Object data query method and device, computer equipment and storage medium

Abstract

The invention relates to an object data query method and device, computer equipment, a storage medium and a computer program product. The method can be applied to various scenes such as cloud technology, artificial intelligence, intelligent traffic and auxiliary driving. In response to the object data query request, determining a query condition indicated by the object data query request; when the query condition contains the object attribute information, generating an object attribute array query instruction suitable for a data storage table in the database according to the object attribute information; in response to the object attribute array query instruction, querying an object attribute tag array containing at least one object attribute tag from an object attribute tag array field in the data storage table; wherein the object attribute tags in the object attribute tag array in the data storage table belong to the same object; and according to the object attribute tag array containing the at least one object attribute tag, obtaining a query result meeting the query condition. By adopting the method, the query efficiency of the labels of the objects can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of computer data processing, and particularly to an object data query method, apparatus, computer device, and storage medium. Background Art

[0002] With the continuous development of big data, various objects in production and life will generate various types of information. Objects can include products, users, devices, and so on. By summarizing and analyzing the information of the objects, "labels" with highly generalized object attribute information and some other types of information can be obtained for the objects. Using these labels, various attributes of the objects can be accurately understood.

[0003] However, with the increase in the number of objects, the problem of querying object labels has gradually emerged. During querying, due to the large amount of label data, the query speed will be severely affected. Therefore, how to improve the efficiency of querying object labels has become an urgent problem to be solved. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide an object data query method, apparatus, computer device, and storage medium that can improve the query efficiency of object labels.

[0005] In a first aspect, the present disclosure provides an object data query method. The method includes:

[0006] Responding to an object data query request, determining a query condition indicated by the object data query request;

[0007] When the query condition includes object attribute information, generating an object attribute array query instruction applicable to a data storage table in a database according to the object attribute information, where the object attribute array query instruction includes at least one object attribute label determined based on the object attribute information;

[0008] Responding to the object attribute array query instruction, querying an object attribute label array including at least one object attribute label from an object attribute label array field in the data storage table; where the object attribute labels in the object attribute label array in the data storage table belong to the same object;

[0009] Obtaining a query result that meets the query condition according to the object attribute label array including at least one object attribute label.

[0010] In a second aspect, the present disclosure further provides an object data query apparatus. The apparatus includes:

[0011] A query condition determination module, configured to respond to an object data query request and determine a query condition indicated by the object data query request;

[0012] A query instruction generation module, configured to generate an object attribute array query instruction applicable to a data storage table in a database according to object attribute information when the query condition includes the object attribute information, where the object attribute array query instruction includes at least one object attribute label determined based on the object attribute information;

[0013] A query module, configured to, in response to the object attribute array query instruction, query an object attribute label array including at least one object attribute label from an object attribute label array field in the data storage table; wherein, the object attribute labels in the object attribute label array in the data storage table belong to the same object;

[0014] A query result obtaining module, configured to obtain a query result that meets the query condition according to the object attribute label array including at least one object attribute label.

[0015] In a third aspect, the present disclosure further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps in any one of the above method embodiments are implemented.

[0016] In a fourth aspect, the present disclosure further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0017] In a fifth aspect, the present disclosure further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0018] In the above embodiments, in response to an object data query request, the query conditions indicated by the object data query request are determined. The query conditions include various information that needs to be queried. When the query conditions include target object attribute information, an object attribute array query instruction applicable to the data storage table in the database is generated according to the object attribute information, which can convert the object attribute information into a query instruction. Using this query instruction, the required information can be accurately retrieved, ensuring the accuracy of the query result. In response to the object attribute array query instruction, an object attribute label array containing at least one object attribute label is queried from the object attribute label array field in the data storage table. And the object attribute labels in the object attribute label array in the data storage table belong to the same object. Therefore, the object attribute labels of the same object are all stored in an array, avoiding different labels of the same object being distributed in different regions of the data table, and being able to reduce data redundancy. In addition, when querying object attribute labels using the object attribute array query instruction, since the object attribute labels of the same object have been classified into an object attribute label array and stored in a data storage table, when querying object attribute labels, it can avoid the problem that the object attribute labels of different objects are stored in different data tables, and during the query process, it is necessary to merge and compare the data in multiple tables at the same time, resulting in a decline in query performance and query speed. During the query process, it is only necessary to query in different object attribute label arrays, and there will be no problem of cross-query, which can reduce the query scope and thus improve the query speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present disclosure. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0020] Figure 1 It is a schematic flowchart of an object data query method in an embodiment;

[0021] Figure 2A It is a schematic diagram of the object attribute label array field in the data storage table in an embodiment;

[0022] Figure 2B It is a schematic diagram of the object attribute label array in the data storage table in an embodiment;

[0023] Figure 3A It is a schematic diagram of the object attribute label value array field in the data storage table in an embodiment;

[0024] Figure 3BSchematic diagram of the object attribute label array and the object attribute label value array in an embodiment;

[0025] Figure 4 Schematic diagram of the object identifier, the object attribute label array, and the object attribute label value array in an embodiment;

[0026] Figure 5 Schematic diagram of the time information, the object attribute label array, and the object attribute label value array in an embodiment;

[0027] Figure 6 Graph showing the relationship between the number of object attribute label values and the time information satisfying the time condition in an embodiment;

[0028] Figure 7 Graph showing the relationship between the time information and the quantity ratio under each time information in an embodiment;

[0029] Figure 8 Schematic diagram of the storage structure of Clickhouse in an embodiment;

[0030] Figure 9 Schematic flowchart of a method for querying object data in another embodiment;

[0031] Figure 10 Schematic diagram of another method for querying object data in another embodiment;

[0032] Figure 11 Schematic diagram of the query of two object attribute labels in another embodiment;

[0033] Figure 12 Schematic diagram of the variation relationship of male users over time and the variation relationship of the number of objects interested in catering over time in another embodiment;

[0034] Figure 13 Schematic diagram of the structure of an object data query device in an embodiment;

[0035] Figure 14 Schematic diagram of the internal structure of a computer device in an embodiment. Detailed implementation manners

[0036] In order to make the objectives, technical solutions, and advantages of the present disclosure clearer and more understandable, the present disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure and are not used to limit the present disclosure.

[0037] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or equipment that includes a series of steps or units need not be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or equipment.

[0038] In this article, the term "and / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Additionally, in this article, the character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0039] As mentioned in the background art, the current methods for label query are as follows: 1. Using impala + kudu as the query and storage engine to query data for labels. Since this method is usually applied in interactive queries, real-time analysis, and complex multi-table association queries, and impala is based on in-memory computing and has a high dependence on memory and does not support high-concurrency scenarios. Therefore, this method consumes a large amount of resources when querying labels. 2. Using wide tables and indexes for label query. Due to the design of wide tables, when there are many labels, this storage structure will cause storage waste and data redundancy. Additionally, since data from multiple data sources needs to be obtained, when splicing wide tables (which need to carry hundreds of millions or even tens of billions of data), the calculation time will increase significantly when multiple wide tables are joined, thereby reducing the query efficiency. 3. Storing the labels that describe user data in databases such as Hive, HBase, and Elasticsearch. Hive is suitable for batch data processing and supports SQL language, but its query performance is poor and it takes a long time when dealing with complex queries or multi-table associations. HBase is suitable for storing massive amounts of data and supports random read and write. Since the query language of HBase is relatively simple, usually single-row or range queries based on row keys, it is not very suitable for complex multi-table associations, aggregation analysis, and other complex query operations, and its ability to handle complex query operations is weak. Elasticsearch has poor performance in scenarios where a large amount of result data needs to be scanned and exported, as it needs to retrieve data across the entire cluster, and the larger the result set returned, the slower it is.

[0040] Therefore, the embodiments of the present disclosure provide an object data query method, device, computer device, storage medium, and computer program product, aiming to solve the above-mentioned problems.

[0041] In one embodiment, as Figure 1As shown, an object data query method is provided. In this embodiment, the method is exemplified by being applied to a terminal. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. The terminal includes but is not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, etc. The embodiments of the present disclosure can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, intelligent transportation, assisted driving, etc. In this embodiment, the method includes the following steps:

[0042] S102, in response to an object data query request, determine the query conditions indicated by the object data query request.

[0043] Among them, the object data query request may be a request for obtaining specific data. The object data query request may be a request input to the terminal in text form on the client side, or a request input to the terminal through a graphical user interface on the client side. The query conditions may be conditions including various information of the data to be queried, such as time conditions of the data to be queried, conditions of the type of the data to be queried, etc. The query conditions can be determined according to different object data query requests. In addition, when there are multiple different conditions in the query conditions, there may also be some relationships among the multiple different conditions. For example, the sequence relationship of the query order of different conditions in the query conditions, the query logic relationship of different conditions. The logical relationship may include: AND, OR, etc. According to different object data query requests, the query conditions are also different. In some embodiments of the present disclosure, the information and relationships included in the query conditions are not absolutely limited, and those skilled in the art can flexibly set the query conditions according to actual application requirements.

