user data to obtain integrated data; The integrated data is loaded into the data center to obtain the user domain.

[0008] Preferably, the attribute data includes one or more of the following data: the user's gender, age, contact number, ID card, device identification number, and device operating system code; The behavior data includes one or more of the following data: the user's access records, collection records, purchase history records, and the page IDs visited by the user.

[0009] Preferably, the method for obtaining the user characteristics of each piece of the user data in the user domain is: Converting the initial feature set formed based on the user data into a data matrix through linear transformation; Based on the data matrix, obtain the covariance matrix: Where X is the data matrix, where each row is a sample and each column is a feature; the standardized data matrix is , the covariance matrix is ​​C, n is the number of samples, is the transpose of the normalized data matrix; Solve for the eigenvalues ​​of the covariance matrix C and the corresponding eigenvector ; Sort all the eigenvalues ​​and eigenvectors by eigenvalue from large to small, select the eigenvectors corresponding to the first k eigenvalues, and form a new eigenspace with them to obtain the principal component space ; The standardized data matrix X′ is compared with the principal component space Multiply to get the data features after dimensionality reduction : The data features after dimension reduction are set as user features.

[0010] Preferably, the method for obtaining the number of parameters k is: Get the cumulative variance contribution rate M: Where d is the total number of features, is the i-th eigenvalue, t is the parameter; Based on the variance contribution rate M, a scree plot was constructed, and t that could explain 85%-95% of the variance was set as the value of k.

[0011] Preferably, based on the user characteristics, similarity calculation is performed to obtain the similarity calculation result by: Get any two of the user features and obtain the cosine similarity calculation result C between the two user features: , Among them, a and b are two user features whose cosine similarity results are to be calculated.

[0012] Preferably, when the cosine similarity result exceeds a similarity threshold, it is determined that the two user features involved in the similarity calculation belong to the same user; The value interval of the similarity threshold is [0.94-0.95].

[0013] Another object of an embodiment of the present application is to provide an ID Mapping device based on similarity calculation, the ID Mapping device based on similarity calculation comprising: The user domain acquisition module is used to extract some user data from some business systems to the data center to obtain the user domain; A user feature acquisition module, used to acquire user features of each piece of user data in the user domain; A similarity calculation acquisition module is used to perform similarity calculation based on the user characteristics to obtain a similarity calculation result; The same source determination module is used to obtain the similarity calculation result and determine the user features whose similarity calculation results exceed a preset threshold as user features belonging to the same source.

[0014] Another object of an embodiment of the present application is to provide a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the processor executes the steps of the ID Mapping method based on similarity calculation as described above.

[0015] Another object of an embodiment of the present application is to provide an ID Mapping system based on similarity calculation, wherein when the system is running, the steps of the ID Mapping method based on similarity calculation as described above are executed.

[0016] An ID Mapping method based on similarity calculation provided in an embodiment of the present application has the outstanding advantage that: this solution adopts a method of first synchronizing all data to the data center for unified calculation and then performing similarity comparison, and does not rely on a large amount of real-time calculation. The collected data is compared for similarity in the user center, and it can quickly determine which data originates from the same user based on the comparison results, thereby obtaining a higher user matching accuracy without significantly increasing the amount of calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 An application environment diagram of an ID Mapping method based on similarity calculation provided in an embodiment of the present application; Figure 2 A flowchart of an ID Mapping method based on similarity calculation provided in an embodiment of the present application; Figure 3 A similarity threshold-accuracy curve diagram provided in an embodiment of the present application; Figure 4 A structural block diagram of an ID Mapping device based on similarity calculation provided in an embodiment of the present application; Figure 5 FIG. 4 is a block diagram of the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first script may be referred to as a second script, and similarly, a second script may be referred to as a first script without departing from the scope of this application.

[0020] Figure 1 The application environment diagram of the ID Mapping method based on similarity calculation provided in the embodiment of the present application is as follows: Figure 1 As shown, in the application environment, a terminal 110 and a computer device 120 are included.

[0021] The computer device 120 may be an independent physical server or terminal, or a server cluster composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud servers, cloud databases, cloud storage, and CDN.

