Data processing method and device based on attribution scene

Abstract

The application discloses a data processing method and device based on an attribution scene, and aims to solve the problems of low attribution accuracy and low efficiency of current advertisement click data. The method comprises the following steps: obtaining click data corresponding to advertisement data, and generating a data identifier corresponding to the click data; according to a business attribute corresponding to the click data, the click data is split into attribution dimension data and click information; a first target database narrow table corresponding to the attribution dimension data is determined; a second target database narrow table corresponding to the click information is determined; each database narrow table is created based on different time units; the attribution dimension data and the data identifier are stored in the first target database narrow table in association, and the click information and the data identifier are stored in the second target database narrow table in association, so that an attribution party obtains the click data through different database narrow tables and performs attribution processing. Based on the technical scheme, the data processing method can improve the attribution accuracy and efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, and in particular to a data processing method and device based on an attribution scenario. BACKGROUND

[0002] In today's Internet era, advertising as one of the effective channels to obtain new users has become indispensable. Precise positioning in the advertising process is caused by which specific advertisement click brings installation and activation, which is the attribution process. Attribution is faced with a series of problems and challenges such as large amount of click data, high concurrency of click data, availability of attribution system, performance of attribution system, multiple attribution dimensions, and attribution accuracy. How to solve the above problems has become a headache for various advertisers.

[0003] Currently, for the processing of advertising click data, the general practice in the industry is usually divided into the following three ways: way one, store advertising click data through a traditional relational database. This traditional relational database cannot store a large amount of data and cannot withstand a large amount of concurrency. Way two, use a new database that combines OLAP (On-Line Analytical Processing) and OLTP (On-Line Transaction Processing), such as TIDB database. This database can support OLTP and OLAP, and can withstand massive concurrency of data. However, the TIDB database still has performance bottlenecks, and its capacity for massive concurrency of large data is limited, and it cannot store cold and hot data separately, which leads to the inability to distinguish between cold and hot data when attributing data, further leading to low attribution accuracy. Way three, use a big data solution such as an OLAP database to store advertising click data. This big data solution has problems in supporting the effectiveness of multi-condition structural data queries, resulting in inaccurate data query results and low data attribution accuracy. It can be seen that the several existing data processing methods used in the attribution scenario have low data attribution accuracy and low attribution efficiency. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a data processing method and device based on an attribution scenario, to solve the problem of low attribution accuracy and low efficiency of advertising click data.

[0005] To solve the above technical problems, the embodiments of the present application are implemented as follows:

[0006] On the one hand, the embodiments of the present application provide a data processing method based on an attribution scenario, comprising:

[0007] obtaining click data corresponding to the advertising data, and generating a data identifier corresponding to the click data;

[0008] splitting the click data into attribution dimension data and click information according to a service attribute corresponding to the click data; the service attribute includes user identifier information and / or device identifier information corresponding to the click data; and the click information includes at least one of click content, timestamp information, and the device identifier information;

[0009] determining a first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, a number of first databases, and a number of first database narrow tables included in each of the first databases; and determining a second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, a number of second databases, and a number of second database narrow tables included in each of the second databases; each database narrow table is created based on a different time unit;

[0010] storing the attribution dimension data and the data identifier in the first target database narrow table in association, and storing the click information and the data identifier in the second target database narrow table in association, so that an advertising attribution party obtains the click data through different database narrow tables and performs attribution processing.

[0011] In another aspect, an embodiment of the present application provides a data processing apparatus based on an attribution scenario, including:

[0012] an obtaining and generating module configured to obtain click data corresponding to advertising data, and generate a data identifier corresponding to the click data;

[0013] a splitting module configured to split the click data into attribution dimension data and click information according to a service attribute corresponding to the click data; the service attribute includes user identifier information and / or device identifier information corresponding to the click data; and the click information includes at least one of click content, timestamp information, and the device identifier information;

[0014] a first determining module configured to determine a first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, a number of first databases, and a number of first database narrow tables included in each of the first databases; and determine a second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, a number of second databases, and a number of second database narrow tables included in each of the second databases; each database narrow table is created based on a different time unit;

[0015] The association storage module is configured to store the attribution dimension data and the data identifier in the first target database narrow table and store the click information and the data identifier in the second target database narrow table, so that an advertisement attribution party obtains the click data through different database narrow tables and performs attribution processing.

[0016] In another aspect, an embodiment of the present application provides a data processing device based on an attribution scenario, comprising a processor and a memory electrically connected to the processor, the memory storing a computer program, and the processor is configured to call and execute the computer program from the memory to implement:

[0017] obtaining click data corresponding to advertisement data and generating a data identifier corresponding to the click data;

[0018] According to the business attribute corresponding to the click data, the click data is split into attribution dimension data and click information; the business attribute comprises user identifier information and / or device identifier information corresponding to the click data; and the click information comprises at least one of click content, timestamp information and the device identifier information;

[0019] According to the attribution dimension data, the timestamp information, the number of first databases and the number of first database narrow tables included in each of the first databases, a first target database narrow table corresponding to the attribution dimension data is determined; and according to the data identifier, the timestamp information, the number of second databases and the number of second database narrow tables included in each of the second databases, a second target database narrow table corresponding to the click information is determined; each database narrow table is created based on different time units;

[0020] The attribution dimension data and the data identifier are stored in the first target database narrow table, and the click information and the data identifier are stored in the second target database narrow table, so that an advertisement attribution party obtains the click data through different database narrow tables and performs attribution processing.

[0021] In another aspect, an embodiment of the present application provides a storage medium for storing a computer program, the computer program being executed by a processor to implement the following processes:

[0022] obtaining click data corresponding to advertisement data and generating a data identifier corresponding to the click data;

[0023] According to the business attribute corresponding to the click data, the click data is split into attribution dimension data and click information; the business attribute comprises user identifier information and / or device identifier information corresponding to the click data; and the click information comprises at least one of click content, timestamp information and the device identifier information;

[0024] determine the first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of the first databases and the number of the first database narrow tables included in each of the first databases; and determine the second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of the second databases and the number of the second database narrow tables included in each of the second databases; each of the database narrow tables is created based on different time units;

[0025] store the attribution dimension data and the data identifier in the first target database narrow table and store the click information and the data identifier in the second target database narrow table, so that an advertisement attribution party can obtain the click data through different database narrow tables and perform attribution processing.

[0026] According to the technical scheme of the embodiments of the present application, the click data corresponding to the advertisement data is obtained, the data identifier corresponding to the click data is generated, the click data is split into attribution dimension data and click information according to the business attribute corresponding to the click data, the first target database narrow table corresponding to the attribution dimension data is determined according to the attribution dimension data, the timestamp information, the number of the first databases and the number of the first database narrow tables included in each of the first databases, and the second target database narrow table corresponding to the click information is determined according to the data identifier, the timestamp information, the number of the second databases and the number of the second database narrow tables included in each of the second databases. Each of the database narrow tables is created based on different time units, so that each of the click data can be stored in the database narrow table corresponding to the corresponding time unit based on the timestamp information of each of the click data, the cold and hot data (i.e. data in different periods) is separated and stored, the influence of the overall business and performance of the database in the process of cleaning the cold data can be avoided. Moreover, the attribution dimension data and the data identifier are stored in the first target database narrow table, and the click information and the data identifier are stored in the second target database narrow table, the storage effect of the click data is achieved, the storage of massive click data is supported, and massive concurrent data of large data volume can be tolerated, the efficient storage effect of massive click data is achieved, and the failure of click data storage is avoided. Moreover, the advertisement attribution party can quickly and accurately obtain the click data through different database narrow tables and perform attribution processing in the technical scheme, the search accuracy and search speed of the click data are improved, and the accuracy and efficiency of the advertisement attribution are improved. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to make the technical solutions in the application or the prior art clearer, the accompanying drawings needed in the description of the embodiments or the prior art will be briefly introduced. Obviously, the accompanying drawings in the following description only represent some embodiments described in the application, and other accompanying drawings can be obtained by those skilled in the art without creative labor.

[0028] Figure 1 is a schematic flow chart of a data processing method based on an attribution scene according to an embodiment of the application;

[0029] Figure 2 is a schematic block diagram of a data processing method based on an attribution scene according to an embodiment of the application;

[0030] Figure 3 is a schematic flow chart of a data processing method based on an attribution scene according to another embodiment of the application;

[0031] Figure 4 is a schematic flow chart of a processing method of an attribution request according to an embodiment of the application;

[0032] Figure 5 is a schematic block diagram of a data processing apparatus based on an attribution scene according to an embodiment of the application;

[0033] Figure 6 is a schematic hardware structure diagram of a data processing device based on an attribution scene according to an embodiment of the application. DETAILED DESCRIPTION

[0034] The embodiments of the application provide a data processing method and apparatus based on an attribution scene, to solve the problems of low accuracy and low efficiency of advertisement click data attribution.

