Method, apparatus and medium for recognizing person

Abstract

The application discloses a same-person identification method and device, equipment and medium, and relates to the technical field of Fintech (Fintech), which comprises the following steps: when a same-person identification instruction is received, corresponding target ID pair data is acquired; whether a unique user identification number of a target user corresponding to the same-person identification instruction can be acquired from a preset identification tool is determined according to the target ID pair data; in the preset identification tool, a plurality of unique user identification numbers are generated according to a plurality of ID relationship chains, and the ID relationship chains are generated according to the association relationship between the ID pairs extracted from the ID pair data corresponding to the same user; if the unique user identification number of the target user corresponding to the same-person identification instruction can be acquired from the preset identification tool, it is determined that the user corresponding to the unique user identification number and the target user are the same person. The application solves the technical problems of complex identification and weak identification scalability in the existing same-person identification process.

Description

Technical Field

[0001] This invention relates to the field of financial technology (Fintech), and in particular to a method, apparatus, device, and medium for identifying individuals. Background Technology

[0002] With the development of computer technology, more and more technologies are being applied in the financial field. The traditional financial industry is gradually transforming into financial technology (Fintech), and identity recognition technology is no exception. However, due to the security and real-time requirements of the financial industry, higher demands are being placed on the technology.

[0003] Currently, banks and other financial institutions often need to identify users who read content through different accounts to make targeted recommendations. Existing technologies use machine learning to judge based on user behavior characteristics or graph computing to connect user ID data to determine whether users reading through different accounts are the same person. Graph computing is complex, has poor scalability, and is significantly affected by interference data. Machine learning requires collecting user behavior data and then training it, which is cumbersome, complex, and has a high barrier to entry. In other words, existing methods for identifying users with the same account suffer from high identification costs, complex processes, and poor scalability, making it impossible to achieve efficient and low-cost information recommendation in the future. Summary of the Invention

[0004] The main objective of this invention is to provide a method, apparatus, device, and medium for identifying fan fiction, aiming to solve the technical problems of high identification cost, complex identification process, and weak scalability in the existing fan fiction identification process.

[0005] To achieve the above objectives, embodiments of the present invention provide a method for identifying fanfiction, the method comprising:

[0006] Upon receiving a fanfiction identification instruction, acquire the target ID pair data corresponding to the fanfiction identification instruction;

[0007] Based on the target ID, determine whether the unique user identification number of the target user corresponding to the fan identification instruction can be obtained from the preset identification tool;

[0008] In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains, and the ID relationship chains are generated by processing the association between ID pairs extracted from the data of ID pairs corresponding to the same user.

[0009] If the unique user identification number of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool, then it is determined that the user corresponding to the unique user identification number is the same person as the target user.

[0010] Optionally, the step of generating the ID relationship chain based on the association relationship between ID pairs extracted from the data corresponding to the same user's ID pairs includes:

[0011] Data on the ID pairs corresponding to the same user is obtained at preset time intervals. The ID pairs are then cleaned according to preset regular expression validation rules to obtain cleaned ID pairs.

[0012] Retrieve the preset priority order of each ID;

[0013] Based on the priority order, multiple ID pairs in the cleaned ID pair data are processed to supplement the relationship by using a common ID as an indirect association, so as to generate a multi-dimensional ID relationship;

[0014] Based on the multi-ID relationship, the relationship supplementation processing is continued for each ID pair in the cleaned ID pair data until each ID pair and ID are used in the processing, so as to generate ID relationship chains of varying lengths composed of different IDs in the cleaned ID pair data.

[0015] Optionally, the step of obtaining the preset priority order of each ID is followed by:

[0016] According to the priority order, the first map chain of the ID with the highest priority is obtained to obtain the first type of ID relationship chain. The map chain of any ID is a multi-dimensional ID relationship formed with the ID as the highest priority and starting from the ID.

[0017] Based on the priority order, continuously obtain second map chains of IDs with other priority order, and find IDs with higher priority than the other priority order IDs for the second map chains to obtain a second type of ID relationship chain;

[0018] Based on the first type of ID relationship chain and the second type of ID relationship chain, all ID relationship pairs of the cleaned ID pair data are processed into ID relationship chains.

[0019] Optionally, the step of processing all ID relationship pairs of the cleaned ID pair data into ID relationship chains based on the first type of ID relationship chain and the second type of ID relationship chain includes:

[0020] Obtain the highest priority ID in each ID relationship chain, and set the highest priority ID as the seed. Encrypt the seed using a preset encryption function to generate a unique identifier UID in the ID relationship chain.

[0021] Based on the level of the unique identifier UID of the corresponding ID relationship chain when the lowest priority ID is in different ID relationship chains, as well as the occurrence time and number of occurrences of the lowest priority ID in different ID relationship chains, the final UID corresponding to the lowest priority ID in different ID relationship chains is determined;

[0022] Obtain the final UID of each ID other than the lowest priority ID, and associate all IDs to form the final ID relationship chain.

[0023] Optionally, the unique user identification number includes the latest unique user identification number, and the preset identification tool includes multiple historical unique user identification numbers. The step of generating multiple unique user identification numbers based on multiple ID relationship chains includes:

[0024] Based on the priority of each ID in any ID relationship chain, determine in order whether each ID in the ID relationship chain can find the associated target historical unique user identification number from the multiple historical unique user identification numbers;

[0025] If it is determined that none of the IDs in the ID relationship chain can find a target historical unique user identification number associated with the multiple historical unique user identification numbers, then the latest unique user identification number is generated according to the preset generation rules to obtain multiple latest unique user identification numbers corresponding to the multiple ID relationship chains.

[0026] Optionally, the step of determining whether each ID in any ID relationship chain can find an associated target historical unique user identifier from the plurality of historical unique user identifiers, based on the priority of each ID in the ID relationship chain, includes the following:

[0027] If it is determined that each ID in the ID relationship chain has a target historical unique user identification number that has been associated with the multiple historical unique user identification numbers, the target ID associated with the target historical unique user identification number is obtained.

[0028] Obtain the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain. Based on the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain, determine the latest unique user identifier of the corresponding ID relationship chain.

[0029] Optionally, the step of determining that each ID in the ID relationship chain has a target historical unique user identifier associated with the plurality of historical unique user identifiers, and obtaining the target ID associated with the target historical unique user identifier, includes the following after:

[0030] The ID relationship chain is pre-coded based on each ID appearing in the ID relationship chain to obtain a first code number;

[0031] The ID relationship chain corresponding to the historical unique user identification number is pre-coded to obtain the second code number;

[0032] The first code number and the second code number are summed to obtain the code chain;

[0033] Obtain the encoded bits of other IDs with higher priority than the target ID;

[0034] If the encoded bits of other IDs with higher priority than the target ID are greater than or equal to the corresponding encoded values ​​on the encoding chain, then multiple unique user identification numbers are determined.

[0035] The present invention also provides a fanfiction identification device, the fanfiction identification device comprising:

[0036] The first acquisition module is used to acquire the target ID pair data corresponding to the fan identification instruction when a fan identification instruction is received;

[0037] The first determining module is used to determine whether the unique user identification number of the target user corresponding to the fan identification instruction can be obtained from the preset identification tool based on the target ID;

[0038] In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains, and the ID relationship chains are generated by processing the association between ID pairs extracted from the data of ID pairs corresponding to the same user.

[0039] The second determining module is used to determine that the user corresponding to the unique user identification number is the same person as the target user if the unique user identification number of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool.

[0040] Optionally, the first determining module includes:

[0041] The first acquisition unit is used to acquire data of ID pairs corresponding to the same user at preset time intervals, and clean the ID pair data according to preset regular expression validation rules to obtain cleaned ID pair data.