[0022] The terminal 110 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal 110 and the computer device 120 may be connected via a network, and this application does not limit this.

[0023] like Figure 2 As shown, in one embodiment, an ID Mapping method based on similarity calculation is proposed. This embodiment mainly applies this method to the above Figure 1 An ID Mapping method based on similarity calculation may specifically include the following steps: Step S10, extracting some user data from some business systems to the data center to obtain the user domain.

[0024] In this embodiment, multi-source data is aggregated into a unified platform to lay the foundation for subsequent similarity calculations. User data stored in multiple business systems is usually scattered, so this application uses the data center as the core of data integration to ensure that data from different sources can be shared and interacted.

[0025] Step S20: obtaining user characteristics of each piece of user data in the user domain.

[0026] In this embodiment, user features are basic information used to describe the user's identity, which may include explicit features such as user name, mobile phone number, and email address, and implicit features such as user behavior and transaction history. These features can help accurately identify and match the same user in different business systems. For the extraction of user behavior and interaction data, it is helpful to deal with users who do not have obvious unified identifiers.

[0027] Step S30: performing similarity calculation based on the user features to obtain a similarity calculation result.

[0028] In this embodiment, although the ID Mapping method based on machine learning in the prior art (such as clustering, classification model, etc.) can also perform user matching, it usually requires a large amount of labeled data for training, or requires a large amount of real-time computing, and has poor adaptability to new data. Similarity calculation is the core step in this application, and commonly used methods include Euclidean distance, cosine similarity, Jaccard similarity, etc. This step calculates the similarity between user features to identify user data that are highly similar in features. This step helps solve the problem of user identity matching across systems, especially when the same user has different identifiers in different systems.

[0029] Step S40, obtaining the similarity calculation result, and determining the user features whose similarity calculation results exceed a preset threshold as user features belonging to the same source.

[0030] In the embodiment of the present application, this step is the final matching process, which determines whether two user data are of the same user by setting a threshold (for example, 0.945). If the similarity exceeds the threshold, they are considered to represent the same user. The setting of the threshold is critical and needs to be adjusted according to the actual situation. If it is too low, it may lead to misjudgment, and if it is too high, it may miss the real match.

[0031] In an embodiment of the present application, by extracting user data from multiple business systems and performing similarity calculations, the same users from different systems can be effectively identified and merged. Compared with traditional matching methods based on unique identifiers (such as IDs, mobile phone numbers, etc.), it can cope with situations where users do not have a unified identifier or have different identifiers in different systems. This is a problem often encountered in cross-system ID mapping. This solution adopts a method of first synchronizing all data to the data center for unified calculations and then performing similarity comparisons. It does not rely on a large amount of real-time calculations. It performs similarity comparisons on the collected data in the user center, and can quickly determine which data originates from the same user based on the comparison results, thereby obtaining higher user matching accuracy without significantly increasing the amount of calculations.

[0032] In a preferred embodiment, a method of extracting some user data from some business systems to the data center to obtain the user domain is as follows: Synchronize data from multiple business systems to obtain a number of user data, wherein the user data includes user attribute data and behavior data; Cleaning and regularizing the user data to obtain integrated data; The integrated data is loaded into the data center to obtain the user domain.

[0033] In the embodiment of the present application, the user attribute data and user behavior data in multiple business systems are extracted, cleaned, organized and uniformly loaded into the data center, and finally a complete and standardized user domain is obtained. As a unified data platform, the data center can aggregate multi-source data and provide data support for subsequent business analysis, user profiling, precision marketing, ID Mapping, etc.

[0034] In a preferred embodiment, the attribute data includes one or more of the following data: the user's gender, age, contact number, ID card, device identification number, and device operating system code; The behavior data includes one or more of the following data: the user's access records, collection records, purchase history records, and the page IDs visited by the user.

[0035] In the embodiment of the present application, both the attribute data and the behavior data may contain a large amount of characteristic data of the user, which may not be fully obtained from the business system. For example, in an anonymous business system, key features such as the user's contact number and ID number are difficult to obtain, so it is preferred to obtain all attribute data and behavior data from the business system as much as possible to improve the analysis accuracy of the system.