[0035] In order to make those skilled in the art better understand the technical solutions in the application, the technical solutions in the embodiments of the application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the application. Obviously, the described embodiments only represent some of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the application.

[0036] In one embodiment, the data processing method based on an attribution scene provided by the application can be applied to an advertisement attribution scene, which can include a commodity purchase attribution scene, an application installation and activation attribution scene, etc. The data processing method based on an attribution scene provided by the embodiments of the application will be described in detail below taking the advertisement attribution scene as an example.

[0037] Figure 1 is a schematic flow chart of a data processing method based on an attribution scene according to an embodiment of the present application, as shown in the figure, the method comprises: Figure 1

[0038] S102, acquiring click data corresponding to the advertising data, and generating data identification corresponding to the click data.

[0039] Among them, the click data corresponding to the advertising data can be acquired from an advertising distribution platform used for displaying the advertising data. Exemplarily, the advertising data can be a promotion advertisement of an application, a promotional advertisement of a commodity, etc.

[0040] Optionally, since the snowflake algorithm can generate an identification with uniqueness in a distributed scene, the data identification corresponding to the click data can be generated based on the snowflake algorithm, and the data identification can be represented by a UUID (Universally Unique Identifier, Universal Unique Identifier).

[0041] S104, according to the business attribute corresponding to the click data, the click data is split into attribution dimension data and click information.

[0042] Among them, the business attribute can include user identification information, device identification information, etc. corresponding to the click data. The click information can include click content, timestamp information, device identification information, etc. described above. Optionally, the device identification information can include Android ID (Android Unique Device Identification), Android ID_MD5 (Android Unique Device Identification encrypted by Message Digest Algorithm), OAID (Open Anonymous Device Identifier, Anonymous Device Identifier), OAID_MD5 (Anonymous Device Identifier encrypted by Message Digest Algorithm), IMEI (International Mobile Equipment Identity, International Mobile Equipment Identity), IMEI_MD5 (International Mobile Equipment Identity encrypted by Message Digest Algorithm), IP (Internet Protocol, Internet Protocol, IP address of user clicking the advertising data) + UA (User Agent, User Agent, information used to identify the user's operating system and version, browser and version, etc.), etc. The user identification information can include the user's communication number. Each user identification information or device identification information can correspond to an attribution dimension.

[0043] ​The business attribute can be a business attribute of an advertisement placement platform generating the click data, and the user identifier information and the device identifier information corresponding to the click data obtained by the advertisement placement platforms of different business attributes are the same or different. For example, in a case where the user identifier information and the device identifier information corresponding to the click data include an Android ID, an OAID, an IMEI_MD5, and a communication number of a user, assuming that the same advertisement data is placed on an advertisement placement platform A (according to the business attribute of the advertisement placement platform A, the advertisement placement platform A can only obtain the Android ID, the OAID, and the communication number of the user corresponding to the click data) and an advertisement placement platform B (according to the business attribute of the advertisement placement platform B, the advertisement placement platform B can only obtain the OAID, the IMEI_MD5, and the communication number of the user corresponding to the click data), then the advertisement placement platform A can only obtain the Android ID, the OAID, and the communication number of the user corresponding to the click data, and the advertisement placement platform B can only obtain the OAID, the IMEI_MD5, and the communication number of the user corresponding to the click data. Thus, according to the business attribute corresponding to the click data, the click data generated by the advertisement placement platform A can be split into the click information and the three attribution dimensions of the Android ID, the OAID, and the communication number of the user, and the click data generated by the advertisement placement platform B can be split into the click information and the three attribution dimensions of the OAID, the IMEI_MD5, and the communication number of the user.

[0044] S106, determining, according to the attribution dimension data, the timestamp information, the number of the first databases, and the number of the first database narrow tables included in each of the first databases, a first target database narrow table corresponding to the attribution dimension data, and determining, according to the data identifier, the timestamp information, the number of the second databases, and the number of the second database narrow tables included in each of the second databases, a second target database narrow table corresponding to the click information.

[0045] The first database can be an attribution dimension database, and the second database can be a click detail database.

[0046] S108, storing the attribution dimension data and the data identifier in association in the first target database narrow table, and storing the click information and the data identifier in association in the second target database narrow table, so that an advertisement attribution party obtains the click data through different database narrow tables and performs attribution processing.

[0047] According to the technical scheme of the embodiment of the present application, the click data corresponding to the advertisement data is obtained, the data identifier corresponding to the click data is generated, the click data is split into attribution dimension data and click information according to the business attribute corresponding to the click data, the first target database narrow table corresponding to the attribution dimension data is determined according to the attribution dimension data, the timestamp information, the number of the first databases and the number of the first database narrow tables included in each first database, and the second target database narrow table corresponding to the click information is determined according to the data identifier, the timestamp information, the number of the second databases and the number of the second database narrow tables included in each second database. The database narrow tables are created based on different time units, so that the click data can be stored in the database narrow table corresponding to the corresponding time unit based on the timestamp information of each click data, the effect of separating and storing cold and hot data (i.e. data in different periods) is achieved, and the impact on the overall business and performance of the database during the process of cleaning the cold data can be avoided. Moreover, the attribution dimension data and the data identifier are stored in the first target database narrow table in association, and the click information and the data identifier are stored in the second target database narrow table in association, the storage effect of splitting the click data into databases and tables is achieved, the storage of massive click data is supported, and massive concurrent data of massive data can be tolerated, the efficient storage effect of massive click data is achieved, and the failure of click data storage is avoided. Moreover, the advertisement attribution party can quickly and accurately obtain the click data and perform attribution processing through different database narrow tables, the search accuracy and speed of the click data are improved, and the accuracy and efficiency of the advertisement attribution are improved.

[0048] In one embodiment, before determining the first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of the first databases and the number of the first database narrow tables included in each first database (i.e. S106), the following steps A1-A4 can be used to create the database and the database narrow table.

[0049] Step A1, obtaining historical click data and historical attribution dimensions corresponding to the historical click data.

[0050] The historical click data can be the historical click data corresponding to the advertisement data in the above S102. When obtaining the historical click data, the historical click data in one attribution period can be obtained according to the attribution period. The attribution period can be one week, two weeks, three weeks, etc.

[0051] Step A2, determining the number of the first databases and the number of the second databases according to the data amount corresponding to the historical click data and the dimension number of the historical attribution dimensions.

[0052] The number of the second database can be determined according to the data volume corresponding to the historical click data and the TPS (Transactions Per Second, system throughput) of a single database. Assuming that the TPS of a single database supports 10,000 concurrent, and the TPS of the historical click data business high point is 10,000 concurrent, the number of the second database is 1.

[0053] Each historical attribution dimension corresponding to the historical click data can be regarded as a piece of data. The number of the first database can be determined by evaluating the dimension number of the historical click data and combining the TPS of a single database. Assuming that the TPS of a single database supports 10,000 concurrent, and the TPS of the historical click data business high point is 100,000 concurrent, the number of the first database is 10.

[0054] In step A3, the number of the first database narrow table and the number of the second database narrow table are determined according to the data volume corresponding to the historical click data and the data volume storage threshold of each database narrow table.

[0055] In this embodiment, the number of the first database narrow table and the number of the second database narrow table can be analyzed according to the data volume corresponding to the historical click data, the dimension number of the historical attribution dimension, and the data volume storage threshold of each database narrow table. Since each database narrow table is created based on different time units, assuming that one week of click data is stored in one time unit, then analyzing the number of the first database narrow table and the number of the second database narrow table is to analyze how many database narrow tables are needed to be established every week.

[0056] Assuming that there are about 200 million data volume in the historical click data every day, and the dimension number of the historical attribution dimension is 10, then the data volume corresponding to the historical click data every day is about 2000 million. If it is divided into 10 databases, then each database has about 200 million data every day, and about 1400 million data every week. If the data volume that can be stored by each database narrow table is 50 million, then about 28 database narrow tables are needed every week. In order to make the database support elastic expansion, at least 64 database narrow tables can be set every week. It should be noted that the database is generally expanded by 2 to the power of N, 2 to the power of 5 is 32, and 2 to the power of 6 is 64. Therefore, in order to make the database more expandable, at least 64 database narrow tables are set every week. In actual application, at least 32 database narrow tables can be set every week, at least 128 database narrow tables can be set every week, and the like, which are not limited in the present application.

[0057] In step A4, the first database and the second database are respectively created according to the number, the first database narrow table is created in the first database, and the second database narrow table is created in the second database.

[0058] In this embodiment, the number of databases and the number of database narrow tables are estimated according to historical click data and historical attribution dimensions, so that each database and database narrow table is created according to the number, so that the number of each database and database narrow table is more reasonable, and as the historical click data is continuously updated, the number of databases and database narrow tables can be changed, so that the corresponding number of databases and database narrow tables can be flexibly created, and the effect of elastic expansion and contraction is realized.