[0042] The second acquisition unit is used to acquire the preset priority order of each ID;

[0043] The first relationship supplementation unit is used to supplement the relationship of multiple ID pairs in the cleaned ID pair data with a common ID as an indirect association according to the priority order, so as to generate a multi-dimensional ID relationship;

[0044] The second relationship supplementation unit is used to continue to supplement the relationship of each ID pair in the cleaned ID pair data according to the multi-dimensional ID relationship, until each ID pair and ID are used in the processing, so as to generate ID relationship chains of different lengths composed of different IDs in the cleaned ID pair data.

[0045] Optionally, the same person identification device further includes:

[0046] The second acquisition module is used to acquire the first map chain of the highest priority ID according to the priority order, so as to obtain the first type of ID relationship chain, wherein the map chain of any ID is a multi-dimensional ID relationship formed with the ID as the highest priority and starting from the ID;

[0047] The third acquisition module is used to continuously acquire second map chains of IDs with other priority order according to the priority order, and to find IDs with higher priority than the other priority order IDs for the second map chain to obtain a second type of ID relationship chain;

[0048] The processing module is used to process all ID relationship pairs of the cleaned ID pair data into ID relationship chains based on the first type of ID relationship chain and the second type of ID relationship chain.

[0049] Optionally, the same person identification device further includes:

[0050] The fourth acquisition module is used to acquire the highest priority ID in each ID relationship chain. The highest priority ID is set as the seed, and the seed is encrypted using a preset encryption function to generate a unique identifier UID in the ID relationship chain.

[0051] The third determining module is used to determine the final UID corresponding to the lowest priority ID in different ID relationship chains based on the level of the unique identifier UID of the corresponding ID relationship chain when the lowest priority ID is in different ID relationship chains, as well as the occurrence time and number of occurrences of the lowest priority ID in different ID relationship chains.

[0052] The fifth acquisition module is used to acquire the final UID of each ID other than the lowest priority ID, and associate all IDs to form the final ID relationship chain.

[0053] Optionally, the unique user identification number includes the latest unique user identification number, and the preset identification tool includes multiple historical unique user identification numbers. The first determining module includes:

[0054] The first determining unit is used to determine, in an orderly manner, whether each ID in any ID relationship chain can find the associated target historical unique user identification number from the plurality of historical unique user identification numbers, based on the priority of each ID in any ID relationship chain.

[0055] The second determining unit is used to generate the latest unique user identification number according to a preset generation rule if it is determined that none of the IDs in the ID relationship chain can find the associated target historical unique user identification number from the multiple historical unique user identification numbers, so as to obtain multiple latest unique user identification numbers corresponding to the multiple ID relationship chains.

[0056] Optionally, the same person identification device further includes:

[0057] The sixth acquisition module is used to acquire the target ID associated with the target historical unique user identification number if it is determined that each ID in the ID relationship chain has a target historical unique user identification number that has been associated with the multiple historical unique user identification numbers.

[0058] The seventh acquisition module is used to acquire the priority of the target ID, the priority of each ID associated with the target historical unique user identification number associated with the target ID, and the priority of each ID on the ID relationship chain. Based on the priority of the target ID, the priority of each ID associated with the target historical unique user identification number associated with the target ID, and the priority of each ID on the ID relationship chain, the latest unique user identification number of the corresponding ID relationship chain is determined.

[0059] Optionally, the same person identification device further includes:

[0060] The first encoding module is used to pre-encode the ID relationship chain according to each ID appearing in the ID relationship chain to obtain a first encoding number;

[0061] The second encoding module is used to pre-encode the ID relationship chain corresponding to the historical unique user identification number to obtain the second encoding number;

[0062] The summation module is used to sum the first code number and the second code number to obtain the code chain;

[0063] The eighth acquisition module is used to acquire the encoding bits of other IDs with higher priority than the target ID;

[0064] The fourth determining module is used to determine multiple unique user identification numbers if the encoded bits of other IDs with higher priority than the target ID are greater than or equal to the corresponding encoded values ​​on the encoding chain.

[0065] The present invention also provides a medium storing a fan fiction identification program, which, when executed by a processor, implements the steps of the fan fiction identification method described above.

[0066] Upon receiving a fanfiction identification instruction, this invention first acquires the target ID pair data corresponding to the instruction. Based on the target ID pair data, it determines whether the unique user identification number of the target user corresponding to the instruction can be obtained from a preset identification tool. If the unique user identification number can be obtained, then the user corresponding to the unique user identification number is determined to be the same person as the target user. In other words, in this application, fanfiction identification can be performed simply by determining whether the target ID pair data in the fanfiction identification instruction matches the unique user identification number of the target user. The fanfiction identification process is simple, and the target ID pair data is not limited, meaning it is scalable. Therefore, the fanfiction identification process in this application is highly scalable, and can be performed simply by determining whether the target ID pair data in the fanfiction identification instruction matches the unique user identification number of the target user. The reason for the co-person identification is that multiple unique user identification numbers are generated based on multiple ID relationship chains. That is, each user pair is assigned a unique user identification number. Furthermore, in the preset identification tool, the ID relationship chain is generated by processing the association relationships between ID pairs extracted from the data of ID pairs corresponding to the same user. Specifically, the preset identification tool of this application has already obtained the association relationships between user ID pairs and processed them to generate the ID relationship chain, or it has already extracted, transformed, and loaded previously obtained ID pair data to generate the ID relationship chain. This ensures that different IDs of the same user use the same unique user identification number; that is, the IDs are already associated through the unique user identification number rather than being separated. Therefore, if the unique user identification number of the target user corresponding to the co-person identification instruction can be obtained from the preset identification tool, it is determined that the user corresponding to the unique user identification number is the same person as the target user. Compared to graph computing and machine learning methods, this invention reduces the identification cost, simplifies the co-person identification process, and has strong scalability, facilitating subsequent efficient and low-cost information recommendation. Attached Figure Description

[0067] Figure 1 This is a flowchart illustrating the first embodiment of the method for identifying likenesses according to the present invention;

[0068] Figure 2 This is a detailed flowchart illustrating the steps of obtaining the node to be processed based on the content of all files in the second embodiment of the method for identifying fanfiction of the present invention.

[0069] Figure 3 This is a schematic diagram of the device structure of the hardware operating environment involved in the method of the embodiment of the present invention.

[0070] Figure 4 This is a schematic diagram of the first scenario in the method for identifying likenesses in this invention;

[0071] Figure 5 This is a schematic diagram of the second scenario in the method for identifying similar individuals according to the present invention;

[0072] Figure 6 This is a schematic diagram of the third scenario in the method for identifying likenesses in this invention;

[0073] Figure 7 This is a schematic diagram of the fourth scenario in the method for identifying fanfiction in this invention.

[0074] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0075] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0076] This invention provides a method for identifying fanfiction. In one embodiment of the method, reference is made to... Figure 1 The method for identifying fellow individuals includes:

[0077] Step S10: Upon receiving a fan identification instruction, obtain the target ID pair data corresponding to the fan identification instruction;

[0078] Step S20: Determine whether the unique user identification number of the target user corresponding to the fan identification instruction can be obtained from the preset identification tool based on the target ID.

[0079] In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains, and the ID relationship chains are generated by processing the association between ID pairs extracted from the data of ID pairs corresponding to the same user.

[0080] Step S30: If the unique user identification number of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool, then it is determined that the user corresponding to the unique user identification number is the same person as the target user.