[0044] Specifically, in response to an object data query request, the query conditions indicated by the object data query request can be determined from the mapping relationship containing the object data query request and the query conditions according to the object data query request. It is also possible to parse the object data query request and obtain the determined query conditions from the parsed object data query request. It should be noted that there may be other types of ways to determine the query conditions according to the object data query request, not limited to the above ways.

[0045] S104, when the query conditions include object attribute information, generate an object attribute array query instruction applicable to the data storage table in the database according to the object attribute information, and the object attribute array query instruction includes at least one object attribute label determined based on the object attribute information.

[0046] Among them, the objects can include users, various production devices, various application programs, etc. The object attribute information can be one or more pieces of information at various levels of the object. Different objects have different object attribute information. If the object is a user, the object attribute information can be the user's age, height, graduation school, hobbies, etc. If the object is various production devices, the object attribute information can be the production time of the production device, the production location of the device, the number of times the device has malfunctioned, etc. If the object is various application programs, the object attribute information can be the online time of the application program, the number of updates of the application program, the storage size of the installation package of the application program, etc. The database can be a columnar storage database in some embodiments of the present disclosure. For example, it can be a Clickhouse database. The data storage table can be a data table for storing the data to be queried. The object attribute array query instruction can be an instruction for querying the data in the array. The object attribute label can be regarded as an index in some embodiments of the present disclosure. The object attribute label can be determined based on the object attribute information. For example, if the object attribute information is age, the object attribute label can be the index corresponding to age.

[0047] Specifically, when the query condition includes object attribute information, since the object attribute information cannot be directly used as a query statement for querying. At least one object attribute label to be queried can be determined according to the object attribute information first. Then, an object attribute array query instruction applicable to the data storage table in the database can be generated based on the object attribute label and the object attribute information.

[0048] S106, in response to the object attribute array query instruction, query an object attribute label array containing the at least one object attribute label from the object attribute label array field in the data storage table; wherein, the object attribute labels in the object attribute label array in the data storage table belong to the same object. Figure 2AAs shown, it is a schematic diagram of the object attribute label array field in the data storage table in the embodiments of the present disclosure. The data storage table may include an object attribute label array field. Under the object attribute label array field, multiple object attribute label arrays may be included, such as object attribute label array A, object attribute label array B, and object attribute label array C. And each different object attribute label array belongs to the same object. For example, object attribute label array A belongs to object A, object attribute label array B belongs to object B, and object attribute label array C belongs to object C. In this way, when storing the object attribute labels of different objects, the object attribute labels of the same object can be placed in an object attribute label array, which can reduce data redundancy. Furthermore, placing the object attribute labels of the same object in an object attribute label array and in a data storage table can avoid the problem that when querying object attribute labels, since the object attribute labels corresponding to different objects are stored in different data tables, it is necessary to merge and compare the data in multiple tables simultaneously during the query process, resulting in a decrease in query performance and query speed.

[0049] Specifically, in response to the object attribute array query instruction, the object array query instruction can be used to query from under the object attribute label array field of the data storage table, query all the object attribute label arrays under the object attribute label array field, and then query the object attribute label array containing at least one object attribute label determined by the object attribute information.

[0050] In some exemplary embodiments, the object attribute array query instruction may be an instruction corresponding to the array_contain function or an instruction corresponding to the has All function. As Figure 2B shown, when the object attribute label to be queried is tag1, the instruction corresponding to the array_contain function can be used to query the array containing tag1 under the Tag_list field, and the arrays [tag1, tag2] and [tag1, tag3] can be obtained. When the object attribute labels to be queried are tag1 and tag2, the instruction corresponding to the has All function can be directly used to query the array containing both tag1 and tag2 under the Tag_list field, and the array [tag1, tag2] can be obtained. It is also possible to use the instruction corresponding to the array_contain function to query the array containing tag1 under the Tag_list field, obtain the arrays [tag1, tag2] and [tag1, tag3], and then use the instruction corresponding to the array_contain function to query the array containing tag2 in the arrays [tag1, tag2] and [tag1, tag3] to obtain the array [tag1, tag2].

[0051] S108. Obtain a query result that meets the query condition according to the object attribute label array including the at least one object attribute label.

[0052] Specifically, after obtaining the object attribute label array containing at least one object attribute label, the query result that meets the query condition can be directly obtained according to the object attribute label array containing at least one object attribute label.

[0053] In the above object data query method, in response to an object data query request, determine the query condition indicated by the object data query request. The query condition includes various information that needs to be queried. When the query condition contains target object attribute information, according to the object attribute information, generate an object attribute array query instruction applicable to the data storage table in the database, which can convert the object attribute information into a query instruction. Using this query instruction can accurately retrieve the required information and ensure the accuracy of the query result. In response to the object attribute array query instruction, query the object attribute label array containing at least one object attribute label from the object attribute label array field in the data storage table. And the object attribute labels in the object attribute label array in the data storage table belong to the same object. Therefore, the object attribute labels of the same object are all stored in one array, avoiding the different labels of the same object being distributed in different areas of the data table, and being able to reduce data redundancy. In addition, when querying object attribute labels using the object attribute array query instruction, since the object attribute labels of the same object have been classified into an object attribute label array and stored in one data storage table, when querying object attribute labels, it can avoid the problem that the object attribute labels of different objects are stored in different data tables, and during the query process, it is necessary to merge and compare the data in multiple tables at the same time, resulting in a decrease in query performance and query speed. During the query process, only queries need to be performed in different object attribute label arrays, and there will be no problem of cross-query, which can reduce the query scope and thus improve the query speed.

[0054] In one embodiment, when the number of object attribute labels determined based on the object attribute information is at least two, the object attribute array query instruction includes a query array, and the query array includes at least two object attribute labels determined based on the object attribute information.

[0055] The step of, in response to the object attribute array query instruction, querying the object attribute label array containing the at least one object attribute label from the object attribute label array field in the data storage table includes:

[0056] In response to the object attribute array query instruction, obtain the query array included in the object attribute array query instruction.

[0057] Determine the at least two object property tags included in the query array.

[0058] Under the object property tag array field in the data storage table, query the object property tag array that contains the at least two object property tags.

[0059] Specifically, when the number of object property tags is at least two, if the object property array query instruction is the instruction corresponding to the array_contain function at this time, it is necessary to query each object property tag in turn, which will reduce the query speed. Therefore, the query array can be determined based on the at least two object property tags. For example, the at least two object property tags will be combined into the query array. At this time, the generated object property array query instruction will contain this query array. In response to the object property array query instruction, obtain the query array included in the object property array query instruction. Then determine the at least two object property tags included in the query array, and directly query the object property tag array that contains the at least two object property tags under the object property tag array field in the data storage table.

[0060] In some exemplary embodiments, for example, when the object property tag array field is tag, the number of object property tags is 2, and they are tag1 and tag2, tag1 and tag2 can be combined into the query array [tag1, tag2], and the object property array query instruction can be hasAll(tag, [tag1, tag2]). Use this instruction to check whether the array under the tag field contains both the object property tags tag1 and tag2 at the same time. If the array under the tag field contains both tag1 and tag2 at the same time, this instruction will return 1 (true), otherwise this instruction will return 0 (false). According to the return result of each array under the tag field, it can be queried whether both the object property tags tag1 and tag2 are included at the same time.

[0061] In this embodiment, when the number of object property tags determined based on the object property information is at least two, the object property array query instruction contains the query array. When querying, it is only necessary to determine whether all of the at least two object property tags in the query array are included in the object property tag array. And when querying, it is only necessary to locate the object property tag array under the object property tag array field, without scanning the entire data table, which can further improve the query speed.

[0062] In one embodiment, obtaining a query result that meets the query condition according to the object property tag array that contains the at least one object property tag includes:

[0063] According to the object attribute label array including the at least one object attribute label, query an object attribute label value array that matches the object attribute label array including the at least one object attribute label from the object attribute label value array field in the data storage table.

[0064] Specifically, in addition to including an object attribute label array field, the data storage table may further include an object attribute label value array field corresponding to the object attribute label array field. After querying the object attribute label array including at least one object attribute label, it is possible to query under the object attribute label value array field according to the object attribute label array including at least one object attribute label to obtain an object attribute label value array that matches the object attribute label array of the at least one object attribute label.