[0036] In a preferred embodiment, the method for obtaining the user characteristics of each piece of the user data in the user domain is: Converting the initial feature set formed based on the user data into a data matrix through linear transformation; Based on the data matrix, obtain the covariance matrix: Where X is the data matrix, where each row is a sample and each column is a feature; the standardized data matrix is , the covariance matrix is ​​C, n is the number of samples, is the transpose of the normalized data matrix; Solve for the eigenvalues ​​of the covariance matrix C and the corresponding eigenvector ; Sort all the eigenvalues ​​and eigenvectors by eigenvalue from large to small, select the eigenvectors corresponding to the first k eigenvalues, and form a new eigenspace with them to obtain the principal component space ; The standardized data matrix X′ is compared with the principal component space Multiply to get the data features after dimensionality reduction : The data features after dimension reduction are set as user features.

[0037] In the embodiments of the present application, the accuracy of obtaining user features is a key factor affecting the accuracy of ID Mapping. In order to further balance the requirements of computational complexity and matching accuracy, the present application improves the existing matching method. This solution first reduces the dimension of the data, effectively reducing the dimension of the feature space. The reduction of features not only reduces the complexity of storage and calculation, but also avoids the problem of "dimensionality disaster", making subsequent analysis more efficient. In addition, after dimensionality reduction, the redundancy between features is removed, the data becomes more streamlined, but at the same time the maximum amount of information (i.e., the features in the direction of maximum variance) is retained.

[0038] In this embodiment, the principal components are selected by eigenvalue sorting, so that the feature space after dimensionality reduction has stronger explanatory power. The selected principal components are usually the most important dimensions in the data that best reflect user characteristics, which helps to understand the inherent structure of the data. The system automatically identifies features with high variance and compresses or discards redundant or noisy features with low variance. This helps to reduce unnecessary effects caused by data noise and improve the accuracy and robustness of the model.

[0039] In a preferred embodiment of the present application, for obtaining user features, the method is as follows: first, the feature set is reduced in dimension, and 70% of the samples are classified as training sets, 20% of the samples are classified as test sets, and 10% are classified as verification sets, thereby obtaining a feature set. Among them, the steps of reducing the dimension are as follows: first, by calculating the covariance matrix, the correlation coefficient of the two features is obtained; then, after the eigenvector and eigenvalue of the technical covariance matrix, the principal component is selected according to the size of the eigenvalue, and it can be specified here to select the principal component with an eigenvalue> 1; finally, by multiplying the sample data by each eigenvector, the score of each principal component is obtained, and 10 main features are selected, so that the final feature set of the sample is obtained.

[0040] In a preferred embodiment, the method for obtaining the parameter number k is: Get the cumulative variance contribution rate M: Where d is the total number of features, is the i-th eigenvalue, t is the parameter; Based on the variance contribution rate M, a scree plot was constructed, and t that could explain 85%-95% of the variance was set as the value of k.

[0041] In the embodiment of the present application, the scree plot is a visualization technique used to help select the number of principal components k. The scree plot converts the eigenvalues ​​of each principal component into Plot them in descending order according to their size. By observing the "elbow point" of the scree plot, you can intuitively see the trend of the eigenvalue decrease. After the elbow, the rate of decrease of the eigenvalue slows down significantly, which shows that the gain in data explanatory power by adding more principal components becomes very small. Based on the cumulative variance contribution rate M, select the number of principal components t that can explain 85%-95% of the variance, that is, select k as the minimum t that satisfies the condition M∈[85%,95%].

[0042] In a preferred embodiment, similarity calculation is performed based on the user characteristics, and a method for obtaining a similarity calculation result is as follows: Get any two of the user features and obtain the cosine similarity calculation result C between the two user features: Among them, a and b are two user features whose cosine similarity results are to be calculated.

[0043] In the embodiment of the present application, the smaller the angle C is, the more similar the samples a and b are.

[0044] In a preferred embodiment, it is characterized in that when the cosine similarity result exceeds the similarity threshold, it is determined that the two user features involved in the similarity calculation belong to the same user; The value interval of the similarity threshold is [0.94-0.95].