[0081] The specific steps are as follows:

[0082] Step S10: Upon receiving a fan identification instruction, obtain the target ID pair data corresponding to the fan identification instruction;

[0083] It should be noted that in this embodiment, the fan identification method is applied to a fan identification system, which is subordinate to a fan identification device. In this system, each time a user login or entry is detected, a fan identification command is triggered. When the fan identification system triggers and receives the fan identification command, it obtains the target ID pair data (in general form) or target ID data corresponding to the command. Specifically, after obtaining the target ID data, target ID pair data is generated according to the mapping relationship. Specifically, the mapping relationship between different IDs is processed using Hive (a distributed big data warehouse) - ETL (extract-transform-load). Hive-ETL specifically transforms and loads different ID data (e.g., converting and transforming data...). Different ID data are converted and loaded through phone numbers to obtain the mapping relationship between different IDs. In this embodiment, an ID pair refers to a mapping relationship containing two IDs, while an ID chain refers to a mapping relationship containing two or more IDs. ID pairs can be processed into ID chains. It should be noted that the method of processing ID pairs into ID chains can be implemented by manual programming or graph computation, but the implementation method is complex. If an error occurs, the data must be cleaned and recalculated from scratch, which is time-consuming. In this embodiment, the intermediate process is broken down step by step using Hive-ETL. If an error occurs in the intermediate process, the task can continue from the point of error, thus improving scalability. In addition, in this embodiment, when adding a user ID, it can also be implemented during the process of processing ID pairs into ID chains.

[0084] Obtaining the target ID pair data corresponding to the fan identification instruction means, upon receiving the fan identification instruction, retrieving the target ID pair data corresponding to the fan identification instruction from the fan identification instruction itself, or retrieving the target ID pair data corresponding to the fan identification instruction from the big data system or big data warehouse pointed to by the fan identification instruction. This target ID pair data can be ID pair data composed of preset regular IDs. Preset regular IDs (which can be simply referred to as regular IDs) include customer number (ecifno), ID card number (idno), phone number (telephone), WeChat ID (unionid), and device number (imei / idfa). Among them, WeChat ID (unionid) refers to the WeChat identification ID; the unionid of an application user under the same WeChat developer platform account is unique. Device number (imei / idfa) refers to the unique identifier ID of a mobile device. The device number for Android devices is imei, and the device number for iOS devices is idfa. The preset regular IDs can be processed and supplemented with ID association relationships, such as ID pairs. <idno:telephone>and ID pair<telephone:unionid> This can be supplemented by associating with the intermediate IDtelephone.<idno:unionid> Relationship between ID pairs<ecifno:unionid> and ID pair <unionid:imei idfa>This can be supplemented by associating through unionid.<ecifno:imei / idfa> In addition, the preset regular IDs have priorities, and the priority order can be: ecifno>idno>telephone>unionid>device number (imei / idfa). Of course, the priority order can also be other, but in this embodiment, the priority order of ecifno>idno>telephone>unionid>device number (imei / idfa) is used as an example for specific explanation. It should be noted that in this embodiment, the target ID pair data can also be composed of non-preset regular IDs, that is, the target ID pair data can be composed of extended class IDs, which can include QQ number, E-mail, etc. Since the target ID pair data can also be composed of non-preset regular IDs, in this embodiment, when a non-preset regular ID is detected, the mapping relationship between the non-preset regular ID and the preset regular ID can be processed. Specifically, the mapping relationship between the non-preset regular ID and the preset regular ID is processed by using Hive (a distributed big data warehouse)-ETL (extract-transform-load), that is, the non-preset regular ID is loaded after data extraction and transformation to achieve the scalability of user ID.

[0085] Step S20: Determine whether the unique user identification number of the target user corresponding to the fan identification instruction can be obtained from the preset identification tool based on the target ID.

[0086] In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains, and the ID relationship chains are generated by processing the association between ID pairs extracted from the data of ID pairs corresponding to the same user.

[0087] Determining whether the unique user identification number of the target user corresponding to the fanfiction identification instruction can be obtained from the preset identification tool based on the target ID includes two cases: it is determined that the unique user identification number of the target user corresponding to the fanfiction identification instruction can be obtained from the preset identification tool, or it is determined that the unique user identification number of the target user corresponding to the fanfiction identification instruction cannot be obtained from the preset identification tool. If it cannot be obtained, then the unique user identification number of the target user corresponding to the fanfiction identification instruction is generated. The specific method of generation will be explained later.

[0088] In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains. The ID relationship chains are generated by processing the association relationships between ID pairs extracted from the data of ID pairs corresponding to the same user. In this embodiment, same-person identification can be performed simply by determining whether the target ID pair data in the same-person identification instruction matches the unique user identification number of the target user. The same-person identification process is simple, and the target ID pair data is randomly expandable without requiring changes to the system's results. Therefore, the same-person identification process in this application has strong scalability, and same-person identification can be performed simply by determining whether the target ID pair data in the same-person identification instruction matches the unique user identification number of the target user. The identification is based on the following: multiple unique user identification numbers are generated according to multiple ID relationship chains, that is, a unique user identification number is set for each user pair. In the preset identification tool, the ID relationship chain is generated by processing the association relationship between ID pairs extracted from the data of the same user's corresponding ID pair. That is, in the preset identification tool of this embodiment, the association relationship between user ID pairs has already been obtained and processed to generate the ID relationship chain, or the previously obtained ID pair data has been extracted, transformed and loaded to generate the ID relationship chain, so as to ensure that different IDs of the same user use the same unique user identification number. That is, the IDs have been associated with each other through the unique user identification number instead of being separated.

[0089] Specifically, refer to Figure 2 The step of generating the ID relationship chain based on the association relationship between ID pairs extracted from the data corresponding to the same user's ID pairs includes:

[0090] Step S21: Obtain the ID pair data corresponding to the same user at preset time intervals, and clean the ID pair data according to preset regular expression validation rules to obtain cleaned ID pair data.

[0091] In this embodiment, data of ID pairs corresponding to the same user are obtained at preset time intervals. That is, at preset time intervals, ID relationship pairs (containing at least two IDs) are obtained from different data tables obtained from a large database. ID pairs that do not conform to the encoding rules or preset regular expression verification rules are cleaned and filtered. The encoding rules or verification rules may include character length rules and character composition rules. That is, it is to determine whether idno conforms to the preset first character length rule and first character composition rule, whether telphone conforms to the preset second character length rule and second character composition rule, whether unionid conforms to the preset third character length rule and third character composition rule, whether idfa conforms to the preset fourth character length rule and fourth character composition rule, and whether imei conforms to the preset fifth character length rule and fifth character composition rule, etc. In other words, the ID pair data is cleaned according to the preset regular expression verification rules to obtain cleaned ID pair data.

[0092] Specifically, the regular expression validation rules for idno are as follows:

[0093] (^[1-9]\d{5}(18|19|(

[23] \d))\d{2}((0[1-9])|(10|11|12))(([0-2][1-9])|10|20|30|31)\d{ 3}[0-9Xx]$)|(^[1-9]\d{5}\d{2}((0[1-9])|(10|11|12))(([0-2][1-9])|10|20|30|31)\d{2}$).

[0094] The regular expression validation rule for telphone is: (^1(3|4|5|6|7|8|9)\d{9}$).

[0095] The regular expression validation rule for unionid is: ([\\\x00-\\\xff]+), with a length of 28-29 characters.

[0096] The regular expression validation rule for idfa is: length 32-36 characters of numbers or letters.

[0097] Imei's regular expression validation rule is: length 15-18 characters of pure numbers.