[0065] In some exemplary embodiments, for example Figure 3A as shown, after obtaining the object attribute label array B after querying, it is possible to search for an object attribute label value array B1 that matches the object attribute label array B under the object attribute label value array field. Generally, the matching object attribute label array and object attribute label value array are located in the same row in the data storage table, which can improve the query efficiency.

[0066] According to the positions of the at least one object attribute label in the object attribute label array, determine object attribute label values that respectively match the at least one object attribute label in the object attribute label value array.

[0067] Among them, the object attribute label value may be a data value corresponding to the object label. For example, if the object attribute label is age, the object attribute label value may be a numerical value corresponding to the age. If the object attribute label is the usage times of the device, the object attribute label value may be a numerical value corresponding to the usage times.

[0068] Specifically, generally, the position of the object attribute label in the object attribute label array corresponds to the position of the object attribute label value in the object attribute label value array. Therefore, it is possible to determine object attribute label values that respectively match the at least one object attribute label in the object attribute label value array according to the positions of the at least one object attribute label in the object attribute label array.

[0069] In some exemplary embodiments, taking Figure 3B as an example, in the object attribute label array [tag1, tag2], tag1 is in the first position, then in the object attribute label value array [1, 0.1], the first value 1 is the object attribute label value corresponding to tag1. The second value 0.1 is the object attribute label value corresponding to tag2.

[0070] Obtain a query result that meets the query conditions according to the object attribute label array including the at least one object attribute label and the object attribute label values respectively matching the at least one object attribute label.

[0071] Specifically, according to the queried object attribute label array including at least one object attribute label and the queried object attribute label values respectively matching the at least one object attribute label, a query result that meets the query conditions can be obtained.

[0072] In a specific application scenario, for example, in the scenario of needing to obtain the number of times each function is used in assisted driving, taking the statistics of the number of times the adaptive cruise control function is used as an example for illustration. The object attribute label can be determined according to the adaptive cruise control function, so as to generate an object attribute array query instruction. Using this instruction, an object attribute label array containing the object attribute label of the adaptive cruise control function can be queried. Then, an object attribute label value array corresponding to the object attribute label array can be found. Then, according to the position of the object attribute label of the adaptive cruise control function in the object attribute label array, the label value of the object attribute label of the adaptive cruise control function can be determined. Thus, the number of times the adaptive cruise control function is used can be obtained according to the label value.

[0073] In this embodiment, by using the positions of the at least one object attribute label in the object attribute label array, the object attribute label values respectively matching the at least one object attribute label can be accurately matched. It only needs to be matched under the object attribute label value array field, which can improve the efficiency of label value query.

[0074] In one embodiment, the obtaining a query result that meets the query conditions according to the object attribute label array including the at least one object attribute label and the object attribute label values respectively matching the at least one object attribute label includes:

[0075] According to the object attribute label array including the at least one object attribute label, query the object identifier matching the object attribute label array including the at least one object attribute label from the object identifier field in the data storage table;

[0076] According to the queried object identifier, the object attribute label array including the at least one object attribute label, and the object attribute label value, obtain a query result that meets the query conditions.

[0077] Among them, the object identifier in the object identifier field can be unique data used to represent each object. The object identifier can be in the form of numbers, letters, symbols, etc.

[0078] Specifically, the data storage table may further include an object identification field. Multiple object identifications may be included under the object identification field. After querying and obtaining an object attribute label value array that matches an object attribute label array of at least one object attribute label, it is also possible to perform a query under the object identification field based on the object attribute label array containing at least one object attribute label to obtain an object identification that matches the object attribute label array containing at least one object attribute label. Then, based on the queried object attribute label array containing at least one object attribute label, the queried object attribute label value, and the object identification, a query result that meets the query conditions is obtained. It should be noted that, usually, the matching object attribute label array, object attribute label value array, and object identification are located in the same row of the data storage table, so that the query speed can also be improved when querying the object identification.

[0079] In some exemplary embodiments, such as Figure 4 shown, after obtaining the object attribute label array [tag1, tag2] containing the object attribute label tag2, usually, the object identification in the corresponding row of the data storage table can be determined according to the object attribute label array [tag1, tag2]. In FIG. 3, the object attribute label array [tag1, tag2] is located in the first row, so the object identification can be searched for in the first row to obtain the object identification u1.

[0080] In a specific application scenario, for example, in the scenario of driving a car, if you want to obtain the number of times that a certain type of car uses the cruise control function in autonomous driving, you can first query the object attribute label array containing the object attribute label corresponding to the cruise control function in the data storage table, and then use the object attribute label array to query the corresponding object identification and the value of the object attribute label containing the cruise control function. Usually, there is a one-to-one correspondence between the object identification and the car. Then, the object identification can be used to determine each car, and the value of the object attribute label containing the cruise control function can be used to determine the number of times the cruise control function is used.

[0081] In this embodiment, by using the object attribute label array, the object identification that matches the object attribute label array of at least one object can be accurately matched from the data storage table. Only matching needs to be performed under the object identification field, which can improve the efficiency of object identification query. In addition, since the object identification can be queried, it can adapt to queries in different scenarios.

[0082] In one embodiment, the obtaining of the query result that meets the query conditions according to the object attribute label array containing at least one object attribute label includes:

[0083] According to the object attribute label array including the at least one object attribute label, query, from the object identification field in the data storage table, the object identification that matches the object attribute label array including the at least one object attribute label;

[0084] Obtain a query result that meets the query condition according to the matched object identification and the object attribute label array including the at least one object attribute label.

[0085] Specifically, according to different query scenarios, there may also be a situation where it is necessary to only query the object identification and the object attribute label array associated with the object attribute label array. Therefore, after querying the object attribute label array including at least one object attribute label, it is also possible to query, from the object identification field in the data storage table, according to the object attribute label array including at least one object attribute label, the object identification in the same row as the object attribute label array including at least one object attribute label. The object identification in the same row is the object identification that matches the object attribute label array. Then, obtain a query result that meets the query condition according to the matched object identification and the object attribute label array including at least one object attribute label.

[0086] In a specific application scenario, for example, it is necessary to query cars with the automatic parking function in autonomous driving. First, the object attribute label corresponding to the automatic parking function can be determined, then the object attribute label array containing the object attribute label corresponding to the automatic parking function can be found, and then the object identification that matches can be found by using the object attribute label array containing the object attribute label corresponding to the automatic parking function. According to this object identification, the cars with the automatic parking function in autonomous driving can be determined.

[0087] In this embodiment, when querying an object, since the object identification can be quickly queried by using the object attribute label array including at least one object attribute label, and then the object to be queried can be determined, the query speed of querying the object can be improved.

[0088] In one embodiment, the query condition further includes a time condition. The obtaining of the query result that meets the query condition according to the object attribute label array including the at least one object attribute label and the object attribute label values respectively matching the at least one object attribute label includes:

[0089] From the object attribute label values respectively matching the at least one object attribute label, filter out the object attribute label values that match the time information meeting the time condition according to the time condition.

[0090] Among them, the time condition can usually be a condition for querying information between a certain time period or only for a certain time. For example, the time condition can be a condition for querying information between December 1, 2023 and December 20, 2023. The time condition can also be a condition for only querying the information on December 1, 2023. The time condition can also be a condition for querying the information at 12:00 on December 1, 2023. According to different scenarios, the time condition is also different. In some embodiments of the present disclosure, the time condition is not specifically limited.

[0091] Specifically, as mentioned in the above embodiments, the object attribute tag values that match the object attribute tags have been queried. Generally, each object attribute tag value will have corresponding time information. Therefore, it is also possible to screen the time information that matches the object attribute tag values again from the queried object attribute tag values according to the time condition. Screen out the time information that meets the time condition, and then further determine the object tag values that match the time information of the screened time condition.

[0092] From the data storage table, find the object attribute tag array that matches the screened object attribute tag values;

[0093] According to the screened object attribute tag values and the found object attribute tag array, obtain the query result that meets the query condition.

[0094] Specifically, after screening out the object tag values, since the object tag values usually match the object attribute tag array in the same row. Therefore, it is also possible to find the object attribute tag array that matches the screened object tag values from the data storage table. Finally, according to the screened object attribute tag values and the found object attribute tag array, obtain the query result that meets the query condition.

[0095] In another implementation manner of this embodiment, it is also possible to first screen out the object attribute tag array that matches the time information that meets the time condition from the found object attribute tag array according to the time condition. Then, according to the object attribute tag array that matches the time information that meets the time condition and the screened object attribute tag values, further perform secondary screening on the screened object attribute tag values to obtain the tag values after secondary screening. Then, use the tag values after secondary screening and the object attribute tag array that matches the time information that meets the time condition to obtain the query result that meets the query condition.