[0059] In one embodiment, the first database can include a plurality of first sub-databases corresponding to a plurality of time units, and each first sub-database includes a plurality of first database narrow tables.

[0060] Because the data volume of click data corresponding to the advertisement data is huge, it can reach tens of thousands of clicks per second, and the click data that has exceeded the attribution period has little significance for attribution, so it is particularly important to realize the hot and cold separation of click data, and the click data that has exceeded the attribution period can be quickly cleaned up without affecting the performance of the database. In the attribution process, the attribution period generally does not exceed seven days when searching for click data according to the activation operation of the advertisement data, so dividing the time unit by week is an effective method to realize the hot and cold separation of data. New click data needs to be stored in the database space of the current week, and the click data of the previous week needs to be queried during attribution, so the click data before the previous week can be emptied at any time. Therefore, three weeks of click data need to be stored to realize the hot and cold separation of click data storage.

[0061] For example, the first database can include three first sub-databases corresponding to three time units, and according to the example in step A3, each first sub-database can include 64 first database narrow tables.

[0062] In this embodiment, the first target database narrow table corresponding to the attribution dimension data is determined according to the attribution dimension data, the timestamp information, the number of first databases, and the number of first database narrow tables included in each first database (i.e., S106), which can be specifically performed as steps B1-B4:

[0063] Step B1, determining the attribution dimension representation value corresponding to the attribution dimension data according to the attribution dimension data.

[0064] Because in the attribution process, the activation information generated based on the activation operation of the advertisement data is matched with multiple attribution dimension data such as Android ID, Android ID_MD5, OAID, OAID_MD5, IMEI, IMEI_MD5, IP+UA, etc., how to realize fast positioning and searching of data and realize average distribution of data is the most important thing in database and table separation, and if multiple attribution dimension data are used with the same database and table separation strategy, a same transformation or encryption algorithm needs to be used.

[0065] Optionally, in the embodiment, the CRC64 (Cyclic Redundancy Check) algorithm can be used to encrypt each attribution dimension data to obtain a CRC64 value (i.e., an attribution dimension representation value) corresponding to each attribution dimension data. The algorithm can ensure that the CRC64 value obtained in the application scenario of massive click data is not repeated, has a high processing efficiency, and the generated data length is moderate. In actual application, other transformation or encryption algorithms can also be used, which are not limited in the present application.

[0066] Step B2, determining a first target database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of the first databases.

[0067] In the embodiment, the number of the first databases can be taken modulo (i.e., taken remainder) according to the attribution dimension representation value to obtain a modulo result, and the sequence number of the first target database corresponding to the attribution dimension data is determined according to the modulo result.

[0068] In the case that the attribution dimension representation value is a CRC64 value and the number of the first databases is 10, step B2 can be specifically performed as follows: taking 10 modulo of the CRC64 value, i.e., CRC64%10, to obtain a modulo result (one of 0-9), and the sequence number of the first target database corresponding to the attribution dimension data is determined according to the modulo result.

[0069] Similarly, in the case that the attribution dimension representation value is a CRC64 value and the number of the first databases is 4, step B2 can be specifically performed as follows: taking 4 modulo of the CRC64 value, i.e., CRC64%4, to obtain a modulo result (one of 0-3), and the sequence number of the first target database corresponding to the attribution dimension data is determined according to the modulo result.

[0070] Step B3, determining a first target time unit corresponding to the timestamp information, and determining a first target sub-database corresponding to the first target time unit as the first target sub-database corresponding to the attribution dimension data.

[0071] In the embodiment, since the time units are divided by weeks, and the first database includes three first sub-databases corresponding to three time units, the number of milliseconds in each week (i.e., 604800000 milliseconds) can be taken modulo according to the timestamp information to obtain how many weeks the timestamp information is since January 1, 1970, and the number 3 can be taken modulo according to the obtained number of weeks to determine the sequence number of the first target time unit corresponding to the timestamp information according to the modulo result (one of 0-2).

[0072] Step B4, determining a first target database narrow table in the first target sub-database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of the first database narrow tables included in the first target sub-database.

[0073] The first target database narrow table corresponding to the attribution dimension data is determined according to the result of the modulo operation.

[0074] In the example in step A3, if the first database includes 64 first database narrow tables in the first target sub-database, and the attribution dimension representation value is a CRC64 value, step B4 can be specifically implemented as follows: a modulo operation of the CRC64 value on 64, i.e., CRC64%64, is performed to obtain a result of the modulo operation (one of 0-63), and the first target database narrow table corresponding to the attribution dimension data is determined according to the result of the modulo operation.

[0075] In the example in the embodiment, the first database includes three first sub-databases corresponding to three time units, and each first sub-database includes 64 first database narrow tables. The overall table name of the first database narrow table storing the attribution dimension data can be "click_${0-2}_${0-63}". This design facilitates quick positioning of the attribution dimension data in a specific database narrow table of a specific time unit, greatly improves the search speed of the attribution dimension data in the attribution process, and thus improves the attribution efficiency.

[0076] In the embodiment, the attribution dimension data, the attribution dimension representation value, and the data identifier corresponding to the click data can be used as table fields of the first target database narrow table, and a common index can be created using the attribution dimension representation value.

[0077] In the embodiment, the attribution dimension representation value corresponding to the attribution dimension data is determined, and then the first target database corresponding to the attribution dimension data is determined according to the attribution dimension representation value and the number of the first database. The first target time unit corresponding to the timestamp information is determined, and the first sub-database corresponding to the first target time unit is determined as the first target sub-database corresponding to the attribution dimension data. The first target database narrow table in the first target sub-database corresponding to the attribution dimension data is determined according to the attribution dimension representation value and the number of the first database narrow table included in the first target sub-database. The effect of cold and hot data separation storage is achieved, and the data distribution in each database and each database narrow table is more uniform.

[0078] In one embodiment, the second database can include a plurality of second sub-databases corresponding to a plurality of time units, and each second sub-database can include a plurality of second database narrow tables.

[0079] Based on the same reasons as the first database, in order to realize the hot and cold separation of the click data stored in the second database, three second sub-databases corresponding to three time units can be set in the second database. Following the example in step A3, each second sub-database can include 64 second database narrow tables.

[0080] In this embodiment, according to the data identifier, the timestamp information, the number of second databases, and the number of second database narrow tables included in each second database, the second target database narrow table corresponding to the click information (i.e., S106) is determined, which can be specifically implemented as steps C1-C3:

[0081] Step C1, according to the data identifier and the number of second databases, the second target database corresponding to the click information is determined.

[0082] Among them, the number of second databases can be taken modulo according to the data identifier to obtain a modulo result, and the serial number of the second target database corresponding to the click information is determined according to the modulo result.

[0083] Assuming that the data identifier is represented by UUID, and the number of second databases is 3, then step C1 can be specifically implemented as: taking 3 modulo according to UUID, i.e., UUID%3, to obtain a modulo result (one of 0-2), and according to the modulo result, the serial number of the second target database corresponding to the click information is determined.

[0084] Step C2, determine the second target time unit corresponding to the timestamp information, and determine the second sub-database corresponding to the second target time unit as the second target sub-database corresponding to the click information.

[0085] Alternatively, the timestamp information can be calculated inversely according to the data identifier generated based on the snowflake algorithm. In this embodiment, since the time unit is divided by week, and the second database includes three second sub-databases corresponding to three time units, the timestamp information can be taken modulo according to the number of milliseconds per week (i.e., 604800000 milliseconds) to obtain how many weeks since January 1, 1970, and according to the obtained number of weeks, the modulo result (one of 0-2) is obtained, and the serial number of the second target time unit corresponding to the timestamp information is determined according to the modulo result.

[0086] Step C3, according to the data identifier and the number of second database narrow tables included in the second target sub-database, the second target database narrow table in the second target sub-database corresponding to the click information is determined.

[0087] Among them, the number of second database narrow tables can be taken modulo according to the data identifier to obtain a modulo result, and the serial number of the second target database narrow table corresponding to the click information is determined according to the modulo result.

[0088] With the example in step A3, in the case that the second target sub-database includes 64 second database narrow tables, if the data identifier is represented by a UUID, step C3 can be specifically implemented as: obtaining a modulo result (one of 0-63) according to the modulo of the UUID against 64, and determining the serial number of the second target database narrow table corresponding to the click information according to the modulo result.

[0089] In this embodiment, taking the second database including 3 second sub-databases corresponding to 3 time units respectively and each second sub-database including 64 second database narrow tables as an example, the overall table name of the second database narrow table storing the click information can be "click_detail_${0-2}_${0-63}". This design facilitates quick positioning of the click information in a specific database narrow table of a specific time unit, greatly improves the search speed of the click information in the attribution process, and thus improves the attribution efficiency.

[0090] In this embodiment, the data identifier corresponding to the click information and the click data can be taken as a table field of the second target database narrow table, and a unique index can be created by using the data identifier corresponding to the click data.