[0098] Specifically, for example, when cleaning a telephone, the process begins by determining whether the telephone is entirely composed of numbers according to the second character composition rule. If it is determined that the telephone is not entirely composed of numbers, then the telephone is filtered. If it is determined that the telephone is entirely composed of numbers, the process continues by determining whether each digit of the telephone conforms to the second character composition rule. For example, it determines whether the first digit of the telephone is 1. If it is uncertain whether the first digit of the telephone is 1, then the telephone is filtered. If it is determined that the first digit of the telephone is 1, then the other digits of the telephone are identified until it is determined that the telephone conforms to the second character composition rule. If it does not conform, then it is filtered. After determining that the telephone conforms to the second character composition rule, the process then determines whether the length of the telephone's characters meets the requirements according to the second character length rule. If it meets the requirements, then the telephone is retained.

[0099] Step S22: Obtain the preset priority order of each ID;

[0100] After obtaining the cleaned ID pairs, the priority order of each ID is obtained. Specifically, the priority order is ecifno>idno>telephone>unionid>device number (imei / idfa).

[0101] Step S23: According to the priority order, perform relationship supplementation processing on multiple ID pairs in the cleaned ID pair data with common ID as indirect association to generate multi-dimensional ID relationships;

[0102] Step S24: Based on the multi-dimensional ID relationship, continue to process the relationship of each ID pair in the cleaned ID pair data until each ID pair and ID are used in the processing, so as to generate ID relationship chains of varying lengths composed of different IDs in the cleaned ID pair data.

[0103] To fully utilize the mapping relationship between IDs, ID pairs are processed into ID chains. Specifically, according to the priority order, multiple ID pairs in the cleaned ID pair data are indirectly associated using a common ID to supplement the relationship, generating a multi-dimensional ID relationship. Specifically, ID pairs are combined pairwise and indirectly associated using each ID. For example, ID pairs...<ecifno,idno> and<idno,telephone> Using IDno as an indirect association can generate an ID chain.<ecifno,idno,telephone> Based on the multi-ID relationship, the relationship supplementation processing is continued for each ID pair in the cleaned ID pair data until each ID pair and ID are used in the processing, so as to generate ID relationship chains of different lengths composed of different IDs in the cleaned ID pair data. That is, each two ID pairs need to be indirectly associated once, and so on, until each ID pair and ID are used in the processing of the relationship chain, so as to generate ID relationship chains of different lengths composed of different IDs in the cleaned ID pair data.

[0104] For example, the obtained ID pairs can be categorized into the following two types:

[0105] LINK1:<ecifno,idno,telephone,unionid,imei,idfa>

[0106] LINK2:<NULL,idno,telephone,unionid,imei,idfa>

[0107] The device first breaks down LINK1 and LINK2 into multiple ID pairs:<ecifno:idno> ,<ecifno:telephoe> , <idno:telephone>Then, the ID pairs are combined two by two to generate <ecifno:idno:telephone>, <ecifno:telephone:unionid>, and the like, and the ternary relation pairs are combined two by two to generate quaternary relation pairs, and so on until all the IDs are finally combined into an ID chain.

[0108] It should be noted that in the above process of obtaining the ID relation chain, the relation pairs can be repeatedly used, for example, there are 15 kinds of ID pairs composed of 6 kinds of IDs, and according to the 15 kinds of ID pairs, corresponding ternary relations and the like need to be further generated, so there are problems of tedious and complex process and low efficiency.

[0109] The step of obtaining the priority order of each ID includes:

[0110] In step S25, according to the priority order, a first map chain of the ID with the highest priority is obtained to obtain a first type of ID relation chain, wherein the map chain of any ID is a multi-ID relation formed by taking the ID as the highest priority and taking the ID as the starting point.

[0111] In step S26, according to the priority order, a second map chain of other priority order IDs is continuously obtained, and an ID with a higher priority than the other priority order IDs is found for the second map chain to obtain a second type of ID relation chain.

[0112] In step S27, according to the first type of ID relation chain and the second type of ID relation chain, all the ID relation pairs of the cleaned ID pair data are processed into an ID relation chain.

[0113] In this embodiment, another processing and association method of the ID chain is also provided to solve the problems of tedious and complex process and low efficiency in the process of generating the ID relation chain. Specifically, in this embodiment, in the processing and association method of the ID chain, the existing ID relation chain is not split into relation pairs for association, but is directly associated based on the ID relation chain or the map chain. For example, for the ID pair data LINK1 and LINK2, the common ID relation chain is directly used for association, that is, the common IDs idno, unionid, telephone, and imei / idfa are associated, and the common IDs idno, unionid, telephone, and imei / idfa are associated (the specific association will be described in detail later). In total, 4 times of association are performed, which completely preserves the original ID relation and improves the generation efficiency of the ID chain.

[0114] In this embodiment, the processing step of the ID chain is specifically as follows: Figure 4 It is to be noted that the IDidno_map, telephone_map and the like in the figure are no longer simple ID pairs, but Map chains, wherein the Map chain structure is <ID1, ID2, ID3... IDn>, wherein n <= 5 (ID types), that is, the map chain of any ID is a multi-element ID relationship formed with the ID as the highest priority and as the starting point. For example, the map chain of ecifno can include <ecifno, telephone, imei / idfa>, or <ecifno, telephone> or <ecifno, imei / idfa>, and the process of processing the ID chain is actually to find an ID with a higher priority for the ID, and to associate the ID with a lower priority to an ID with a higher priority as much as possible.

[0115] In the embodiment, according to the priority order, a first map chain of the ID with the highest priority is obtained to obtain a first type of ID relationship chain, and according to the priority order, a second map chain of the ID with the other priority order is continuously obtained, and an ID with a higher priority than the ID with the other priority order is found for the second map chain to obtain a second type of ID relationship chain; and all ID relationship pairs of the cleaned ID pair data are processed into an ID relationship chain according to the first type of ID relationship chain and the second type of ID relationship chain.

[0116] Specifically, a first map chain of the ID with the highest priority is obtained first, such as a first map chain of ecifno, and after the first map chain of ecifno is obtained, a second map chain of the ID with the other priority order, such as idno, is obtained, and an ID with a higher priority than the ID with the other priority order is found for the second map chain to obtain a second type of ID relationship chain, including: finding an ecifno with a higher priority for the Map chain containing idno to obtain a second type of ID relationship chain, that is, generating an ID chain containing at least ecifno or idno; and finding an ecifno, idno with a higher priority for the Map chain containing telephone to obtain a second type of ID relationship chain, that is, generating an ID chain containing at least ecifno, idno or telephone; and finding an ecifno, idno, unionid with a higher priority for the Map chain containing unionid to obtain a second type of ID relationship chain, that is, generating an ID chain containing at least ecifno, idno, telephone or unionid; and finding an ecifno, idno, telephone or unionid with a higher priority for the Map chain containing the device number (imei / idfa) to obtain a second type of ID relationship chain, that is, generating an ID relationship chain including all user IDs.

[0117] According to the first type of ID relationship chain and the second type of ID relationship chain, the processing logic of processing all ID relationship pairs of the cleaning ID pair data into an ID relationship chain can be: processing is performed using all IDs, for example, in the Map chain of ecifno, there can be four types of IDs, idno, telephone, unionid, and equipment number, and when association is performed, idno, telephone, unionid, and equipment number are sequentially used to associate with other Map chains to process into an ID relationship chain. In this embodiment, it is obvious that four times of association are performed through idno, telephone, unionid, and equipment number.

[0118] In all the above association processes, if a one-to-many ID relationship pair (for example, ecifno: telephone is 1:N) is encountered, a certain ID relationship pair with the highest priority (the priority is determined by the time or the number of occurrences of the one-to-many ID relationship pair) can be used for association, and all ID relationship pairs are processed into an ID relationship chain.

[0119] In this embodiment, after the ID relationship chain is processed, the preset identification tool generates a plurality of unique user identification numbers corresponding to a plurality of ID relationship chains.