[0096] In some exemplary embodiments, for example, in the query condition, the time condition is to query information for October 10, 2023. The object attribute tags determined by the object attribute information are tag1 and tag3. Then, the object attribute tag values of the object attribute tag array containing at least one object attribute tag can be, at different times, the tag1 tag value for 20231010, the tag1 and tag3 tag values respectively for 20231011, the tag3 tag value for 20231012, and the tag1 and tag3 tag values for 20231013, a total of four groups. Then, use 20231010 to filter the above tag values, and only retain the tag1 tag value for 20231010. Then, use the tag1 tag value corresponding to 20231010 to find the corresponding object attribute tag array [tag1, tag2]. Then, according to the object attribute tag array [tag1, tag2] and the tag1 tag value corresponding to 20231010, obtain the query result that meets the query condition.

[0097] As another query method, if the object attribute tag array containing at least one object attribute tag that is queried is an array containing tag1 and tag2, then two groups can be obtained: [tag1, tag2] and [tag1, tag2, tag4]. Among them, the time information corresponding to [tag1, tag2] is October 10, 2023, and the time information corresponding to [tag1, tag2, tag4] is October 13, 2023. If the time condition in the query condition still needs to query information for October 10, 2023, then after filtering using the time condition, [tag1, tag2] can be obtained. Then, use the [tag1, tag2] obtained from the query and the tag1 and tag2 tag values in the [tag1, tag2] obtained from the query to obtain the query result that meets the query condition.

[0098] In this embodiment, by using the time condition, further filtering can be performed, and thus various information that meets the time condition can be queried, and information that meets different time conditions can be queried.

[0099] In one embodiment, the query condition can also be used to filter only the object attribute tag array. Therefore, the query condition further includes a time condition. Obtaining a query result that meets the query condition according to the object attribute tag array containing the at least one object attribute tag includes:

[0100] From the object attribute tag array containing the at least one object attribute tag, filter out the object attribute tag array that matches the time information that meets the time condition according to the time condition;

[0101] Obtain a query result that meets the described query conditions according to the array of object attribute tags filtered out.

[0102] Regarding the specific implementation manner of querying the array of object attribute tags using time conditions in this embodiment, reference may be made to the above embodiment, and details will not be repeated here.

[0103] In a specific application scenario, for example, continuing with the autonomous driving scenario as an example, if it is necessary to query whether the automatic parking function was used within a certain time period, corresponding time conditions can be set. If the automatic parking function has been used, the corresponding object attribute tags of the automatic parking function will be stored in the corresponding data storage table. Therefore, according to the object attribute tags corresponding to the automatic parking function, find the array of object attribute tags containing the object attribute tags corresponding to the automatic parking function, and then use the time conditions to further filter the found array of object attribute tags containing the object attribute tags corresponding to the automatic parking function to determine whether the automatic parking function was used at a certain time. If, after further filtering the found array of object attribute tags containing the object attribute tags corresponding to the automatic parking function by the time conditions, an array of object attribute tags containing the object attribute tags corresponding to the automatic parking function under a certain time information is obtained, it can be determined that the automatic parking function was used at a certain time. If no result is finally obtained, it can be determined that the automatic parking function was not used at a certain time.

[0104] In this embodiment, by further filtering the array of object attribute tags that can be further queried using time conditions, different query scenarios can be satisfied.

[0105] In an embodiment, the query conditions further include at least one object attribute value condition corresponding to the object attribute information. The obtaining of the query result that meets the query conditions according to the filtered object attribute tag values and the found array of object attribute tags includes:

[0106] Determine at least one object attribute tag value condition according to the at least one object attribute value condition.

[0107] Among them, the object attribute value condition can usually be a condition for filtering object attribute information. For example, if the object attribute information is the age of a user, the object attribute value condition can be the condition that the age is greater than forty years old. If the object attribute information is the number of times a device has malfunctioned, the object attribute value condition can be the condition that the number of times the device has malfunctioned is greater than five times. According to different object attribute information and different application scenarios, the object attribute value conditions are also different. The object attribute label value condition can be a condition for filtering the object attribute label value. For example, if the object attribute label tag1 is age, and the object attribute value condition is that the age is greater than forty years old, then the object attribute label value condition is that the label value of tag1 > 40.

[0108] Specifically, since the object attribute value condition is usually a condition for filtering object attribute information, it processes the object attribute information and cannot directly process the object attribute label value. Therefore, the object attribute value condition can be converted into an object attribute label value condition that can process the object attribute label value.

[0109] From the filtered object attribute label values, determine the object attribute label values that meet the object attribute label value condition.

[0110] Specifically, as mentioned in the above embodiment, the object attribute label value usually is a label value that meets the time condition. Therefore, after determining at least one object attribute label value condition, the object attribute label values filtered in the above embodiment can be further filtered according to the object attribute label value condition to determine the object attribute label values that meet the object attribute label value condition.

[0111] From the data storage table, find the object attribute label array that matches the object attribute label value that meets the object attribute label value condition;

[0112] According to the object attribute label value that meets the object attribute label value condition and the found object attribute label array, obtain the query result that meets the query condition.

[0113] Specifically, after determining the object attribute label value that meets the object attribute label value condition, the object attribute label array that matches the object attribute label value that meets the object attribute label value condition can also be found from the data storage table. Then, according to the found object attribute label array and the object attribute label value that meets the object attribute label value condition, the query result that meets the query condition is obtained. It should be noted that the query condition in this embodiment includes object attribute information, at least one object attribute value condition corresponding to the object attribute information, and a time condition.

[0114] In a specific application scenario, taking the autonomous driving scenario as an example for further illustration. For example, it is necessary to find the results where the time condition is from November 11, 2023 to November 30, 2023 (the time condition can be from 20231111 to 20231130), the automatic parking function is used, and the number of times the automatic parking function is used is greater than 10 times. The object attribute tag values that meet the time condition have been queried above. Therefore, the object attribute tag value condition can be determined according to the object attribute value condition that the number of times the automatic parking function is greater than 10 times. If the object attribute tag corresponding to the automatic parking function is tag1, then the object attribute tag value condition can be tag1 > 10. The condition of tag1 > 10 can be found in the object attribute tag values that meet the time condition, and then the object attribute tag values that meet the object attribute tag value condition can be determined. Then, from the data storage table, find the object attribute tag array that matches the object attribute tag value that meets the object attribute tag value condition.

[0115] In one embodiment, when the query condition includes object attribute information and at least one object attribute value condition corresponding to the object attribute information. Obtaining the query result that meets the query condition according to the object attribute tag array including the at least one object attribute tag and the object attribute tag values respectively matching the at least one object attribute tag includes:

[0116] Determine at least one object attribute tag value condition according to the at least one object attribute value condition.

[0117] Determine the object attribute tag values that meet the object attribute tag value condition from the object attribute tag values respectively matching the at least one object attribute tag.

[0118] From the data storage table, find the object attribute tag array that matches the object attribute tag value that meets the object attribute tag value condition;

[0119] Obtain the query result that meets the query condition according to the object attribute tag value that meets the object attribute tag value condition and the found object attribute tag array.

[0120] For the specific implementation manner of querying using the object attribute value condition here, reference can be made to the above embodiment. It will not be repeated here. The difference between this embodiment and the above embodiment is that the query condition in the above embodiment also includes a time condition. The query condition in this embodiment does not include a time condition.

[0121] In this embodiment, by using the object attribute tag value condition, the results that meet the object attribute tag value condition can be further found, which can meet various query conditions, so as to query more accurate results.

[0122] In one embodiment, the method further includes:

[0123] Obtain the object attribute tag value that meets the object attribute tag value condition.

[0124] For each time information that meets the time condition, respectively, under the targeted time information, count the number of object attribute tag values that meet the object attribute tag value condition.

[0125] According to the number of object attribute tag values respectively counted for each time information that meets the time condition, construct a relationship graph between the counted number of object attribute tag values and the time information that meets the time condition.

[0126] Specifically, as mentioned in the above embodiment, after the query condition meets the time condition and at least one object attribute value condition, a query result that meets the query condition is obtained. Since the time condition is used, after obtaining the query result, the change trend of the query result under the time condition can also be counted. Therefore, the object attribute tag value in the above-obtained query result can be obtained first. It should be noted that since the change trend needs to be counted in this embodiment, the time condition in the embodiments of the present disclosure usually consists of multiple consecutive time information. For example, if the time condition is the time period from October 10, 2023 to October 15, 2023, then for each time information that meets the time condition, such as October 10, 2023, October 11, 2023, October 12, 2023, October 13, 2023, October 14, 2023, and October 15, 2023, the number of object attribute tag values in the above query result corresponding to each time information can be obtained accordingly. Then, for each time information that meets the time condition and the number of object attribute tag values in the query result under each time information, a relationship graph between the counted number of object attribute tag values and the time information that meets the time condition is constructed.