[0045] In the embodiment of the present application, the cosine similarity between samples can be used. When the cosine similarity exceeds a threshold, they are classified as the same "person". For example, the method of threshold confirmation is: traverse all thresholds in the interval [0, 1] with a step size of 0.002, and count the accuracy of each time in the test set. Figure 3 As shown in FIG. 1 , a curve diagram of similarity threshold-accuracy is given, and the threshold corresponding to the highest accuracy is selected. Through experiments, this paper obtains that the interval [0.94-0.95] is the threshold value range of cosine similarity.

[0046] like Figure 4 As shown, in one embodiment, an ID Mapping device based on similarity calculation is provided. The ID Mapping device based on similarity calculation can be integrated into the above-mentioned computer device 120, and specifically may include: The user domain acquisition module 510 is used to extract a number of user data from a number of business systems to the data center to obtain the user domain; A user feature acquisition module 520, configured to acquire a user feature of each piece of user data in the user domain; A similarity calculation acquisition module 530 is used to perform similarity calculation based on the user characteristics to obtain a similarity calculation result; The same-source determination module 540 is used to obtain the similarity calculation result, and determine the user features whose similarity calculation results exceed a preset threshold as user features belonging to the same source.

[0047] In the embodiment of the present application, for the explanation and description of the above-mentioned ID Mapping device based on similarity calculation, please refer to the explanation and description of the above-mentioned corresponding method. For the description of the above-mentioned ID Mapping method based on similarity calculation, please refer to the above text and will not be repeated here.

[0048] In the embodiment of the present application, this solution adopts the method of first synchronizing all data to the data center for unified calculation and then performing similarity comparison. It does not rely on a large amount of real-time calculation. The collected data is compared for similarity in the user center, and it can quickly determine which data originates from the same user based on the comparison results, thereby obtaining a higher user matching accuracy without significantly increasing the amount of calculation.

[0049] Figure 5 The internal structure diagram of a computer device in one embodiment is shown. The computer device may specifically be Figure 1 The computer device 120 in FIG. Figure 5 As shown, the computer device includes a processor, a memory, a network interface, an input device and a display screen connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement an ID Mapping method based on similarity calculation. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute an ID Mapping method based on similarity calculation. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0050] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0051] In one embodiment, the ID Mapping device based on similarity calculation provided by the present application can be implemented in the form of a computer program. The computer program can be used in Figure 5 The memory of the computer device can store various program modules constituting the ID Mapping device based on similarity calculation, for example, Figure 4 The user domain acquisition module 510, the user feature acquisition module 520, etc. are shown. The computer program composed of various program modules enables the processor to execute the steps of the ID Mapping method based on similarity calculation in various embodiments of the present application described in this specification.

[0052] For example, Figure 5 The computer device shown can be Figure 4 The module 510 in the ID Mapping device based on similarity calculation shown in the figure executes step S10. The computer device can execute step S20 through the module 520. And so on.

[0053] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the processor executes the steps of the ID Mapping method based on similarity calculation as described above.

[0054] In the embodiment of the present application, please refer to the above for the description of the above-mentioned ID Mapping method based on similarity calculation, which will not be repeated here.

[0055] In the embodiment of the present application, the program executed by the method stored in the storage medium of the embodiment of the present application adopts the method of first synchronizing all data to the data center for unified calculation and then performing similarity comparison, which does not rely on a large amount of real-time calculation, performs similarity comparison on the collected data in the user center, and can quickly determine which data originates from the same user based on the comparison results, thereby obtaining a higher user matching accuracy without significantly increasing the amount of calculation. .

[0056] In one embodiment, an ID Mapping system based on similarity calculation is provided. When the system is running, the system executes the steps of the ID Mapping method based on similarity calculation as described above.

[0057] In the embodiment of the present application, the system can be a computer software or program system, which runs in hardware containing a processing device, and when the above system is running, the corresponding method is executed. For the description of the above IDMapping method based on similarity calculation, please refer to the above text, which will not be repeated here.