[0120] In step S30, if the unique user identification number of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool, it is determined that the user corresponding to the unique user identification number and the target user are the same person.

[0121] After obtaining the same person identification instruction, if the unique user identification number CMID (Cdvd-Master-Id) of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool, it is determined that the user corresponding to the unique user identification number and the target user are the same person. If the unique user identification number of the target user corresponding to the same person identification instruction cannot be obtained from the preset identification tool, it is determined that the user corresponding to the unique user identification number and the target user are not the same person.

[0122] Upon receiving a fanfiction identification instruction, this invention first acquires the target ID pair data corresponding to the instruction. Based on the target ID pair data, it determines whether the unique user identification number of the target user corresponding to the instruction can be obtained from a preset identification tool. If the unique user identification number can be obtained, then the user corresponding to the unique user identification number is determined to be the same person as the target user. In other words, in this application, fanfiction identification can be performed simply by determining whether the target ID pair data in the fanfiction identification instruction matches the unique user identification number of the target user. The fanfiction identification process is simple, and the target ID pair data is not limited, meaning it is scalable. Therefore, the fanfiction identification process in this application is highly scalable, and can be performed simply by determining whether the target ID pair data in the fanfiction identification instruction matches the unique user identification number of the target user. The reason for the co-person identification is that multiple unique user identification numbers are generated based on multiple ID relationship chains. That is, each user pair is assigned a unique user identification number. Furthermore, in the preset identification tool, the ID relationship chain is generated by processing the association relationships between ID pairs extracted from the data of ID pairs corresponding to the same user. Specifically, the preset identification tool of this application has already obtained the association relationships between user ID pairs and processed them to generate the ID relationship chain, or it has already extracted, transformed, and loaded previously obtained ID pair data to generate the ID relationship chain. This ensures that different IDs of the same user use the same unique user identification number; that is, the IDs are already associated through the unique user identification number rather than being separated. Therefore, if the unique user identification number of the target user corresponding to the co-person identification instruction can be obtained from the preset identification tool, it is determined that the user corresponding to the unique user identification number is the same person as the target user. Compared to graph computing and machine learning methods, this invention reduces the identification cost, simplifies the co-person identification process, and has strong scalability, facilitating subsequent efficient and low-cost information recommendation.

[0123] Furthermore, based on the above embodiments, in another embodiment of the method for identifying individuals, the step of processing all ID relationship pairs of the cleaned ID pair data into an ID relationship chain according to the first type of ID relationship chain and the second type of ID relationship chain includes:

[0124] Step A1: Obtain the highest priority ID in each ID relationship chain, and set the highest priority ID as the seed. Encrypt the seed using a preset encryption function to generate a unique identifier UID in the ID relationship chain.

[0125] In this embodiment, the ID relationship chain (which can be simply referred to as the ID chain) generated in the above manner may not be the final ID chain. This is because the source data layer may include one-to-many ID relationship pairs, such as ecifno:telephone being 1:N. When processing the ID relationship chain, these one-to-many ID relationship pairs may be retained, resulting in a certain ID belonging to multiple relationship chains. Therefore, the generated ID relationship chain can be processed to generate the final ID chain. The specific process is as follows: Figure 5 As shown, the highest priority ID in each ID relationship chain is obtained. Let the highest priority ID be the seed. The seed is encrypted using a preset encryption function (optionally MD5 algorithm) to generate a unique identifier UID in the ID relationship chain. That is, in this embodiment, a unique ID is generated for each relationship chain in the ID relationship chain. If the highest priority ID in the ID chain is idno, then MD5 is performed on idno to generate a UID. If the highest priority ID is telephone, then MD5 is performed on telephone to generate a UID. That is, in this embodiment, the highest priority ID in the ID chain is selected as the seed, and the seed is encrypted using an encryption function to ensure uniqueness.

[0126] Step A2: Based on the level of the unique identifier UID of the corresponding ID relationship chain when the lowest priority ID is in different ID relationship chains, as well as the occurrence time and number of occurrences of the lowest priority ID in different ID relationship chains, determine the final UID corresponding to the lowest priority ID in different ID relationship chains.

[0127] Step A3: Obtain the final UID of each ID other than the lowest priority ID, and associate all IDs to form the final ID relationship chain.

[0128] Based on the level of the unique identifier UID of the corresponding ID relationship chain when the lowest priority ID is in different ID relationship chains, and the occurrence time and frequency of the lowest priority ID in different ID relationship chains, the final UID corresponding to the lowest priority ID in different ID relationship chains is determined. The final UIDs of all other IDs besides the lowest priority ID are obtained, and all IDs are associated to form the final ID relationship chain. Specifically, the highest priority ecifno is associated with the UID of its relationship chain, and idno is associated with a UID. The association method is as follows: if idno has an associated ecifno, the UID associated with ecifno in the previous step is used as the UID of idno; if idno does not have an ecifno, the UID of the ID relationship chain is used. Using this method, telephone, unionid and device number are associated with UIDs in sequence. The association method is the same as that of idno, that is, in this embodiment, ecifno, idno, telephone, unionid and the UID associated with the relationship chain are used in sequence according to priority.

[0129] After associating all IDs with UIDs, the occurrence time and frequency of the lowest priority ID in different ID relationship chains are obtained (where the closer the time, the higher the priority, and time has the first weight in the corresponding UID consideration). The frequency of occurrence of the lowest priority ID in different ID relationship chains is also obtained (the more frequency, the higher the priority, and the frequency of occurrence has the second weight in the corresponding UID consideration). Based on the occurrence time, frequency, first weight, and second weight, the target ID relationship chain corresponding to the lowest priority ID is obtained. The highest UID corresponding to the target ID relationship chain corresponding to the lowest priority ID is obtained as the final UID of that ID. Finally, all IDs are associated using UIDs to form the final ID chain.

[0130] In this embodiment, the highest-priority ID in each ID relationship chain is obtained, and this highest-priority ID is designated as the seed. The seed is encrypted using a preset encryption function to generate a unique identifier (UID) for the corresponding ID relationship chain. Based on the level of the unique identifier (UID) of the lowest-priority ID in different ID relationship chains, and the occurrence time and frequency of the lowest-priority ID in different ID relationship chains, the final UID corresponding to the lowest-priority ID in different ID relationship chains is determined. The final UIDs of all other IDs besides the lowest-priority ID are obtained, and all IDs are associated to form the final ID relationship chain. In this embodiment, the final ID relationship chain is accurately generated.

[0131] Furthermore, based on the above embodiments, this invention provides another embodiment of the method for identifying duplicate users. In this embodiment, the unique user identification number includes the latest unique user identification number, and the preset identification tool includes multiple historical unique user identification numbers. The step of generating multiple unique user identification numbers corresponding to multiple ID relationship chains includes:

[0132] Step S31: Based on the priority of each ID in any ID relationship chain, determine in order whether each ID in the ID relationship chain can find the associated target historical unique user identification number from the multiple historical unique user identification numbers.

[0133] In this embodiment, after generating the ID relationship chain, the user's CMID can be generated according to a preset generation rule (the generation rule can be a function or a regular expression to ensure that the CMID generated for each ID relationship chain is unique) and the ID relationship chain (using the ID relationship chain UID as the seed). Specifically, for example, using the ID relationship chain UID as the seed, the ID relationship chain is encrypted using an MD5 encryption function to obtain the MD5 value of the ID relationship chain as the corresponding CMID. To ensure the stability, uniqueness, and lifetime of the user's unique identifier CMID, specifically, after generating the ID relationship chain, it is necessary to combine it with historical CMID data. Further confirmation is needed regarding whether to adopt the CMID generated based on the ID relationship chain, or to directly use the corresponding historical CMID (historical unique user identification number) as the unique user identification number corresponding to the same person identification instruction. Specifically, in this embodiment, based on the priority of each ID on any ID relationship chain, it is also determined in an orderly manner whether each ID on the ID relationship chain can find the target historical unique user identification number associated with the multiple historical unique user identification numbers. Specifically, according to the order of ID priority on the ID chain, ecifno, idno, telephone, unionid, and device number are used in sequence to search for the historical unique user identification number associated with each ID in the CMID historical data, i.e., the historical CMID.