[0091] In this embodiment, according to the data identifier corresponding to the click data and the number of the second database, the second target database corresponding to the click information is determined, the second target time unit corresponding to the timestamp information is determined, and the second sub-database corresponding to the second target time unit is determined as the second target sub-database corresponding to the click information, so as to determine the second target database narrow table in the second target sub-database corresponding to the click information according to the data identifier and the number of the second database narrow tables included in the second target sub-database, thereby realizing the effect of cold and hot data separation storage and making the data distribution in each database and each database narrow table more uniform.

[0092] In one embodiment, after the attribution dimension data and the data identifier are stored in the first target database narrow table and the click information and the data identifier are stored in the second target database narrow table (i.e., S108), the target time unit meeting a preset condition in each time unit can be determined, and the click data stored in the target sub-database corresponding to the target time unit is deleted, so that the space in the target sub-database is released.

[0093] The preset condition can include at least one of the following: the time length from the current time exceeds a preset time length, and the usage rate of the click data stored in the sub-database corresponding to the time unit is lower than a preset threshold.

[0094] Since the attribution period is generally no more than seven days in the attribution process, the time unit is divided in a week unit, which is an effective method to separate cold and hot data. Since the click data of the last week needs to be queried in the attribution process, the click data before the last week can be emptied at any time, and the storage of three weeks of click data in the database can realize the separation of cold and hot click data. Therefore, the preset time length can be two weeks, and the time length from the current time is more than two weeks, which indicates that the click data stored in the sub-database corresponding to the time unit has expired, that is, the part of the click data is cold data.

[0095] When the usage rate of the click data stored in the sub-database corresponding to the time unit is lower than the preset threshold, it indicates that the click data stored in the sub-database corresponding to the time unit is not significant to the attribution, that is, the demand for the part of the click data is not high in the attribution process, that is, the part of the click data is cold data. In this embodiment, the target time unit that meets the preset condition is the time unit corresponding to the cold data.

[0096] Alternatively, a script can be written to periodically traverse each database narrow table every week, and each database narrow table in the target sub-database corresponding to the cold data is emptied by the truncate command to release the hard disk space, thereby realizing the recycling of the database narrow table.

[0097] In this embodiment, the automatic release of the database narrow table space is realized by periodically cleaning the cold data that is no longer used, the purpose of recycling the database narrow table is achieved, the cost is greatly saved, and since each click data is stored based on the time unit, the effect of separating and storing cold and hot data is achieved, so the periodic cleaning of the cold data does not affect the overall business and performance of the database.

[0098] In one embodiment, after the attribution dimension data and the data identifier are stored in the first target database narrow table, and the click information and the data identifier are stored in the second target database narrow table (i.e., S108), the attribution processing of the click data can be realized by the following steps D1-D3:

[0099] Step D1, when receiving the attribution request for the click data, determining the target attribution dimension corresponding to the attribution request.

[0100] The attribution request is generated based on the activation operation of the user on the advertisement data. The attribution request can carry the activation data of the advertisement data, and the activation data can include the attribution dimension data. Determining the target attribution dimension corresponding to the attribution request is to split the activation data into each target attribution dimension according to each attribution dimension data included in the activation data.

[0101] Step D2, according to the target attribution dimension, searching the target data identifier corresponding to the target attribution dimension from the first target database narrow table.

[0102] In this embodiment, the first target database narrow table is indexed by the attribution dimension. When performing step D2, the attribution dimension representation value corresponding to the target attribution dimension can be determined first. Since the attribution dimension data, the attribution dimension representation value, and the data identifier corresponding to the click data are table fields of the first target database narrow table, and the attribution dimension representation value is the index of the first target database narrow table, the target data identifier corresponding to the target attribution dimension can be found from the first target database narrow table based on the attribution dimension representation value corresponding to the target attribution dimension.

[0103] In step D3, the click information corresponding to the target data identifier is found from the second target database narrow table according to the target data identifier, so that the attribution processing is performed by using the click information in the click data.

[0104] In this embodiment, the second target database narrow table is indexed by the data identifier. Since the click information and the data identifier corresponding to the click data are table fields of the second target database narrow table, and the data identifier corresponding to the click data is the index of the second target database narrow table, the click information corresponding to the target data identifier can be found from the second target database narrow table based on the target data identifier.

[0105] In this embodiment, by determining the target attribution dimension corresponding to the attribution request, finding the target data identifier corresponding to the target attribution dimension from the first target database narrow table according to the target attribution dimension, and finding the click information corresponding to the target data identifier from the second target database narrow table according to the target data identifier, the attribution processing is performed by using the click information in the click data, the effect of quickly locating the related database and the specific database narrow table according to the index of the database narrow table is achieved, the data searching accuracy and speed are improved, and thus the attribution accuracy and efficiency are greatly improved.

[0106] In one embodiment, the sharding JDBC (open source middleware for database and table splitting) can be used to carry the above data processing method based on the attribution scene. According to the data processing method based on the attribution scene provided in the embodiments of the present application, by using the sharding method, the CRC64 value of the multiple attribution dimensions and the data identifier of the click data are used for database and table splitting, not only the effects of separating the cold and hot data and evenly distributing the data to different database narrow tables are achieved, but also the purposes of flexibly expanding or shrinking the capacity according to the business, regularly arranging the expired data to support the business needs and save the cost are achieved. The method has the advantages of low cost, fast attribution, uniform data distribution, cold and hot data separation, and supporting elastic expansion, and has a profound reference significance in the attribution field.

[0107] For example, the attribution dimension representation value is CRC64 value, and the data identifier corresponding to the click data is represented by UUID. The first database (i.e. Figure 2The number of attribution dimension libraries (23) is 4, and the second database (i.e. Figure 2 The number of click detail databases 24 is 4, and the time units are created in weeks. Each attribution dimension database includes 3 attribution dimension sub-databases 25 corresponding to each time unit. Each attribution dimension sub-database includes 64 attribution dimension database narrow tables. Each click detail database includes 3 click detail sub-databases 26 corresponding to each time unit. Each click detail sub-database includes 64 click detail database narrow tables. Taking this example, the data processing method based on attribution scenarios provided in the embodiments of this application is described in detail.

[0108] like Figure 2 As shown, click data 20 can be split into attribution dimension data 21 and click information 22. Attribution dimension data 21 includes Android ID and IMEI, while click information 22 includes data identifier UUID. For attribution dimension data 21, firstly, CRC64 values ​​corresponding to Android ID and IMEI can be generated respectively. Then, the modulo of the CRC64 value with the number of attribution dimension libraries 23 (i.e., CRC64%4) is taken to obtain the modulo result (one of 0-3). Based on the modulo result, the attribution dimension libraries corresponding to Android ID and IMEI can be determined. Secondly, based on the timestamp information clickTime of the click data, the modulo of the number of milliseconds per week (604,800,000) can be taken. The modulo result is then taken with 3 to obtain the modulo result. Based on the modulo result, it can be determined in which time unit corresponding to attribution dimension sub-database 25 Android ID and IMEI are stored. Finally, the modulo of the CRC64 value with the number of narrow tables in the attribution dimension library (i.e., CRC64%64) is taken to obtain the modulo result. Based on the modulo result, it can be determined in which narrow table of the attribution dimension library for the corresponding time unit Android ID and IMEI are stored. For click information 22, firstly, the modulo of the number of items in the second database can be used to obtain a modulo result (one of 0-3). Based on the modulo result, the click details database corresponding to the click information can be determined. Secondly, the modulo of the number of milliseconds per week (604,800,000) can be used based on the timestamp information clickTime of the click data. The modulo result can then be used to obtain a modulo result of 3. Based on the modulo result, it can be determined in which time unit the click information is stored in the click details sub-database 26. Finally, the modulo of the number of items in the narrow table of the click details database (i.e., UUID%64) can be used to obtain a modulo result. Based on the modulo result, it can be determined in which narrow table of the click details database is stored in the corresponding time unit.

[0109] Figure 3 This is a schematic flowchart of a data processing method based on attribution scenarios according to another embodiment of this application, such as... Figure 3 As shown, the method includes:

[0110] S301, acquire click data corresponding to the advertisement data, and generate data identification corresponding to the click data.

[0111] S302, according to the service attribute corresponding to the click data, split the click data into attribution dimension data and click information. Then, S303 and S308 are executed respectively.

[0112] The service attribute includes user identification information, device identification information, etc. corresponding to the click data. The click information includes click content, timestamp information, device identification information, etc.

[0113] In an embodiment, before S303 and S308 are executed, the number of the first database and the number of the second database can be determined according to the data amount of the historical click data and the dimension number of the historical attribution dimension corresponding to the historical click data, the number of the first database narrow table and the number of the second database narrow table can be determined according to the data amount of the historical click data and the data amount threshold of each database narrow table, and then the first database and the second database are respectively created according to the number, the first database narrow table is created in the first database, and the second database narrow table is created in the second database.