[0127] In a specific application scenario, for example, continuing with the autonomous driving scenario as an example, it is necessary to count the change trend of the use of the automatic parking function between the 10th and the 15th. When obtaining each time information that meets the time condition between the 10th and the 15th, the number of object attribute tag values corresponding to the automatic parking function is obtained. The number of times the automatic parking function is used daily between the 10th and the 15th can be based on. Then, according to the number of times the automatic parking function is used daily between the 10th and the 15th and the daily time information between the 10th and the 15th, it can be obtained as Figure 6 shown, the change trend of the automatic parking function between the 10th and the 15th.

[0128] In this embodiment, by constructing a relationship graph between the number of object attribute tag values and the time information that meets the time condition, the change trends of various data within the time condition can be statistically analyzed, and different change trends can be better targeted. When the change trend of the number of object attribute tag values in the relationship graph gradually decreases over time, it can be determined that there is a problem with the object attribute tag corresponding to the object attribute tag value. Therefore, various aspects of the product can be optimized specifically according to the object attribute tag whose number of object attribute tag values has decreased.

[0129] In one embodiment, the method further includes:

[0130] Obtain the object attribute tag values that meet the object attribute tag value condition and the array of object attribute tags that match the filtered object attribute tag values included in the query result.

[0131] Under each piece of the time information, calculate the quantity ratio between the number of the obtained object attribute tag values and the number of the obtained array of object attribute tags.

[0132] Construct a relationship graph between the time information and the quantity ratio under each piece of the time information according to the time information and the quantity ratio under each piece of the time information.

[0133] Specifically, it is also possible to obtain the proportion of the object attribute tag values that meet the object attribute tag value condition within the time condition among all the queried ones containing the object attribute tag value. For example, it is necessary to calculate the proportion of male users among all users. Or it is necessary to calculate the proportion of devices with more than three failure times among the devices that have failed. At this time, the object attribute tag values in the query result need to be obtained. The object attribute tag value usually meets the time condition and the object attribute tag value condition. It is also possible to obtain the array of object attribute tags that match the object attribute tag values that match the time information that meets the time condition. Under each piece of the time information that meets the time condition, calculate the quantity ratio between the number of object attribute tag values in the obtained query result and the number of the array of object attribute tags that match the object attribute tag values that match the time information that meets the time condition. Then, according to each piece of the time information of the time condition and the quantity ratio under each piece of the time information, construct a relationship graph between the time information and the quantity ratio under each piece of the time information.

[0134] In another embodiment, it is also possible to obtain the number of object attribute tag values that match the time information meeting the time condition. Under each time information meeting the time condition, calculate the quantity ratio between the number of object attribute tag values in the obtained query result and the number of object attribute tag values that match the time information meeting the time condition. Then, based on each time information of the time condition and the quantity ratio under each time information, construct a relationship graph between the time information and the quantity ratio under each time information.

[0135] In a specific application scenario, continue to take the application in the autonomous driving scenario as an example for illustration. For example, it is necessary to count the proportion trend of the number of times the automatic parking function is used more than 5 times under a certain time condition. When obtaining each time information that meets the time condition between the 10th and the 15th, the number of object attribute tag values corresponding to the automatic parking function that are greater than 5 is obtained. It is also possible to obtain the number of object attribute tag values corresponding to the automatic parking function or the number of object attribute tag arrays corresponding to the automatic parking function between the 10th and the 15th. Then calculate the ratio between the number of object attribute tag values corresponding to the automatic parking function that are greater than 5 and the number of object attribute tag arrays corresponding to the automatic parking function, and construct a relationship graph between the time information and the quantity ratio under each time information as shown in Figure 6 the figure.

[0136] In this embodiment, by constructing a relationship graph between the time information and the quantity ratio under each time information, it is possible to reflect the proportion trend of different types of object attribute tag values that meet the query conditions, so that the object attribute tags corresponding to the object attribute tag values can be better analyzed using the proportion trend, and the products corresponding to the object attribute tags can be optimized and improved targeted to improve the usage rate of the products.

[0137] In one embodiment, the object attribute tag array in the data storage table is obtained by including the following method:

[0138] Obtain the object information of at least one object stored in multiple data sources.

[0139] Among them, multiple data sources can usually be storage sources for various attribute information of the object. For example, the data storage data source generated when the object operates the application program. Or the data source for storing various object information when the object registers, etc. Usually, the object information of the object is obtained after being authorized by the object or other third parties with permissions.

[0140] Perform data cleaning on the object information and delete the non-compliant data in the object information.

[0141] Among them, data cleaning refers to the process of processing and screening object information to obtain high-quality object information that can be used for analysis. The goal of data cleaning is to eliminate noise, errors, and redundancy in object information, as well as fill in missing values to make the object information more complete and accurate.

[0142] Specifically, SQL statements can be used to clean the data of object information. First, preprocess the dirty data that does not conform to common sense. For example, there is information in the object information such as an age of 1000 years or an age of -1 year, and the equipment usage time is 100,000 hours, etc. These object information that obviously do not conform to the specifications are deleted.

[0143] Perform data analysis on the object information after data cleaning to determine the object attribute labels in the object information after data cleaning.

[0144] Specifically, after cleaning the data of the object information, data analysis can be performed on the object information after data cleaning. For example, extract keywords, key phrases, etc. from the object information, and then match them with some predefined attribute labels and the keywords, key phrases, etc. extracted from the object information to determine the object attribute labels in the object information. In addition, some trained neural network models can also be used to process the object information to determine the object attribute labels in the object information.

[0145] Merge the object attribute labels of the same object into an object attribute label array.

[0146] Specifically, usually, the object attribute labels in the object information of the same object can be merged into an object attribute label array.

[0147] In this embodiment, by cleaning the data of the object information, performing data analysis, and merging the object attribute labels of the same object into an object attribute label array and storing it in a data storage table, during subsequent queries, the object attribute labels of the same object are only stored in one object attribute label array, which can avoid cross-queries and thus improve the query speed.

[0148] In addition, the data storage table is usually stored in a columnar database. In some embodiments of the present disclosure, queries can be performed in the columnar database, and the columnar database can be a ClickHouse database. For example Figure 8As shown in the figure, since ClickHouse is a columnar database, each column has a separate data file (*.bin), and all values of this column are stored in a compressed format, which can reduce the storage space. The granule in the columnar database is a logical data block divided on each column and is also the smallest read unit in ClickHouse. Instead of reading individual rows each time from ClickHouse, the entire row group (granule) is always read (in a streaming and parallel manner). Usually, the size of the granule is determined by the configuration item, and the default value is 8192. In addition, since ClickHouse locates the specific granule based on the primary key and then reads all eligible granules at once, it can improve the concurrency and computing performance of the query, thereby increasing the query speed. Additionally, since the data storage table is stored in a columnar database, and the columnar database adopts a column storage structure, each column of data can be stored separately on the disk, which can improve the query performance and reduce the overhead of disk input / output and CPU computing, and reduce the resource consumption during querying.

[0149] In one embodiment, the method further includes:

[0150] Obtain the label quality and / or application degree of each object attribute label in the object attribute label array of the data storage table, where the label quality is determined based on the data related to the object attribute label and the proportion of the data, and the application degree is determined based on the operations related to the object attribute label;

[0151] Determine the useless object attribute labels of each object attribute label in the object attribute label array of the data storage table according to the label quality and / or the application degree of each object attribute label in the object attribute label array of the data storage table;

[0152] Delete the useless object attribute labels in the object attribute label array of the data storage table.

[0153] Among them, the label quality can usually be a kind of data for evaluating the quality of the object attribute label. The application degree can be a kind of data for evaluating the usage degree of the label. The data related to the object attribute label can be object information or the source of the object information. The proportion of the data can usually be the proportion of the object information. The operations related to the object attribute label can include: query operation, click operation, operation of applying for permissions, favorite operation, and so on. The useless object attribute label is a label without value.

[0154] Specifically, the data completeness of each object attribute label value can be determined according to the data source credibility to which the object attribute label value of each object attribute label in the data storage table belongs and the quality of the data in the data source. The data completeness may refer to the comprehensiveness and accuracy degree of the object attribute labels and object attribute label values included in the data storage package. The data source credibility and the quality of the data in the data source can be determined in advance according to different requirements. When the data source credibility to which a certain object attribute label value belongs is higher than the credibility threshold, and / or the quality index of the data in the data source is greater than the quality threshold, it can be determined that the data completeness of the object attribute label corresponding to the object attribute label value is relatively high.