[0058] In the embodiment of the present application, based on the present system, the present solution adopts a method of first synchronizing all data to the data center for unified calculation and then performing similarity comparison. It does not rely on a large amount of real-time calculation. The collected data is compared for similarity in the user center, and it can quickly determine which data originates from the same user based on the comparison results, thereby obtaining a higher user matching accuracy without significantly increasing the amount of calculation.

[0059] It should be understood that, although each step in the flow chart of each embodiment of the present application is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is clear explanation in this article, the execution of these steps does not have strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0060] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0061] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An ID Mapping method based on similarity calculation, characterized in that: The method comprises: Extract some user data from some business systems to the data center to obtain the user domain; Obtaining user characteristics of each piece of user data in the user domain; Based on the user characteristics, similarity calculation is performed to obtain a similarity calculation result; The similarity calculation result is obtained, and the user features whose similarity calculation results exceed a preset threshold are determined to be user features belonging to the same source.

2. According to claim 1, an ID Mapping method based on similarity calculation is characterized in that: Extract some user data from some business systems to the data center, and obtain the user domain by: Synchronize data from multiple business systems to obtain a number of user data, wherein the user data includes user attribute data and behavior data; Cleaning and regularizing the user data to obtain integrated data; The integrated data is loaded into the data center to obtain the user domain.

3. The ID Mapping method based on similarity calculation according to claim 2, characterized in that: The attribute data includes one or more of the following data: the user's gender, age, contact number, ID card, device identification number, and device operating system code; The behavior data includes one or more of the following data: the user's access records, collection records, purchase history records, and the page IDs visited by the user.

4. The ID Mapping method based on similarity calculation according to claim 1, characterized in that: The method for obtaining the user characteristics of each piece of the user data in the user domain is: Converting the initial feature set formed based on the user data into a data matrix through linear transformation; Based on the data matrix, obtain the covariance matrix: Where X is the data matrix, where each row is a sample and each column is a feature; the standardized data matrix is , the covariance matrix is ​​C, n is the number of samples, is the transpose of the normalized data matrix; Solve for the eigenvalues ​​of the covariance matrix C and the corresponding eigenvector ; Sort all the eigenvalues ​​and eigenvectors by eigenvalue from large to small, select the eigenvectors corresponding to the first k eigenvalues, and form a new eigenspace with them to obtain the principal component space ; The standardized data matrix X′ is compared with the principal component space Multiply to get the data features after dimensionality reduction : The data features after dimension reduction are set as user features.

5. The ID Mapping method based on similarity calculation according to claim 4, characterized in that: The method for obtaining the number of parameters k is: Get the cumulative variance contribution rate M: Where d is the total number of features, is the i-th eigenvalue, t is the parameter; Based on the variance contribution rate M, a scree plot was constructed, and t that could explain 85%-95% of the variance was set as the value of k.

6. The ID Mapping method based on similarity calculation according to claim 1, characterized in that: Based on the user characteristics, similarity calculation is performed to obtain the similarity calculation result as follows: Get any two of the user features and obtain the cosine similarity calculation result C between the two user features: Among them, a and b are two user features whose cosine similarity results are to be calculated.

7. The ID Mapping method based on similarity calculation according to claim 6, characterized in that: When the cosine similarity result exceeds the similarity threshold, it is determined that the two user features involved in the similarity calculation belong to the same user; The value interval of the similarity threshold is [0.94-0.95].

8. An ID Mapping device based on similarity calculation, characterized in that: The IDMapping device based on similarity calculation includes: The user domain acquisition module is used to extract some user data from some business systems to the data center to obtain the user domain; A user feature acquisition module, used to acquire user features of each piece of user data in the user domain; A similarity calculation acquisition module is used to perform similarity calculation based on the user characteristics to obtain a similarity calculation result; The same source determination module is used to obtain the similarity calculation result and determine the user features whose similarity calculation results exceed a preset threshold as user features belonging to the same source.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the steps of the ID Mapping method based on similarity calculation according to any one of claims 1 to 7.

10. An ID Mapping system based on similarity calculation, characterized in that: When the system is running, the steps of the ID Mapping method based on similarity calculation as claimed in any one of claims 1 to 7 are executed.

Images