[0114] S303, according to the attribution dimension data, determine the attribution dimension representation value corresponding to the attribution dimension data.

[0115] Optionally, each attribution dimension data can be encrypted by using the CRC64 algorithm to obtain the CRC64 value (i.e. the attribution dimension representation value) corresponding to each attribution dimension data.

[0116] S304, according to the attribution dimension representation value and the number of the first database, determine the first target database corresponding to the attribution dimension data.

[0117] The first database can be an attribution dimension database. The first database can include a plurality of first sub-databases corresponding to time units, and each first sub-database can include a plurality of first database narrow tables.

[0118] S305, determine the first target time unit corresponding to the timestamp information, and determine the first sub-database corresponding to the first target time unit as the first target sub-database corresponding to the attribution dimension data.

[0119] S306, according to the attribution dimension representation value and the number of the first database narrow table included in the first target sub-database, determine the first target database narrow table in the first target sub-database corresponding to the attribution dimension data.

[0120] S307, store the attribution dimension data and the data identifier into the first target database narrow table.

[0121] S308, determine the second target database corresponding to the click information according to the data identifier and the number of the second databases.

[0122] The second database can be a click detail database. The second database can include a plurality of second sub-databases corresponding to a plurality of time units, and each second sub-database can include a plurality of second database narrow tables.

[0123] S309, determine the second target time unit corresponding to the timestamp information, and determine the second target sub-database corresponding to the second target time unit as the second target sub-database corresponding to the click information.

[0124] S310, determine the second target database narrow table in the second target sub-database corresponding to the click information according to the data identifier and the number of the second database narrow tables included in the second target sub-database.

[0125] S311, store the click information and the data identifier into the second target database narrow table.

[0126] In an embodiment, after S307 and S311 are performed, the target time unit satisfying a preset condition in each time unit can be determined, and the click data stored in the target sub-database corresponding to the target time unit can be deleted, so that the space in the target sub-database is released.

[0127] The preset condition can include at least one of the following: the time length from the current time exceeds a preset time length, and the usage rate of the click data stored in the sub-database corresponding to the time unit is lower than a preset threshold.

[0128] The specific processes of S301-S311 are described in detail in the above embodiments, and will not be repeated here.

[0129] According to the technical scheme of the embodiment of the present application, the click data corresponding to the advertisement data is obtained, the data identifier corresponding to the click data is generated, the click data is split into attribution dimension data and click information according to the service attribute corresponding to the click data, the first target database narrow table corresponding to the attribution dimension data is determined according to the attribution dimension data, the timestamp information, the number of the first databases and the number of the first database narrow tables included in each first database, and the second target database narrow table corresponding to the click information is determined according to the data identifier, the timestamp information, the number of the second databases and the number of the second database narrow tables included in each second database. Each database narrow table is created based on different time units, so that each click data can be stored in the database narrow table corresponding to the corresponding time unit based on the timestamp information of each click data, the effect of separating and storing cold and hot data (i.e. data in different periods) is realized, and the influence of the overall business and performance of the database in the process of cleaning the cold data can be avoided. Moreover, the attribution dimension data and the data identifier are stored in the first target database narrow table in association, and the click information and the data identifier are stored in the second target database narrow table in association, the storage effect of splitting the click data into databases and tables is realized, not only the storage of massive click data is supported, but also massive concurrency of massive data can be tolerated, the efficient storage effect of massive click data is realized, and the situation of click data storage failure is avoided. Moreover, the advertisement attribution party can quickly and accurately obtain the click data and perform attribution processing through different database narrow tables in the technical scheme, the search accuracy and speed of the click data are improved, and the accuracy and efficiency of the advertisement attribution are improved.

[0130] Figure 4 is a schematic flow chart of a processing method of an attribution request according to an embodiment of the present application, as shown in Figure 4 , the method comprises:

[0131] S401, when receiving the attribution request for the click data, determining the target attribution dimension corresponding to the attribution request.

[0132] Wherein, the attribution request is generated based on the activation operation of the user on the advertisement data.

[0133] S402, according to the target attribution dimension, searching the target data identifier corresponding to the target attribution dimension from the first target database narrow table.

[0134] Wherein, the first target database narrow table is indexed by the attribution dimension.

[0135] S403, according to the target data identifier, searching the click information corresponding to the target data identifier from the second target database narrow table, so as to perform attribution processing by using the click information in the click data.

[0136] The second target database narrow table takes the data identifier as an index.

[0137] The specific processes of S401-S403 are described in the above embodiments, and will not be repeated here.

[0138] By determining the target attribution dimension corresponding to the attribution request, finding the target data identifier corresponding to the target attribution dimension from the first target database narrow table according to the target attribution dimension, and finding the click information corresponding to the target data identifier from the second target database narrow table according to the target data identifier, the attribution processing is performed by using the click information in the click data, the effect of quickly locating the related database and the specific database narrow table according to the database narrow table index is achieved, the data searching accuracy and speed are improved, and the attribution accuracy and efficiency are greatly improved.

[0139] In summary, specific embodiments of the present subject matter have been described. Other embodiments are within the scope of the following claims. In some cases, the actions noted in the claims can be performed in a different order and still achieve the desired results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0140] The above is the data processing method based on the attribution scene provided by the embodiment of the present application. Based on the same idea, the embodiment of the present application also provides a data processing device based on an attribution scene.

[0141] Figure 5 is a schematic block diagram of a data processing device based on an attribution scene according to an embodiment of the present application, as Figure 5 shown, the data processing device based on the attribution scene can include:

[0142] The acquisition and generation module 510 is configured to acquire the click data corresponding to the advertisement data, and generate the data identifier corresponding to the click data.

[0143] The splitting module 520 is configured to split the click data into attribution dimension data and click information according to the business attribute corresponding to the click data; the business attribute includes at least one of the user identifier information and / or the device identifier information corresponding to the click data; and the click information includes at least one of the click content, the timestamp information, and the device identifier information.

[0144] The first determining module 530 is configured to determine a first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, a quantity of the first databases, and a quantity of the first database narrow tables included in each of the first databases; and determine a second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, a quantity of the second databases, and a quantity of the second database narrow tables included in each of the second databases; and each of the database narrow tables is created based on different time units.

[0145] The association storage module 540 is configured to store the attribution dimension data and the data identifier in the first target database narrow table, and store the click information and the data identifier in the second target database narrow table, so that an advertisement attribution party obtains the click data through different database narrow tables and performs attribution processing.

[0146] In an embodiment, the first database includes a plurality of first sub-databases corresponding to a plurality of time units; and each of the first sub-databases includes a plurality of first database narrow tables.

[0147] The first determining module 530 includes:

[0148] The first determining unit is configured to determine an attribution dimension representation value corresponding to the attribution dimension data according to the attribution dimension data.

[0149] The second determining unit is configured to determine a first target database corresponding to the attribution dimension data according to the attribution dimension representation value and the quantity of the first databases.

[0150] The third determining unit is configured to determine a first target time unit corresponding to the timestamp information, and determine a first target sub-database corresponding to the first target time unit as the first target sub-database corresponding to the attribution dimension data.

[0151] The fourth determining unit is configured to determine a first target database narrow table in the first target sub-database corresponding to the attribution dimension data according to the attribution dimension representation value and a quantity of the first database narrow tables included in the first target sub-database.

[0152] In an embodiment, the second database includes a plurality of second sub-databases corresponding to a plurality of time units; and each of the second sub-databases includes a plurality of second database narrow tables.

[0153] The first determining module 530 includes:

[0154] The fifth determining unit is configured to determine a second target database corresponding to the click information according to the data identifier and the quantity of the second databases.

[0155] The sixth determining unit is configured to determine a second target time unit corresponding to the timestamp information, and determine a second target sub-database corresponding to the second target time unit as the second target sub-database corresponding to the click information.

[0156] The seventh determining unit is configured to determine, according to the data identifier and the number of second database narrow tables included in the second target sub-database, a second target database narrow table in the second target sub-database corresponding to the click information.

[0157] In an embodiment, the data processing apparatus based on an attribution scenario further includes:

[0158] The second determining module is configured to determine a target time unit that meets a preset condition in each time unit; the preset condition includes at least one of the following: a time length from a current time exceeds a preset time length, and a usage rate of click data stored in a sub-database corresponding to the time unit is lower than a preset threshold value.

[0159] The deleting module is configured to delete the click data stored in the target sub-database corresponding to the target time unit, so that a space in the target sub-database is released.

[0160] In an embodiment, the data processing apparatus based on an attribution scenario further includes:

[0161] The third determining module is configured to determine a target attribution dimension corresponding to an attribution request when the attribution request for the click data is received; the attribution request is generated based on an activation operation of a user on advertisement data.

[0162] The first finding module is configured to find, according to the target attribution dimension, a target data identifier corresponding to the target attribution dimension from a first target database narrow table; the first target database narrow table is indexed by an attribution dimension.