[0155] The random sampling method can also be used to sample the object information to obtain the abstract object information, and the accuracy rate of each object attribute label value can be calculated according to the ratio between the same information in the abstract object information and each object attribute label value and the label value corresponding to the sampled object information. Generally, the amount of data in the object information is large. If the full amount is used for statistics, the amount of data is very large, which will increase the time-consuming for calculating the accuracy rate. Therefore, using the sampling method can shorten the time. Using the random sampling method can be consistent with the full amount result, and will not get distorted data due to the sampling method, and can ensure the accurate calculation of the accuracy rate. In addition, to ensure the accuracy of the result after sampling, too few object attribute labels have little practical significance in actual use. Therefore, a certain number, such as 10% of all object attribute labels, can be selected as the target data for sampling. After extracting the sampled object information, the corresponding object attribute labels of the sampled object information can be used to compare with each object attribute label value to determine the accuracy rate. For example, the object attribute labels corresponding to the sampled object information are tag1, tag2, tag3, and their corresponding label values are 1, 0.5, and 0.3. While the label values corresponding to tag1, tag2, tag3 in the data storage table are 1, 0.5, and 0.7, then the final accuracy rate is 2 / 3≈66.7%.

[0156] Determine the coverage rate of each object attribute label value according to the ratio between the number of each object attribute label value and the total number of the object attribute label values.

[0157] The coverage rate of the object property label value of each object property label can be calculated using the ratio between the number of object property label values of each object property label and the total number of object property label values. When the coverage rate is small, it can be determined that the availability rate of this object property label is low. For example, for all the data in a certain application, the object property labels are 100. However, since some objects cannot obtain the data of a certain object property label, there will be a probability of coverage rate. If there are 100 objects in this application, and 80 objects can obtain the A object property label, then the coverage rate is 80%. If only 10 objects can obtain the B object property label, then the coverage rate is 10%. Through the coverage rate, the health of the object property label can be determined. If the coverage rate of a certain object property label is less than 5%, it is determined that the availability of this object property label is not high. Then, according to at least one of the data completeness, accuracy rate, and coverage rate of the object property label value of each object property label, the label quality of each object property label in the object property label array in the data storage table can be determined.

[0158] In addition, the application degree of the object property label can be evaluated from three levels. At the construction level, according to the number of views, the number of favorites, or the number of permission applications required to add the object property label to the object property label array in the data storage table for each object property label in the object property label array in the data storage table, the attention degree of each object property label in the object property label array in the data storage table can be determined. At the query level, according to the number of queries of each object property label in the object property label array in the data storage table, the usage degree of each object property label in the object property label array in the data storage table can be determined. At the application level, the corresponding display data of each object property label in the object property label array in the data storage table can be generated, such as articles, posters, etc. According to the click-through rate of this display data, the click degree of each object property label in the object property label array in the data storage table can be determined. Then, according to at least one of the usage degree, attention degree, and click degree of each object property label in the object property label array in the data storage table, the application degree of each object property label in each object property label array in the data storage table can be determined.

[0159] Then, according to the application degree and / or the label quality, each object property label in each object property label array in the data storage table is screened, and the useless object property labels in the data storage table whose application degree and / or label quality do not meet the requirements are deleted.

[0160] In this embodiment, by using the label quality and / or application degree to delete the useless object attribute labels in the data storage table whose application degree and / or label quality do not meet the requirements, the accuracy and integrity of the object attribute labels can be ensured, and the valuable object attribute labels can be retained. In addition, during the data query and analysis process, removing the useless object attribute labels can more specifically query and analyze the useful object attribute labels, thereby reducing unnecessary analysis operations on the useless object attribute labels.

[0161] In a specific application scenario, the method of the present disclosure can be applied to an application scenario for querying various information of users. For further illustration of the object data query method in the embodiments of the present disclosure, as Figure 9 shown and Figure 10 shown, the object data query includes:

[0162] S902, obtain the object information of at least one object stored in multiple data sources. Perform data cleaning on the object information, delete the non-compliant data in the object information; perform data analysis on the object information after data cleaning to determine the object attribute labels in the object information after data cleaning; merge the object attribute labels of the same object into an object attribute label array and store them in a data storage table.

[0163] Specifically, the object information of multiple objects stored in different data sources can be obtained. Then, data cleaning and data analysis are performed on the object information to determine the object attribute labels in the object information. Then, the object attribute labels can be further classified, for example, into a first attribute, a second attribute, and a third attribute. The first attribute can be various basic attributes of the object, including: demographic attributes (such as gender, age, etc.), device attributes (such as the brand, model, age of the device used, etc.), and geographical attributes (such as place of residence, place of work, etc.). The second attribute can be the consumption attribute of the object, such as consumption frequency, consumption category, and business preference, etc. The third attribute can be various interest categories of the object, such as interest categories (such as swimming, running, fitness, etc.), preferred authors (such as preferring author A, author B, etc.), and long-term and short-term interests (such as the long-term interest is to pay attention to movie information, and the short-term interest is to pay attention to food information, etc.).

[0164] In order to improve the query speed of object attribute labels, it is necessary to build a data storage structure that is convenient for rapid query and analysis, and it is necessary to reconstruct the data structure. The most direct way is usually to build a label table structure ( Figure 10Iteration 1). Store the object property labels in one column and the object property label values in another column. When the value of an object property label is 0, it will not be stored. However, when querying the intersection between labels, such as querying objects that contain both tag1 and tag2, there will be a problem of long connection operation time between data tables. Therefore, iterative optimization can be continued to avoid the connection operation between large tables. The data structure can be further processed to obtain Iteration 2. Among them, Date is the time information field, User_id is the object identification field, and Tag_list is the object property label array field. If there is an object property label, it will be put into the object property label array. If not, it will not be put in, which will not cause storage waste. And the object property labels of the same object on the same day can be put into an object property label array, which can solve the query problem caused by label intersection. In addition, in order to further accurately determine the object property label values of each object property label in the object property label array, the method of Iteration 3 can be used, adopting the Clickhouse Nested structure, storing the object property label array in one column and the object property label value array corresponding to the object property label array in another column. And the corresponding object property label array and object property label value array are on the same row.

[0165] S904, in response to an object data query request, determine the query conditions indicated by the object data query request.

[0166] Specifically, in response to an object data query request, determine that multi-dimensional analysis of object property labels is required, and determine the query conditions indicated according to the object data query request.

[0167] S906, when the query conditions include object property information, generate an object property array query instruction applicable to the data storage table in the database according to the object property information. The object property array query instruction includes at least one object property label determined based on the object property information.

[0168] S908, in response to the object property array query instruction, query the object property label array containing at least one object property label from the object property label array field in the data storage table.

[0169] Specifically, when the query conditions include querying objects interested in catering, an object property array query instruction for querying the data storage table can be generated according to the conditions of being interested in catering. This object property array query instruction can query the object property labels corresponding to the catering category. Then use this object property array query instruction to query the object property label array containing the object property labels corresponding to the catering category.

[0170] As another implementation, the object property tag may also include the degree of interest. For example, when the query condition includes querying for objects with a high degree of interest in dining. Taking tag1 as an example, tag1 can be a high degree of interest in dining, and tag2 can be a low degree of interest in dining. Therefore, an object property tag array containing the tag1 tag (the object property tag with a high degree of interest in dining) can be found in the data storage table, so as to query the object property tag array containing the object property tag with a high degree of interest in dining.

[0171] S910, according to the object property tag array containing the at least one object property tag, query, from the object identifier field in the data storage table, the object identifier that matches the object property tag array containing the at least one object property tag.

[0172] Specifically, after obtaining the object property tag array corresponding to the dining category, the object identifier corresponding to the object property tag array of the object property tag corresponding to the dining category can continue to be queried under the object identifier field in the data storage table, so as to query the objects interested in dining. When the object property tag can also include the degree of interest, the objects with a high degree of interest in dining can be queried.

[0173] S912, according to the object property tag array containing the at least one object property tag, query, from the object property tag value array field in the data storage table, the object property tag value array that matches the object property tag array containing the at least one object property tag.

[0174] S914, according to the positions of the at least one object property tag in the object property tag array, determine, in the object property tag value array, the object property tag values that respectively match the at least one object property tag.

[0175] S916, according to the object property tag array containing the at least one object property tag, query, from the object identifier field in the data storage table, the object identifier that matches the object property tag array containing the at least one object property tag.

[0176] S918, according to the queried object identifier, the object property tag array containing the at least one object property tag, and the object property tag value, obtain a query result that meets the query condition.