[0134] Step S32: If it is determined that none of the IDs in the ID relationship chain can find the associated target historical unique user identification number from the multiple historical unique user identification numbers, then the latest unique user identification number is generated according to the preset generation rule to obtain multiple latest unique user identification numbers corresponding to the multiple ID relationship chains.

[0135] If it is determined that none of the IDs in the ID relationship chain can find the associated target historical unique user identification number from the multiple historical unique user identification numbers, then the latest unique user identification number is generated according to the preset generation rules to obtain multiple latest unique user identification numbers corresponding to the multiple ID relationship chains. The latest unique user identification number is stored in the preset identification tool and used as the unique user identification number corresponding to the fan identification instruction.

[0136] If it is determined that each ID in the ID relationship chain can find the associated target historical unique user ID from the multiple historical unique user IDs, it may cause the same person identification instruction to be associated with the wrong CMID, that is, it may cause a high-priority ID to be associated with the wrong CMID.

[0137] Specifically, such as Figure 6 , Figure 7 As shown, the new CMID generated based on the ID relationship chain UID should have been used as the unique user identification number corresponding to the fan identification command. However, because the telephone is associated with the historical CMID1, the historical CMID1 may have been used as the unique user identification number corresponding to the fan identification command. This is mainly due to the change of ownership of the low-priority ID, which causes the user corresponding to the fan identification command to be identified as an existing user when it should be a newly added user. It should be noted that not all associations of high-priority IDs with existing historical CMIDs of low-priority IDs are incorrect. For example, if Figure 7 In the CMID historical data, telephone:001 is just a standalone number (if there is no ecifno or idno under the historical CMID associated with this telephone, it means that the system has obtained the telephone but there is no other data). It is correct to directly use the historical CMID in the CMID historical data.

[0138] To address the issue of incorrect CMID association with the aforementioned fan identification command, this embodiment includes the following step after determining whether each ID in any ID relationship chain can find an associated target historical unique user identification number from the plurality of historical unique user identification numbers, based on the priority of each ID in that ID relationship chain:

[0139] Step B1: If it is determined that each ID in the ID relationship chain has a target historical unique user identification number that has been associated with the multiple historical unique user identification numbers, obtain the target ID associated with the target historical unique user identification number.

[0140] Step B2: Obtain the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain. Based on the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain, determine the latest unique user identifier of the corresponding ID relationship chain.

[0141] In this embodiment, specifically, the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain are obtained. Based on the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain, the latest unique user identifier of the corresponding ID relationship chain is determined. Specifically, the priority of the target ID is obtained, such as the priority being ecifno>idno>telephone>unionid>device number (imei / idfa). Based on this priority, if ecifno is used to associate with a historical CMID in historical CMID data, then the historical CMID associated with ecifno is used as the latest CMID in the ID chain.

[0142] If `idno` is used to associate with a historical CMID (historical unique user identifier) ​​in historical CMID data, then the following applies: Specifically, if there is no associated `ecifno` in the historical CMID data: use the historical CMID associated with `idno` as the CMID of the ID relationship chain. If there is an associated `ecifno` in the historical CMID data: if there is an `ecifno` in the ID relationship chain, use the CMID generated based on the ID relationship chain UID as the latest CMID of the ID relationship chain; if there is no `ecifno` in the ID relationship chain, use the historical CMID associated with `idno` as the latest CMID of the ID relationship chain.

[0143] If a telephone number is used to associate a historical CMID with historical CMID data, the following scenarios apply: If the historical CMID has no ecifno or idno: use the historical CMID associated with the telephone number as the latest CMID in the ID chain. If the historical CMID has no ecifno but has an idno: if there is no idno in the ID relationship chain, use the historical CMID associated with the telephone number as the latest CMID in the ID relationship chain; otherwise, use the CMID generated based on the ID chain UID as the latest CMID in the ID chain. If the historical CMID has no idno but has an ecifno: if there is no ecifno in the ID relationship chain, use the historical CMID associated with the telephone number as the latest CMID in the ID relationship chain; otherwise, use the CMID generated based on the ID relationship chain UID as the latest CMID in the ID chain. If the historical CMID has both ecifno and idno: use the CMID generated based on the ID relationship chain UID as the latest CMID in the ID chain.

[0144] In this embodiment, if it is determined that each ID in the ID relationship chain has a target historical unique user identifier associated with it from the plurality of historical unique user identifiers, the target ID associated with the target historical unique user identifier is obtained; the priority of the target ID, the priority of each ID associated with the target historical unique user identifier, and the priority of each ID in the ID relationship chain are obtained; based on the priority of the target ID, the priority of each ID associated with the target historical unique user identifier, and the priority of each ID in the ID relationship chain, the latest unique user identifier of the corresponding ID relationship chain is determined. In this embodiment, the latest CMID of the ID chain is accurately obtained.

[0145] Furthermore, based on the above embodiments, in another embodiment of the method for identifying duplicate users provided by the present invention, after the step of determining that each ID in the ID relationship chain has a target historical unique user identification number associated with the plurality of historical unique user identification numbers and obtaining the target ID associated with the target historical unique user identification number, the method includes:

[0146] Step C1: Based on each ID appearing in the ID relationship chain, pre-encode the ID relationship chain to obtain the first code number;

[0147] Step C2: Pre-encode the ID relationship chain corresponding to the historical unique user identification number to obtain the second encoding number;

[0148] Step C3: Sum the first encoding number and the second encoding number to obtain the encoding chain;

[0149] Step C4: Obtain the encoding bits of other IDs with higher priority than the target ID;

[0150] Step C5: If the encoded bits of other IDs with higher priority than the target ID are greater than or equal to the corresponding encoded values ​​on the encoding chain, then determine the corresponding multiple unique user identification numbers.

[0151] To address the cumbersome process of accurately obtaining the latest CMID from the ID chain, this embodiment provides a novel processing method: a preset encoding method. In this method, encoding is based on whether an ID appears; if it appears, the encoding bit is 1, and if it doesn't, it's 0. Specifically, the ID relationship chain is preset-encoded based on each ID appearing in the chain to obtain a first encoding number. The ID relationship chain corresponding to the historical unique user identifier is also preset-encoded to obtain a second encoding number. The first and second encoding numbers are summed to obtain the encoding chain. The encoding bits of other IDs with higher priority than the target ID are then obtained. If the encoding bits of these other IDs with higher priority than the target ID are greater than or equal to the corresponding encoding values ​​on the encoding chain, multiple unique user identifiers are identified. The core principle is that when using IDs associated with historical CMIDs based on the ID relationship chain, if a high-priority ID associated with the historical CMID conflicts with a high-priority ID in the ID chain (the ID encoding bit value is greater than 1), it indicates a change in ID ownership. In this case, the historical CMID cannot be used directly; a new CMID should be generated as the latest CMID.