[0163] The second finding module is configured to find, according to the target data identifier, click information corresponding to the target data identifier from a second target database narrow table, so that the attribution processing is performed by using the click information in the click data; the second target database narrow table is indexed by a data identifier.

[0164] In an embodiment, the data processing apparatus based on an attribution scenario further includes:

[0165] The obtaining module is configured to obtain historical click data and a historical attribution dimension corresponding to the historical click data.

[0166] The fourth determining module is configured to determine the number of first databases and the number of second databases according to a data amount corresponding to the historical click data and a dimension number of the historical attribution dimension.

[0167] The fifth determining module is configured to determine the number of the first database narrow tables and the number of the second database narrow tables according to the data volume corresponding to the historical click data and the data volume storage threshold of each database narrow table;

[0168] The creating module is configured to create the first database and the second database according to the number respectively, and create the first database narrow tables in the first database and the second database narrow tables in the second database.

[0169] The device provided by the embodiment of the present application can obtain the click data corresponding to the advertisement data, generate the data identifier corresponding to the click data, split the click data into attribution dimension data and click information according to the service attribute corresponding to the click data, determine the first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of the first databases and the number of the first database narrow tables included in each first database, and determine the second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of the second databases and the number of the second database narrow tables included in each second database. Each database narrow table is created based on different time units, so that each click data can be stored in the database narrow table corresponding to the corresponding time unit based on the timestamp information of each click data, the cold and hot data (i.e. data in different periods) is separated and stored, and the influence of the overall business and performance of the database in the process of cleaning the cold data can be avoided. In addition, the attribution dimension data and the data identifier are stored in the first target database narrow table in association, and the click information and the data identifier are stored in the second target database narrow table in association, which realizes the storage effect of the click data in the database and the table, supports the storage of massive click data, can withstand massive concurrent data of large data volume, realizes the efficient storage effect of massive click data, and avoids the failure of click data storage. In addition, the advertisement attribution party in the device can quickly and accurately obtain the click data and perform attribution processing through different database narrow tables, which improves the search accuracy and search speed of the click data, and thus improves the accuracy and efficiency of the advertisement attribution.

[0170] Those skilled in the art should understand that, Figure 5 The data processing device based on the attribution scene in the embodiment of the present application can be used to realize the data processing method based on the attribution scene described in the foregoing method part, and the details described in the embodiment should be similar to the description in the foregoing method part. To avoid tediousness, no further description is given here.

[0171] Based on the same idea, the embodiment of the present application also provides a data processing device based on an attribution scene, which comprises a data obtaining unit, a data processing unit and a data storage unit. Figure 6The illustrated. Based on the attribution scene data processing equipment can be different due to configuration or performance, can include one or more processors 601 and memory 602, memory 602 can be stored in one or more storage applications or data. Among them, the memory 602 can be temporary storage or persistent storage. The application stored in the memory 602 can include one or more modules (not shown in the figure), each module can include a series of computer executable instructions in the attribution scene based data processing equipment. Further, the processor 601 can be configured to communicate with the memory 602, execute a series of computer executable instructions in the memory 602 on the attribution scene based data processing equipment. The attribution scene based data processing equipment can also include one or more power supplies 603, one or more wired or wireless network interfaces 604, one or more input output interfaces 605, one or more keyboards 606.

[0172] In particular in this embodiment, the attribution scene based data processing equipment includes a memory, and one or more programs, wherein one or more programs are stored in the memory, and one or more programs can include one or more modules, and each module can include a series of computer executable instructions in the attribution scene based data processing equipment, and the one or more processors are configured to execute the one or more programs include the following computer executable instructions:

[0173] Obtaining the click data corresponding to the advertising data, generating the data identifier corresponding to the click data;

[0174] According to the business attribute corresponding to the click data, the click data is divided into attribution dimension data and click information; the business attribute includes user identification information and / or device identification information corresponding to the click data; the click information includes at least one of the click content, the timestamp information and the device identification information;

[0175] According to the attribution dimension data, the timestamp information, the number of first databases and the number of first database narrow tables included in each first database, the first target database narrow table corresponding to the attribution dimension data is determined; and according to the data identifier, the timestamp information, the number of second databases and the number of second database narrow tables included in each second database, the second target database narrow table corresponding to the click information is determined; each database narrow table is created based on different time units;

[0176] The attribution dimension data and the data identifier are associated and stored in the first target database narrow table, and the click information and the data identifier are associated and stored in the second target database narrow table, so that the advertising attribution party obtains the click data through different database narrow tables and performs attribution processing.

[0177] Optionally, the first database comprises a plurality of first sub-databases corresponding to a plurality of time units; each first sub-database comprises a plurality of first database narrow tables; the computer executable instructions, when executed, further cause the processor to:

[0178] determine, according to the attribution dimension data, an attribution dimension representation value corresponding to the attribution dimension data;

[0179] determine, according to the attribution dimension representation value and the number of the first databases, a first target database corresponding to the attribution dimension data;

[0180] determine a first target time unit corresponding to the timestamp information, and determine a first sub-database corresponding to the first target time unit as a first target sub-database corresponding to the attribution dimension data;

[0181] determine, according to the attribution dimension representation value and the number of the first database narrow tables included in the first target sub-database, a first target database narrow table in the first target sub-database corresponding to the attribution dimension data.

[0182] Optionally, the second database comprises a plurality of second sub-databases corresponding to a plurality of time units; each second sub-database comprises a plurality of second database narrow tables; the computer executable instructions, when executed, further cause the processor to:

[0183] determine, according to the data identifier and the number of the second databases, a second target database corresponding to the click information;

[0184] determine a second target time unit corresponding to the timestamp information, and determine a second sub-database corresponding to the second target time unit as a second target sub-database corresponding to the click information;

[0185] determine, according to the data identifier and the number of the second database narrow tables included in the second target sub-database, a second target database narrow table in the second target sub-database corresponding to the click information.

[0186] Optionally, the computer executable instructions, when executed, further cause the processor to:

[0187] determine a target time unit in each time unit that meets a preset condition; the preset condition comprises at least one of the following: a time length from a current time exceeds a preset time length, a usage rate of click data stored in a sub-database corresponding to the time unit is lower than a preset threshold; delete the click data stored in a target sub-database corresponding to the target time unit, so that a space in the target sub-database is released.

[0188] Optionally, the computer executable instructions, when executed, further cause the processor to:

[0189] When receiving the attribution request for the click data, determine a target attribution dimension corresponding to the attribution request; the attribution request is generated based on the activation operation of the user on the advertisement data;

[0190] According to the target attribution dimension, find a target data identifier corresponding to the target attribution dimension from a first target database narrow table; the first target database narrow table is indexed by the attribution dimension;

[0191] According to the target data identifier, find click information corresponding to the target data identifier from a second target database narrow table, so as to perform attribution processing by using the click information in the click data; the second target database narrow table is indexed by the data identifier.

[0192] Optionally, the computer executable instructions, when executed, can also cause the processor to:

[0193] Obtain historical click data and historical attribution dimensions corresponding to the historical click data;

[0194] According to the data amount corresponding to the historical click data and the dimension number of the historical attribution dimensions, determine the number of the first databases and the number of the second databases;

[0195] According to the data amount corresponding to the historical click data and the data amount storage threshold of each database narrow table, determine the number of the first database narrow tables and the number of the second database narrow tables;

[0196] According to the number, respectively create the first databases and the second databases, and create the first database narrow tables in the first databases, and create the second database narrow tables in the second databases.

[0197] The device of the embodiment of the application generates the data identifier corresponding to the click data by acquiring the click data corresponding to the advertisement data, splits the click data into attribution dimension data and click information according to the service attribute corresponding to the click data, and determines the first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of the first databases and the number of the first database narrow tables included in each first database, and determines the second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of the second databases and the number of the second database narrow tables included in each second database. Each database narrow table is created based on different time units, so that each click data can be stored in the database narrow table corresponding to the corresponding time unit based on the timestamp information of each click data, the effect of separating and storing cold and hot data (i.e. data in different periods) is achieved, and the impact on the overall business and performance of the database during the process of cleaning cold data can be avoided. Moreover, the attribution dimension data and the data identifier are stored in the first target database narrow table in association, and the click information and the data identifier are stored in the second target database narrow table in association, the storage effect of splitting the click data into databases and tables is achieved, not only the storage of massive click data is supported, but also massive concurrent data of large data volume can be tolerated, the efficient storage effect of massive click data is achieved, and the situation of click data storage failure is avoided. Moreover, the advertisement attribution party in the device can quickly and accurately acquire the click data through different database narrow tables and perform attribution processing, the search accuracy and search speed of the click data are improved, and the accuracy and efficiency of advertisement attribution are improved.