[0177] Specifically, when the query condition contains an object for which it is necessary to query the interest in catering and the more accurate degree of interest in catering for each object, since iteration 2 can only query a relatively rough degree of interest, it is necessary to query in the data storage table of the Clickhouse Nested structure. After querying the object property label array containing the object property label corresponding to catering, the corresponding object identifier and object property label value array can be matched under the same row of object identifier fields and object property label value array fields. Then, according to the position of the object property label corresponding to catering in the object property label array, the object property label value of the object property label corresponding to catering is determined. The degree of interest in catering can be determined according to this label value. Generally, the larger the label value, the more interested in catering.

[0178] The above only mentions the query method using a single condition in the query condition. When the query condition also includes querying the gender of the object. For example, when the query condition includes querying males with a relatively high degree of interest in catering. As Figure 11 shown, there are two object property labels. One is tag_id = 'tag1' which refers to the object property label for which the gender needs to be queried, and the label value score = 1 indicates a male user. The other is tag_id = 'tag21' which refers to the object property label for which it is necessary to query the object property label containing catering. Taking score greater than 0.6 indicates a user with a very high degree of enthusiasm for catering. First, use hasall(goals.app, [tag1, tag21]) to directly find the object property label array of the object property label with the degree of interest in catering and the object property label with gender, and then use. tag1 = 1 and tag21 > 0.6 to filter the obtained object property label array, so as to obtain the object property label array corresponding to males with a relatively high degree of interest in catering, and then find the corresponding object identifier according to this object property label array, so as to find the male object with a relatively high degree of interest in catering.

[0179] In addition, when the query condition also includes a time condition, the time condition can be further used for filtering.

[0180] After filtering using the time condition, the variation relationship of various data with time can be determined. As Figure 12As shown, for example, if you want to determine the change relationship of male users over time, you can obtain the number of male users at each time within a certain time period, and then determine the change relationship of the number of male users over time based on the number of male users at each time within that time period. Another example, if you want to determine the change relationship of the number of objects interested in catering over time, you can obtain the number of objects interested in catering at each time within a certain time period, and then determine the change relationship of the number of objects interested in catering over time based on the number of objects interested in catering at each time within that time period.

[0181] Another example, if you want to determine the change relationship of the proportion of the number of male users in all users within a certain time period. You can first determine the number of all users at each time within a certain time period, then determine the number of male users at each time within a certain time period, divide the number of male users by the number of all users to obtain the proportion of male users at each time within a certain time period, and then determine the relationship graph of the proportion of the number of male users in all users within a certain time period based on the proportion of male users at each time within that time period.

[0182] In addition, it is also possible to analyze the relationships in multiple dimensions. For example, if you want to determine the change relationship of the proportion of the number of women who are interested in catering and have a high degree of interest in all female users within a certain time period. You can first determine the number of women who are interested in catering and have a high degree of interest under the time information within a certain time period. Then determine the number of all women under the time information within a certain time period. Then divide the number of women who are interested in catering and have a high degree of interest by the number of all women to obtain the proportion under the time information within a certain time period, and determine the change relationship of the proportion of the number of women who are interested in catering and have a high degree of interest in all female users within a certain time period based on the proportion under the time information within that time period.

[0183] In addition, it is also possible to evaluate the object attribute tags, take offline the useless tags, and at the same time recommend popular tags and high-quality tags. The popular tags, high-quality tags, and useless tags can be determined by using the tag quality and / or application degree of the object attribute tags, and then the useless tags can be deleted from the data storage table.

[0184] This application also provides some application scenarios, which apply the above object data query method. Specifically, the object data query method can also be applied to various function queries in the driverless scenario, various production data queries in the production scenario, various data queries with obvious characteristics in the cloud computing big data scenario, and the query of running data in the application program in the application program data analysis scenario, etc.

[0185] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0186] Based on the same inventive concept, an embodiment of the present disclosure also provides an object data query device for implementing the object data query method involved above. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the object data query device provided below can refer to the limitations on the object data query method in the above text, and will not be repeated here.

[0187] In one embodiment, as Figure 13 shown, an object data query device 1300 is provided, including: a query condition determination module 1302, a query instruction generation module 1304, a query module 1306, and a query result obtaining module 1308, where:

[0188] The query condition determination module 1302 is configured to determine the query condition indicated by the object data query request in response to the object data query request;

[0189] The query instruction generation module 1304 is configured to generate an object attribute array query instruction applicable to the data storage table in the database according to the object attribute information when the query condition includes object attribute information. The object attribute array query instruction includes at least one object attribute label determined based on the object attribute information;

[0190] The query module 1306 is configured to query an object attribute label array including the at least one object attribute label from the object attribute label array field in the data storage table in response to the object attribute array query instruction; wherein, the object attribute labels in the object attribute label array in the data storage table belong to the same object;

[0191] The query result obtaining module 1308 is configured to obtain a query result that meets the query condition according to the object attribute label array including the at least one object attribute label.

[0192] In one embodiment of the device, when the number of object attribute tags determined based on the object attribute information is at least two, the object attribute array query instruction includes a query array, and the query array includes at least two object attribute tags determined based on the object attribute information. The query module 1306 is further configured to, in response to the object attribute array query instruction, obtain the query array included in the object attribute array query instruction; determine the at least two object attribute tags included in the query array; and query, from the object attribute tag array field in the data storage table, an object attribute tag array that includes the at least two object attribute tags.

[0193] In one embodiment of the device, the query result obtaining module 1308 includes:

[0194] A tag value array determining module, configured to query, from the object attribute tag value array field in the data storage table, an object attribute tag value array that matches the object attribute tag array including the at least one object attribute tag, according to the object attribute tag array including the at least one object attribute tag.

[0195] A tag value determining module, configured to determine, in the object attribute tag value array, object attribute tag values that respectively match the at least one object attribute tag, according to the positions of the at least one object attribute tag in the object attribute tag array.

[0196] A query result obtaining sub-module, configured to obtain a query result that meets the query condition, according to the object attribute tag array including the at least one object attribute tag and the object attribute tag values that respectively match the at least one object attribute tag.

[0197] In one embodiment of the device, the query result obtaining sub-module is further configured to query, from the object identifier field in the data storage table, an object identifier that matches the object attribute tag array including the at least one object attribute tag, according to the object attribute tag array including the at least one object attribute tag; and obtain a query result that meets the query condition, according to the queried object identifier, the object attribute tag array including the at least one object attribute tag, and the object attribute tag value.

[0198] In one embodiment of the device, the query condition further includes a time condition. The query result obtaining sub-module is further configured to screen out, from the object attribute tag values respectively matched by the at least one object attribute tag, the object attribute tag values that match the time information meeting the time condition according to the time condition; search in the data storage table for an object attribute tag array that matches the screened-out object attribute tag values; and obtain a query result meeting the query condition according to the screened-out object attribute tag values and the found object attribute tag array.

[0199] In one embodiment of the device, the query condition further includes at least one object attribute value condition corresponding to the object attribute information. The query result obtaining sub-module is further configured to determine at least one object attribute tag value condition according to the at least one object attribute value condition; determine, from the screened-out object attribute tag values, the object attribute tag values that meet the object attribute tag value condition; search in the data storage table for an object attribute tag array that matches the object attribute tag values meeting the object attribute tag value condition; and obtain a query result meeting the query condition according to the object attribute tag values meeting the object attribute tag value condition and the found object attribute tag array.

[0200] In one embodiment of the device, the device further includes: a first relationship graph generating module, configured to obtain the object attribute tag values meeting the object attribute tag value condition; for each time information meeting the time condition, respectively count the number of object attribute tag values meeting the object attribute tag value condition under the targeted time information; and construct a relationship graph between the counted number of object attribute tag values and the time information meeting the time condition.

[0201] In one embodiment of the device, the device further includes: a second relationship graph generating module, which obtains the object attribute tag values meeting the object attribute tag value condition included in the query result and the object attribute tag array that matches the screened-out object attribute tag values;

[0202] Calculate the quantity ratio between the number of the obtained object attribute tag values and the number of the obtained object attribute tag arrays under each of the time information;

[0203] Construct a relationship graph between the time information and the quantity ratio under each of the time information according to the time information and the quantity ratio under each of the time information.

[0204] In one embodiment of the device, the query result obtaining module 1308 is further configured to query, from the object identification field in the data storage table, an object identification that matches the object attribute label array including the at least one object attribute label according to the object attribute label array including the at least one object attribute label; and obtain a query result that meets the query condition according to the matched object identification and the object attribute label array including the at least one object attribute label.

[0205] In one embodiment of the device, the query condition further includes a time condition, and the query result obtaining module 1308 is further configured to filter out, from the object attribute label array including the at least one object attribute label, an object attribute label array that matches time information meeting the time condition according to the time condition; and obtain a query result that meets the query condition according to the filtered object attribute label array.