[0152] Specifically, if the following ID relationship chain exists...<NULL,idno,telephone,unionid,imei / idfa> After being pre-encoded, it can be represented as <01111> ;and CMID<ecifno,idno,telephone,unionid,imei / idfa> After being pre-encoded, it can be represented as <11111> The first and second encoding numbers are summed to obtain the encoding chain. <01111> + <11111> = <12222>, Here, ecifno needs to be considered (because the priority of ecifno is higher than that of idno). The encoding of ecifno is 1 < 2, indicating that there is no conflict between different CMIDs. However, if telephone is used for association (because the priorities of ecifno and idno are higher than that of telephone), then the encoding bits of ecifno and idno need to be considered. The encoding of idno is 2 = 2, which indicates that there is a conflict at this time. Since only the encoding of the ID with higher priority needs to be considered when using different IDs for association to check whether 2 appears, the recognition efficiency can be improved. In this embodiment, after generating the relationship data between the latest user ID and CMID, the relationship between other extended IDs and CMID can be realized. For example, based on the ID pair data of <telephone, QQ>. For example, <CMID: telephone> data can be used to quickly generate <CMID, QQ> data, and finally construct <CMID, user ID (including ecifno, idno, telephone, unionid, device ID, QQ, etc.)>. In this way, using any user ID, we can find other IDs of the user according to CMID and quickly determine whether different types of user IDs belong to the same person.

[0153] In this embodiment, by performing preset encoding on the ID relationship chain according to each ID that appears on the ID relationship chain, a first encoding number is obtained; performing preset encoding on the ID relationship chain corresponding to the historical unique user identification number, a second encoding number is obtained; adding and processing the first encoding number and the second encoding number to obtain an encoding chain; obtaining the encoding bits of other IDs with higher priority than the target ID; if the encoding bits of other IDs with higher priority than the target ID are greater than or equal to the corresponding encoding values on the encoding chain, multiple corresponding unique user identification numbers are determined. In this embodiment, the generation efficiency of generating unique user identification numbers is improved.

[0154] Refer to Figure 3 , Figure 3 FIG. is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0155] The same-person recognition device in the embodiment of the present invention can be a PC, or can also be a terminal device such as a smart phone, a tablet computer, or a portable computer.

[0156] Such as Figure 3 As shown, the fan identification device may include: a processor 1001, such as a CPU, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to establish communication between the processor 1001 and the memory 1005. The memory 1005 may be a high-speed RAM or a stable, non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0157] Optionally, the person identification device may also include a target user interface, a network interface, a camera, RF (Radio Frequency) circuitry, sensors, audio circuitry, a WiFi module, etc. The target user interface may include a display screen and an input unit such as a keyboard; optionally, the target user interface may also include a standard wired interface or a wireless interface. The network interface may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0158] Those skilled in the art will understand that Figure 3 The structure of the fan identification device shown does not constitute a limitation on the fan identification device. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0159] like Figure 3 As shown, the memory 1005, as a computer storage medium, may include an operating system, a network communication module, and a fan-made identification program. The operating system is a program that manages and controls the hardware and software resources of the fan-made identification device, supporting the operation of the fan-made identification program and other software and / or programs. The network communication module is used to enable communication between the various components within the memory 1005, as well as communication with other hardware and software in the fan-made identification device.

[0160] exist Figure 3 In the fanfiction identification device shown, the processor 1001 is used to execute the fanfiction identification program stored in the memory 1005 to implement the steps of the fanfiction identification method described in any of the above claims.

[0161] The specific implementation of the fanfiction identification device of the present invention is basically the same as the various embodiments of the fanfiction identification method described above, and will not be repeated here.

[0162] Furthermore, this invention also proposes a fan identification device, which is applied to a first unit, and the fan identification device includes:

[0163] The first acquisition module is used to acquire the target ID pair data corresponding to the fan identification instruction when a fan identification instruction is received;

[0164] The first determining module is used to determine whether the unique user identification number of the target user corresponding to the fan identification instruction can be obtained from the preset identification tool based on the target ID;

[0165] In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains, and the ID relationship chains are generated by processing the association between ID pairs extracted from the data of ID pairs corresponding to the same user.

[0166] The second determining module is used to determine that the user corresponding to the unique user identification number is the same person as the target user if the unique user identification number of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool.

[0167] Optionally, the first determining module includes:

[0168] The first acquisition unit is used to acquire data of ID pairs corresponding to the same user at preset time intervals, and clean the ID pair data according to preset regular expression validation rules to obtain cleaned ID pair data.

[0169] The second acquisition unit is used to acquire the preset priority order of each ID;

[0170] The first relationship supplementation unit is used to supplement the relationship of multiple ID pairs in the cleaned ID pair data with a common ID as an indirect association according to the priority order, so as to generate a multi-dimensional ID relationship;

[0171] The second relationship supplementation unit is used to continue to supplement the relationship of each ID pair in the cleaned ID pair data according to the multi-dimensional ID relationship, until each ID pair and ID are used in the processing, so as to generate ID relationship chains of different lengths composed of different IDs in the cleaned ID pair data.

[0172] Optionally, the same person identification device further includes:

[0173] The second acquisition module is used to acquire the first map chain of the highest priority ID according to the priority order, so as to obtain the first type of ID relationship chain, wherein the map chain of any ID is a multi-dimensional ID relationship formed with the ID as the highest priority and starting from the ID;

[0174] The third acquisition module is used to continuously acquire second map chains of IDs with other priority order according to the priority order, and to find IDs with higher priority than the other priority order IDs for the second map chain to obtain a second type of ID relationship chain;

[0175] The processing module is used to process all ID relationship pairs of the cleaned ID pair data into ID relationship chains based on the first type of ID relationship chain and the second type of ID relationship chain.

[0176] Optionally, the same person identification device further includes:

[0177] The fourth acquisition module is used to acquire the highest priority ID in each ID relationship chain. The highest priority ID is set as the seed, and the seed is encrypted using a preset encryption function to generate a unique identifier UID in the ID relationship chain.

[0178] The third determining module is used to determine the final UID corresponding to the lowest priority ID in different ID relationship chains based on the level of the unique identifier UID of the corresponding ID relationship chain when the lowest priority ID is in different ID relationship chains, as well as the occurrence time and number of occurrences of the lowest priority ID in different ID relationship chains.

[0179] The fifth acquisition module is used to acquire the final UID of each ID other than the lowest priority ID, and associate all IDs to form the final ID relationship chain.

[0180] Optionally, the unique user identification number includes the latest unique user identification number, and the preset identification tool includes multiple historical unique user identification numbers. The first determining module includes:

[0181] The first determining unit is used to determine, in an orderly manner, whether each ID in any ID relationship chain can find the associated target historical unique user identification number from the plurality of historical unique user identification numbers, based on the priority of each ID in any ID relationship chain.

[0182] The second determining unit is used to generate the latest unique user identification number according to a preset generation rule if it is determined that none of the IDs in the ID relationship chain can find the associated target historical unique user identification number from the multiple historical unique user identification numbers, so as to obtain multiple latest unique user identification numbers corresponding to the multiple ID relationship chains.

[0183] Optionally, the same person identification device further includes:

[0184] The sixth acquisition module is used to acquire the target ID associated with the target historical unique user identification number if it is determined that each ID in the ID relationship chain has a target historical unique user identification number that has been associated with the multiple historical unique user identification numbers.

[0185] The seventh acquisition module is used to acquire the priority of the target ID, the priority of each ID associated with the target historical unique user identification number associated with the target ID, and the priority of each ID on the ID relationship chain. Based on the priority of the target ID, the priority of each ID associated with the target historical unique user identification number associated with the target ID, and the priority of each ID on the ID relationship chain, the latest unique user identification number of the corresponding ID relationship chain is determined.