[0198] The embodiment of the application further provides a storage medium storing one or more computer programs, the one or more computer programs including instructions capable of causing an electronic device including a plurality of application programs to execute the data processing method based on the attribution scenario when the instructions are executed by the electronic device, and specifically used for executing:

[0199] acquiring click data corresponding to advertisement data, and generating a data identifier corresponding to the click data;

[0200] splitting the click data into attribution dimension data and click information according to a service attribute corresponding to the click data; the service attribute includes user identifier information and / or device identifier information corresponding to the click data; and the click information includes at least one of click content, timestamp information and device identifier information;

[0201] determine the first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of the first databases, and the number of the first database narrow tables included in each of the first databases; and determine the second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of the second databases, and the number of the second database narrow tables included in each of the second databases; each of the database narrow tables is created based on different time units;

[0202] store the attribution dimension data and the data identifier in the first target database narrow table in association, and store the click information and the data identifier in the second target database narrow table in association, so that the advertisement attribution party obtains the click data through different database narrow tables and performs attribution processing.

[0203] Optionally, the first database includes a plurality of first sub-databases corresponding to a plurality of time units; each of the first sub-databases includes a plurality of first database narrow tables; when the instruction is executed by the electronic device including a plurality of application programs, the electronic device is further caused to perform:

[0204] determine the attribution dimension representation value corresponding to the attribution dimension data according to the attribution dimension data;

[0205] determine the first target database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of the first databases;

[0206] determine the first target time unit corresponding to the timestamp information, and determine the first sub-database corresponding to the first target time unit as the first target sub-database corresponding to the attribution dimension data;

[0207] determine the first target database narrow table in the first target sub-database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of the first database narrow tables included in the first target sub-database.

[0208] Optionally, the second database includes a plurality of second sub-databases corresponding to a plurality of time units; each of the second sub-databases includes a plurality of second database narrow tables; when the instruction is executed by the electronic device including a plurality of application programs, the electronic device is further caused to perform:

[0209] determine the second target database corresponding to the click information according to the data identifier and the number of the second databases;

[0210] determine the second target time unit corresponding to the timestamp information, and determine the second sub-database corresponding to the second target time unit as the second target sub-database corresponding to the click information;

[0211] determine the second target database narrow table in the second target sub-database corresponding to the click information according to the data identifier and the number of the second database narrow tables included in the second target sub-database.

[0212] Optionally, the instructions, when executed by the electronic device comprising a plurality of application programs, can further cause the electronic device to perform:

[0213] determining a target time unit in each time unit that meets a preset condition; the preset condition comprises at least one of the following: a time length from a current time exceeds a preset time length, a usage rate of click data stored in a sub-database corresponding to the time unit is lower than a preset threshold;

[0214] deleting the click data stored in a target sub-database corresponding to the target time unit, so that the space in the target sub-database is released.

[0215] Optionally, the instructions, when executed by the electronic device comprising a plurality of application programs, can further cause the electronic device to perform:

[0216] when receiving an attribution request for click data, determining a target attribution dimension corresponding to the attribution request; the attribution request is generated based on an activation operation of a user on advertisement data;

[0217] According to the target attribution dimension, find the target data identifier corresponding to the target attribution dimension from the first target database narrow table; the first target database narrow table is indexed by the attribution dimension;

[0218] According to the target data identifier, find the click information corresponding to the target data identifier from the second target database narrow table, so as to perform attribution processing by using the click information in the click data; the second target database narrow table is indexed by the data identifier.

[0219] Optionally, the instructions, when executed by the electronic device comprising a plurality of application programs, can further cause the electronic device to perform:

[0220] obtain historical click data and historical attribution dimensions corresponding to the historical click data;

[0221] According to the data amount corresponding to the historical click data and the dimension number of the historical attribution dimensions, determine the number of the first databases and the number of the second databases;

[0222] According to the data amount corresponding to the historical click data and the data amount storage threshold of each database narrow table, determine the number of the first database narrow tables and the number of the second database narrow tables;

[0223] According to the number, respectively create the first database and the second database, and create the first database narrow table in the first database, and create the second database narrow table in the second database.

[0224] The storage medium of the embodiment of the application can generate data identifiers corresponding to the click data by acquiring the click data corresponding to the advertisement data, split the click data into attribution dimension data and click information according to the service attribute corresponding to the click data, and determine the first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of the first databases, and the number of the first database narrow tables included in each of the first databases, and determine the second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of the second databases, and the number of the second database narrow tables included in each of the second databases. Each database narrow table is created based on different time units, so that each click data can be stored in the database narrow table corresponding to the corresponding time unit based on the timestamp information of each click data, and the effect of separating and storing cold and hot data (i.e. data in different periods) is achieved, which can avoid affecting the overall business and performance of the database during the process of cleaning cold data. Moreover, the attribution dimension data and the data identifier are stored in the first target database narrow table in association, and the click information and the data identifier are stored in the second target database narrow table in association, which achieves the storage effect of splitting the click data into databases and tables, supports the storage of massive click data, can withstand massive concurrent data, achieves the effect of efficient storage of massive click data, and avoids the failure of click data storage. Moreover, the advertisement attribution party in the storage medium can quickly and accurately acquire the click data and perform attribution processing through different database narrow tables, which improves the accuracy and speed of searching the click data, and thus improves the accuracy and efficiency of advertisement attribution.

[0225] The system, apparatus, module or unit illustrated in the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer may, for example, be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0226] For the convenience of description, the above apparatus is described as various units respectively described in functions. Of course, the functions of each unit can be implemented in the same or more software and / or hardware in the implementation of the present application.

[0227] Those skilled in the art will appreciate that embodiments of the application can be readily used as software, hardware, or a combination of software and hardware. In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0228] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing one or more functions specified in the flowchart block or blocks.

[0229] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing one or more functions specified in the flowchart block or blocks.

[0230] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing one or more functions specified in the flowchart block or blocks.

[0231] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0232] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), optical storage, and / or flash memory. The memory is an example of computer readable storage media.

[0233] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0234] It should also be noted that the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, article or apparatus that includes a list of elements does not only include those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article or apparatus. Without more limitations, the element defined by the statement "including a" does not exclude the presence of additional identical elements in the process, method, article or apparatus including the element.

[0235] The present application can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform particular tasks or implement particular abstract data types. The present application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network. In a distributed computing environment, program modules can be located in both local and remote computer storage media including storage devices.

[0236] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment focuses on the difference from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0237] The above merely provides an example of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application should be included in the scope of claims of the present application.

Claims

1. A data processing method based on an attribution scenario, characterized in that, The method comprises: obtaining click data corresponding to advertising data, and generating a data identifier corresponding to the click data; splitting the click data into attribution dimension data and click information according to a service attribute corresponding to the click data; the service attribute comprises user identifier information and / or device identifier information corresponding to the click data; the click information comprises at least one of click content, timestamp information and the device identifier information; determining a number of first databases and a number of second databases according to a data volume corresponding to historical click data and a dimension number of historical attribution dimensions; determining a first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of first databases and a number of first database narrow tables included in each of the first databases, and determining a second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of second databases and a number of second database narrow tables included in each of the second databases; each database narrow table is created based on different time units; storing the attribution dimension data and the data identifier in association in the first target database narrow table, and storing the click information and the data identifier in association in the second target database narrow table, so that an advertising attribution party obtains the click data through different database narrow tables and performs attribution processing; the first database comprises a plurality of first sub-databases corresponding to the time units; each of the first sub-databases comprises a plurality of first database narrow tables; determining a first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of first databases and the number of first database narrow tables included in each of the first databases, comprises: determining an attribution dimension representation value corresponding to the attribution dimension data according to the attribution dimension data; determining a first target database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of first databases; determining a first target time unit corresponding to the timestamp information, and determining the first sub-database corresponding to the first target time unit as the first target sub-database corresponding to the attribution dimension data; determining the first target database narrow table in the first target sub-database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of first database narrow tables included in the first target sub-database; the second database comprises a plurality of second sub-databases corresponding to the time units; each of the second sub-databases comprises a plurality of second database narrow tables; determining a second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of second databases and the number of second database narrow tables included in each of the second databases, comprises: determining a second target database corresponding to the click information according to the data identifier and the number of second databases; determine a second target time unit corresponding to the timestamp information, and determine the second target sub-database corresponding to the second target time unit as a second target sub-database corresponding to the click information; determine the second target database narrow table in the second target sub-database corresponding to the click information according to the data identifier and the number of the second database narrow tables included in the second target sub-database.

2. The method of claim 1, wherein, After the attribution dimension data and the data identifier are stored in the first target database narrow table and the click information and the data identifier are stored in the second target database narrow table, the method further comprises: determine a target time unit in each of the time units that meets a preset condition; the preset condition comprises at least one of the following: a time length from a current time exceeds a preset time length, and a usage rate of the click data stored in a sub-database corresponding to the time unit is lower than a preset threshold; delete the click data stored in a target sub-database corresponding to the target time unit, so that the space in the target sub-database is released.