[0206] In one embodiment of the device, the device further includes: an object attribute label merging module, configured to obtain object information of at least one object stored in multiple data sources; perform data cleaning on the object information to delete non-compliant data in the object information; perform data analysis on the object information after data cleaning to determine object attribute labels in the object information after data cleaning; and merge object attribute labels of the same object into an object attribute label array.

[0207] In one embodiment of the device, the device further includes: a label deletion module, configured to obtain the label quality and / or application degree of each object attribute label in the object attribute label array in the data storage table, where the label quality is determined based on data related to the object attribute label and the proportion of the data, and the application degree is determined based on operations related to the object attribute label; determine useless object attribute labels of each object attribute label in the object attribute label array in the data storage table according to the label quality and / or the application degree of each object attribute label in the object attribute label array in the data storage table; and delete the useless object attribute labels in the object attribute label array in the data storage table.

[0208] Each module in the above object data query device can be implemented in whole or in part by software, hardware, and a combination thereof. The above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or stored in a memory in the computer device in the form of software, so that the processor can call and execute operations corresponding to the above respective modules.

[0209] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be asFigure 14 As shown in the figure. The computer device includes a processor, a memory, and a network interface connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as object attribute information, data storage tables, and object attribute tags. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements an object data query method.

[0210] Those skilled in the art can understand that Figure 14 the structure shown in the figure is only a block diagram of some structures related to the solution of the present disclosure, and does not constitute a limitation on the computer device to which the solution of the present disclosure is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0211] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the steps in the above method embodiments are implemented.

[0212] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0213] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0214] It should be noted that the object attribute information, object attribute tags, object information of the object, etc. involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant laws, regulations, and standards.

[0215] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided by the present disclosure can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided by the present disclosure can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided by the present disclosure can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0216] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0217] The above-described embodiments merely represent several implementation manners of the present disclosure. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present disclosure. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present disclosure, several modifications and improvements can still be made, and these all belong to the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the appended claims.

Claims

1. A method for querying object data, characterized in that, The method includes: In response to an object data query request, determining the query conditions indicated by the object data query request; When the query conditions include object attribute information, generating, according to the object attribute information, an object attribute array query instruction applicable to a data storage table in a database, where the object attribute array query instruction includes at least one object attribute label determined based on the object attribute information; In response to the object attribute array query instruction, querying, from an object attribute label array field in the data storage table, an object attribute label array that contains the at least one object attribute label; wherein, the object attribute labels in the object attribute label array in the data storage table belong to the same object; Obtaining a query result that meets the query conditions according to the object attribute label array that contains the at least one object attribute label.

2. The method according to claim 1, wherein When the number of object attribute labels determined based on the object attribute information is at least two, the object attribute array query instruction includes a query array, and the query array includes at least two object attribute labels determined based on the object attribute information; The step of, in response to the object attribute array query instruction, querying, from an object attribute label array field in the data storage table, an object attribute label array that contains the at least one object attribute label includes: In response to the object attribute array query instruction, obtaining the query array included in the object attribute array query instruction; Determining the at least two object attribute labels included in the query array; Querying, from an object attribute label array field in the data storage table, an object attribute label array that contains the at least two object attribute labels.

3. The method according to claim 1, wherein The step of obtaining a query result that meets the query conditions according to the object attribute label array that contains the at least one object attribute label includes: According to the object attribute label array that contains the at least one object attribute label, querying, from an object attribute label value array field in the data storage table, an object attribute label value array that matches the object attribute label array that contains the at least one object attribute label; Determining, in the object attribute label value array, object attribute label values that respectively match the at least one object attribute label according to the positions of the at least one object attribute label in the object attribute label array; Obtaining a query result that meets the query conditions according to the object attribute label array that contains the at least one object attribute label and the object attribute label values that respectively match the at least one object attribute label.

4. The method according to claim 3, wherein The step of obtaining a query result that meets the query conditions according to the object attribute label array that contains the at least one object attribute label and the object attribute label values that respectively match the at least one object attribute label includes: According to the object attribute label array that contains the at least one object attribute label, querying, from an object identifier field in the data storage table, an object identifier that matches the object attribute label array that contains the at least one object attribute label; Obtain a query result that meets the query conditions based on the queried object identifier, the object attribute label array containing the at least one object attribute label, and the object attribute label value.

5. The method according to claim 3, characterized in that, The query conditions further include a time condition. The obtaining of the query result that meets the query conditions according to the object attribute label array containing the at least one object attribute label and the object attribute label values respectively matching the at least one object attribute label includes: From the object attribute label values respectively matching the at least one object attribute label, filter out the object attribute label values that match the time information meeting the time condition according to the time condition. In the data storage table, search for the object attribute label array that matches the filtered object attribute label values. Obtain a query result that meets the query conditions according to the filtered object attribute label values and the found object attribute label array.

6. The method according to claim 5, characterized in that, The query conditions further include at least one object attribute value condition corresponding to the object attribute information. The obtaining of the query result that meets the query conditions according to the filtered object attribute label values and the found object attribute label array includes: Determine at least one object attribute label value condition according to the at least one object attribute value condition. From the filtered object attribute label values, determine the object attribute label values that meet the object attribute label value condition. In the data storage table, search for the object attribute label array that matches the object attribute label values meeting the object attribute label value condition. Obtain a query result that meets the query conditions according to the object attribute label values meeting the object attribute label value condition and the found object attribute label array.

7. The method according to claim 6, wherein The method further includes: Obtain the object attribute label values that meet the object attribute label value condition. For each time information that meets the time condition, respectively count the number of object attribute label values that meet the object attribute label value condition under the targeted time information. According to the number of object attribute label values respectively counted for each time information that meets the time condition, construct a relationship graph between the counted number of object attribute label values and the time information that meets the time condition.

8. The method according to claim 6, wherein The method further includes: Obtain the object attribute label values that meet the object attribute label value condition included in the query result and the object attribute label array that matches the filtered object attribute label values. Under each time information, calculate the quantity ratio between the number of obtained object attribute label values and the number of obtained object attribute label arrays. According to the time information and the quantity ratio under each time information, construct a relationship graph between the time information and the quantity ratio under each time information.

9. The method according to claim 1, wherein The obtaining of the query result that meets the query conditions according to the object attribute label array containing the at least one object attribute label includes: According to the object attribute label array including the at least one object attribute label, query the object identifier that matches the object attribute label array including the at least one object attribute label from the object identifier field in the data storage table; Obtain a query result that meets the query condition according to the matched object identifier and the object attribute label array including the at least one object attribute label.

10. The method according to claim 1, wherein The query condition further includes a time condition, and obtaining a query result that meets the query condition according to the object attribute label array including the at least one object attribute label includes: From the object attribute label array including the at least one object attribute label, filter out the object attribute label array that matches the time information meeting the time condition according to the time condition; Obtain a query result that meets the query condition according to the filtered object attribute label array.

11. The method according to claim 1, wherein The object attribute label array in the data storage table is obtained by the following method: Obtain the object information of at least one object stored in multiple data sources; Perform data cleaning on the object information, and delete the non-compliant data in the object information; Perform data analysis on the object information after data cleaning, and determine the object attribute labels in the object information after data cleaning; Merge the object attribute labels of the same object into an object attribute label array.

12. The method according to any one of claims 1 to 11, characterized in that, The method further includes: Obtain the label quality and / or application degree of each object attribute label in the object attribute label array in the data storage table, where the label quality is determined based on the data related to the object attribute label and the proportion of the data, and the application degree is determined based on the operations related to the object attribute label; Determine the useless object attribute labels of each object attribute label in the object attribute label array in the data storage table according to the label quality and / or the application degree of each object attribute label in the object attribute label array in the data storage table; Delete the useless object attribute labels in the object attribute label array in the data storage table.

13. An object data query device, characterized in that, The device includes: A query condition determination module, configured to determine the query condition indicated by the object data query request in response to the object data query request; A query instruction generation module, configured to generate an object attribute array query instruction applicable to the data storage table in the database according to the object attribute information when the query condition includes object attribute information, where the object attribute array query instruction includes at least one object attribute label determined based on the object attribute information; A query module, configured to query the object attribute label array including the at least one object attribute label from the object attribute label array field in the data storage table in response to the object attribute array query instruction; wherein, the object attribute labels in the object attribute label array in the data storage table belong to the same object; A query result obtaining module, configured to obtain a query result that meets the query condition according to the object attribute label array including the at least one object attribute label.

14. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 12 are implemented.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 12 are implemented.

16. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 12 are implemented.

Images