[0186] Optionally, the same person identification device further includes:

[0187] The first encoding module is used to pre-encode the ID relationship chain according to each ID appearing in the ID relationship chain to obtain a first encoding number;

[0188] The second encoding module is used to pre-encode the ID relationship chain corresponding to the historical unique user identification number to obtain the second encoding number;

[0189] The summation module is used to sum the first code number and the second code number to obtain the code chain;

[0190] The eighth acquisition module is used to acquire the encoding bits of other IDs with higher priority than the target ID;

[0191] The fourth determining module is used to determine multiple unique user identification numbers if the encoded bits of other IDs with higher priority than the target ID are greater than or equal to the corresponding encoded values ​​on the encoding chain.

[0192] The specific implementation of the fanfiction identification device is basically the same as the various embodiments of the fanfiction identification method described above, and will not be repeated here.

[0193] In addition, the present invention provides a computer medium storing one or more programs, which can be executed by one or more processors to implement the steps of the above-described embodiments of the fanfiction identification method.

[0194] The specific implementation details of the medium (i.e., computer-readable medium) of this invention are basically the same as the embodiments of the above-mentioned methods for identifying similar individuals, and will not be repeated here.

[0195] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0196] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0197] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0198] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.< / idno:telephone> < / unionid:imei> < / idno:telephone>

Claims

1. A method for identifying fanfiction, characterized in that, The method for identifying fellow individuals includes: Upon receiving a fanfiction identification instruction, acquire the target ID pair data corresponding to the fanfiction identification instruction; Based on the target ID, determine whether the unique user identification number of the target user corresponding to the fan identification instruction can be obtained from the preset identification tool; In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains, and the ID relationship chains are generated by processing the association between ID pairs extracted from the data of ID pairs corresponding to the same user. If the unique user identification number of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool, then it is determined that the user corresponding to the unique user identification number is the same person as the target user. The step of obtaining the target ID pair data corresponding to the same person identification instruction includes: The target ID pair data corresponding to the fanfiction identification instruction can be obtained from the fanfiction identification instruction itself, or the target ID pair data corresponding to the fanfiction identification instruction can be obtained from the big data system or big data warehouse pointed to by the fanfiction identification instruction; the target ID pair data includes at least one of the following: preset regular ID, non-preset regular ID; The target ID pair data is inspected. If a non-preset regular ID is detected in the target ID pair data, the mapping relationship between the non-preset regular ID and the preset regular ID is processed by Hive-ETL to improve the scalability of the target ID pair data. The step of generating the ID relationship chain by processing the association relationship between ID pairs extracted from the data of ID pairs corresponding to the same user includes: Data on the ID pairs corresponding to the same user is obtained at preset time intervals. The ID pairs are then cleaned according to preset regular expression validation rules to obtain cleaned ID pairs. Retrieve the preset priority order of each ID; Based on the priority order, multiple ID pairs in the cleaned ID pair data are processed to supplement the relationship by using a common ID as an indirect association, so as to generate a multi-dimensional ID relationship; Based on the multi-ID relationship, the relationship supplementation processing is continued for each ID pair in the cleaned ID pair data until each ID pair and ID are used in the processing, so as to generate ID relationship chains of varying lengths composed of different IDs in the cleaned ID pair data.

2. The method for identifying fanfiction as described in claim 1, characterized in that, The step of obtaining the preset priority order of each ID includes: According to the priority order, the first map chain of the ID with the highest priority is obtained to obtain the first type of ID relationship chain. The map chain of any ID is a multi-dimensional ID relationship formed with the ID as the highest priority and starting from the ID. Based on the priority order, continuously obtain second map chains of IDs with other priority order, and find IDs with higher priority than the other priority order IDs for the second map chains to obtain a second type of ID relationship chain; Based on the first type of ID relationship chain and the second type of ID relationship chain, all ID relationship pairs of the cleaned ID pair data are processed into ID relationship chains.

3. The method for identifying fanfiction as described in claim 2, characterized in that, The step of processing all ID relationship pairs of the cleaned ID pair data into ID relationship chains based on the first type of ID relationship chain and the second type of ID relationship chain includes: Obtain the highest priority ID in each ID relationship chain, and set the highest priority ID as the seed. Encrypt the seed using a preset encryption function to generate a unique identifier UID in the ID relationship chain. Based on the level of the unique identifier UID of the corresponding ID relationship chain when the lowest priority ID is in different ID relationship chains, as well as the occurrence time and number of occurrences of the lowest priority ID in different ID relationship chains, the final UID corresponding to the lowest priority ID in different ID relationship chains is determined; Obtain the final UID of each ID other than the lowest priority ID, and associate all IDs to form the final ID relationship chain.

4. The method for identifying contemporaries as described in any one of claims 1-3, characterized in that, The unique user identification number includes the latest unique user identification number, and the preset identification tool includes multiple historical unique user identification numbers. The step of generating multiple unique user identification numbers based on multiple ID relationship chains includes: Based on the priority of each ID in any ID relationship chain, determine in order whether each ID in the ID relationship chain can find the associated target historical unique user identification number from the multiple historical unique user identification numbers; If it is determined that none of the IDs in the ID relationship chain can find a target historical unique user identification number associated with the multiple historical unique user identification numbers, then the latest unique user identification number is generated according to the preset generation rules to obtain multiple latest unique user identification numbers corresponding to the multiple ID relationship chains.

5. The method for identifying fanfiction as described in claim 4, characterized in that, The step of determining whether each ID in any ID relationship chain can find an associated target historical unique user identification number from the plurality of historical unique user identification numbers, based on the priority of each ID in the ID relationship chain, includes: If it is determined that each ID in the ID relationship chain has a target historical unique user identification number that has been associated with the multiple historical unique user identification numbers, the target ID associated with the target historical unique user identification number is obtained. Obtain the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain. Based on the priority of the target ID, the priority of each ID associated with the target historical unique user identifier associated with the target ID, and the priority of each ID in the ID relationship chain, determine the latest unique user identifier of the corresponding ID relationship chain.

6. The method for identifying fanfiction as described in claim 5, characterized in that, The step of finding the target ID associated with each ID in the ID relationship chain after determining that each ID in the multiple historical unique user identification numbers has a target historical unique user identification number that has been found in the multiple historical unique user identification numbers, and then obtaining the target ID associated with the target historical unique user identification number, includes: The ID relationship chain is pre-coded based on each ID appearing in the ID relationship chain to obtain a first code number; The ID relationship chain corresponding to the historical unique user identification number is pre-coded to obtain the second code number; The first code number and the second code number are summed to obtain the code chain; Obtain the encoded bits of other IDs with higher priority than the target ID; If the encoded bits of other IDs with higher priority than the target ID are greater than or equal to the corresponding encoded values ​​on the encoding chain, then multiple unique user identification numbers are determined.

7. A fanfiction identification device, used to perform the fanfiction identification method as described in any one of claims 1-6, characterized in that, The method for identifying fellow individuals includes: The first acquisition module is used to acquire the target ID pair data corresponding to the fan identification instruction when a fan identification instruction is received; The first determining module is used to determine whether the unique user identification number of the target user corresponding to the fan identification instruction can be obtained from the preset identification tool based on the target ID; In the preset identification tool, multiple unique user identification numbers are generated based on multiple ID relationship chains, and the ID relationship chains are generated by processing the association between ID pairs extracted from the data of ID pairs corresponding to the same user. The second determining module is used to determine that the user corresponding to the unique user identification number is the same person as the target user if the unique user identification number of the target user corresponding to the same person identification instruction can be obtained from the preset identification tool.

8. A device for identifying individuals with similar identities, characterized in that, The device includes: a memory, a processor, and a fanfiction identification program stored in the memory and executable on the processor, wherein the fanfiction identification program, when executed by the processor, implements the steps of the fanfiction identification method as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The medium stores a fanfiction identification program, which, when executed by a processor, implements the steps of the fanfiction identification method as described in any one of claims 1 to 6.

Images