3. The method of claim 1, wherein, After the attribution dimension data and the data identifier are stored in the first target database narrow table and the click information and the data identifier are stored in the second target database narrow table, the method further comprises: when receiving an attribution request for the click data, determine a target attribution dimension corresponding to the attribution request; the attribution request is generated based on an activation operation of a user on the advertisement data; according to the target attribution dimension, find a target data identifier corresponding to the target attribution dimension from the first target database narrow table; the first target database narrow table is indexed by the attribution dimension; according to the target data identifier, find the click information corresponding to the target data identifier from the second target database narrow table, so that the click information in the click data is used for attribution processing; the second target database narrow table is indexed by the data identifier.

4. The method of claim 1, wherein, Before the first target database narrow table corresponding to the attribution dimension data is determined according to the attribution dimension data, the timestamp information, the number of first databases, and the number of first database narrow tables included in each of the first databases, the method further comprises: obtain historical click data and historical attribution dimensions corresponding to the historical click data; determine the number of first databases and the number of second databases according to the data amount corresponding to the historical click data and the dimension number of the historical attribution dimensions; determine the number of first database narrow tables and the number of second database narrow tables according to the data amount corresponding to the historical click data and the data amount storage threshold of each database narrow table; create the first database and the second database according to the number, respectively, and create the first database narrow table in the first database, and create the second database narrow table in the second database.

5. An attribution-scene-based data processing apparatus, characterized by comprising: comprise: an obtaining and generating module, configured to obtain click data corresponding to advertisement data, and generate a data identifier corresponding to the click data; The splitting module is configured to split the click data into attribution dimension data and click information according to a service attribute corresponding to the click data; the service attribute includes user identification information and / or device identification information corresponding to the click data; and the click information includes at least one of click content, timestamp information, and the device identification information; The fourth determining module is configured to determine the number of the first databases and the number of the second databases according to a data volume corresponding to historical click data and a dimension number of historical attribution dimensions; The first determining module is configured to determine a first target database narrow table corresponding to the attribution dimension data according to the attribution dimension data, the timestamp information, the number of the first databases, and a number of first database narrow tables included in each of the first databases, and determine a second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of the second databases, and a number of second database narrow tables included in each of the second databases; Each database narrow table is created based on different time units; The association storage module is configured to store the attribution dimension data and the data identifier in the first target database narrow table, and store the click information and the data identifier in the second target database narrow table, so that an advertisement attribution party obtains the click data through different database narrow tables and performs attribution processing; The first database includes a plurality of first sub-databases corresponding to the time units; and each of the first sub-databases includes a plurality of the first database narrow tables; The first determining module includes: A first determining unit configured to determine an attribution dimension representation value corresponding to the attribution dimension data according to the attribution dimension data; A second determining unit configured to determine a first target database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of the first databases; A third determining unit configured to determine a first target time unit corresponding to the timestamp information, and determine the first sub-database corresponding to the first target time unit as the first target sub-database corresponding to the attribution dimension data; A fourth determining unit configured to determine the first target database narrow table in the first target sub-database corresponding to the attribution dimension data according to the attribution dimension representation value and a number of the first database narrow tables included in the first target sub-database; The second database includes a plurality of second sub-databases corresponding to time units; and each of the second sub-databases includes a plurality of second database narrow tables; The first determining module includes: A fifth determining unit configured to determine a second target database corresponding to the click information according to the data identifier and the number of the second databases; A sixth determining unit configured to determine a second target time unit corresponding to the timestamp information, and determine the second sub-database corresponding to the second target time unit as the second target sub-database corresponding to the click information; A seventh determination unit is configured to determine a second target database narrow table in a second target sub-database corresponding to the click information according to the data identifier and the number of second database narrow tables included in the second target sub-database.

6. An attribution-scene-based data processing device, characterized by comprising: The processor is configured to call and execute the computer program stored in the memory to implement: obtaining click data corresponding to advertising data, and generating a data identifier corresponding to the click data; According to the business attribute corresponding to the click data, the click data is split into attribution dimension data and click information; the business attribute includes user identifier information and / or device identifier information corresponding to the click data; the click information includes at least one of click content, timestamp information and the device identifier information; According to the data volume corresponding to the historical click data and the dimension number of the historical attribution dimension, the number of first databases and the number of second databases are determined; According to the attribution dimension data, the timestamp information, the number of first databases and the number of first database narrow tables included in each of the first databases, a first target database narrow table corresponding to the attribution dimension data is determined; and according to the data identifier, the timestamp information, the number of second databases and the number of second database narrow tables included in each of the second databases, a second target database narrow table corresponding to the click information is determined; Each database narrow table is created based on different time units; The attribution dimension data and the data identifier are stored in the first target database narrow table, and the click information and the data identifier are stored in the second target database narrow table, so that the advertising attribution party obtains the click data through different database narrow tables and performs attribution processing; The first database includes a plurality of first sub-databases corresponding to the time units; each of the first sub-databases includes a plurality of first database narrow tables; According to the attribution dimension data, the timestamp information, the number of first databases and the number of first database narrow tables included in each of the first databases, a first target database narrow table corresponding to the attribution dimension data is determined; and according to the data identifier, the timestamp information, the number of second databases and the number of second database narrow tables included in each of the second databases, a second target database narrow table corresponding to the click information is determined; According to the attribution dimension data, the timestamp information, the number of first databases and the number of first database narrow tables included in each of the first databases, a first target database narrow table corresponding to the attribution dimension data is determined; and according to the data identifier, the timestamp information, the number of second databases and the number of second database narrow tables included in each of the second databases, a second target database narrow table corresponding to the click information is determined; The second database includes a plurality of second sub-databases corresponding to the time units; each of the second sub-databases includes a plurality of second database narrow tables; ​ ​ ​ The second target database narrow table corresponding to the click information is determined according to the data identifier, the timestamp information, the number of second databases, and the number of second database narrow tables included in each of the second databases. The second target database corresponding to the click information is determined according to the data identifier and the number of second databases. The second target time unit corresponding to the timestamp information is determined, and the second sub-database corresponding to the second target time unit is determined as the second target sub-database corresponding to the click information. The second target database narrow table in the second target sub-database corresponding to the click information is determined according to the data identifier and the number of second database narrow tables included in the second target sub-database.

7. A storage medium, characterized by The storage medium is used to store a computer program, which is executed by a processor to implement the following processes: Obtain click data corresponding to advertising data, and generate a data identifier corresponding to the click data; According to the business attribute corresponding to the click data, the click data is split into attribution dimension data and click information; the business attribute includes user identifier information and / or device identifier information corresponding to the click data; the click information includes at least one of click content, timestamp information, and the device identifier information; According to the data volume corresponding to the historical click data and the number of dimensions of the historical attribution dimension, the number of first databases and the number of second databases are determined; According to the attribution dimension data, the timestamp information, the number of first databases, and the number of first database narrow tables included in each of the first databases, the first target database narrow table corresponding to the attribution dimension data is determined; and according to the data identifier, the timestamp information, the number of second databases, and the number of second database narrow tables included in each of the second databases, the second target database narrow table corresponding to the click information is determined; Each database narrow table is created based on different time units; The attribution dimension data and the data identifier are associated and stored in the first target database narrow table, and the click information and the data identifier are associated and stored in the second target database narrow table, so that an advertising attribution party obtains the click data through different database narrow tables and performs attribution processing; The first database includes a plurality of first sub-databases corresponding to the time units; each of the first sub-databases includes a plurality of first database narrow tables; The first target database narrow table corresponding to the attribution dimension data is determined according to the attribution dimension data, the timestamp information, the number of first databases, and the number of first database narrow tables included in each of the first databases, including: According to the attribution dimension data, the attribution dimension representation value corresponding to the attribution dimension data is determined; According to the attribution dimension representation value and the number of first databases, the first target database corresponding to the attribution dimension data is determined; determining a first target time unit corresponding to the timestamp information, and determining the first target sub-database corresponding to the first target time unit as a first target sub-database corresponding to the attribution dimension data; determining the first target database narrow table in the first target sub-database corresponding to the attribution dimension data according to the attribution dimension representation value and the number of the first database narrow tables included in the first target sub-database; the second database includes a plurality of second sub-databases corresponding to the time units; each of the second sub-databases includes a plurality of second database narrow tables; determining a second target database narrow table corresponding to the click information according to the data identifier, the timestamp information, the number of the second databases, and the number of the second database narrow tables included in each of the second databases, comprises: determining a second target database corresponding to the click information according to the data identifier and the number of the second databases; determining a second target time unit corresponding to the timestamp information, and determining the second target sub-database corresponding to the second target time unit as a second target sub-database corresponding to the click information; determining the second target database narrow table corresponding to the click information in the second target sub-database according to the data identifier and the number of the second database narrow tables included in the second target sub-database.

Images