Method and device for determining data table, equipment, medium and product

Abstract

The embodiment of the invention relates to a method and device for determining a data table, equipment, a medium and a product. The method includes determining a target query input from a user and a target set of data tables related to the user. The method further includes determining a first similarity score for the target query input and the data table based on a table description for the data table in the target set of data tables. The method further includes determining a second similarity score for the target query input and the data table based on historical query input corresponding to the data table. The method further comprises the step of determining a popularity score for the data table, wherein the popularity score indicates the frequency degree of accessing the data table by the user. And determining a group of data tables corresponding to the target query input from the target data table set based on the first similarity score, the second similarity score and the popularity score. Through the method, the recall accuracy is improved, the query efficiency is optimized, and the user experience is improved.

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to the field of information retrieval, and more particularly to methods, devices, equipment, media, and products for processing and determining data tables. Background Art

[0002] With the explosive growth of data, the database structure of enterprises and organizations has become increasingly large and complex, and the number of data tables and fields has also increased dramatically. When querying, users no longer simply execute commands in a fixed format, but hope to find relevant data quickly and accurately through natural language.

[0003] There are two main implementation methods for traditional database table recall technology: vector matching and literal matching. Vector matching technology can capture semantic information to a certain extent and is one of the core methods in natural language query. Literal matching technology quickly finds tables and fields that meet possible requirements by matching keywords and characters. This method does not require complex calculations and has a faster response speed.

[0004] With the development of technology, these methods are gradually optimized and applied in database queries, such as introducing more powerful vector representation methods, improving semantic matching capabilities, or using high-speed indexing and caching to optimize literal matching performance. In some specific business scenarios, these technical means enable users to quickly obtain the required table and field information on the natural language query platform to meet the needs of data analysis and user behavior insights. Summary of the invention

[0005] Embodiments of the present disclosure provide a method, apparatus, device, medium, and product for determining a data table.

[0006] According to a first aspect of the present disclosure, a method for determining a data table is provided. The method includes determining a target query input from a user and a target data table set related to the user. The method also includes determining a first similarity score for the target query input and the data table based on a table description for a data table in the target data table set. The method also includes determining a second similarity score for the target query input and the data table based on a historical query input corresponding to the data table. The method also includes determining a heat score for the data table, the heat score indicating how frequently the user accesses the data table. The method also includes determining a group of data tables corresponding to the target query input from the target data table set based on the first similarity score, the second similarity score and the heat score.

[0007] According to a second aspect of the present disclosure, a device for determining a data table is provided. The device includes an input and data table determination module, configured to determine a target query input from a user and a target data table set related to the user; a first similarity score determination module, configured to determine a first similarity score for the target query input and the data table based on a table description for a data table in the target data table set; a second similarity score determination module, configured to determine a second similarity score for the target query input and the data table based on a historical query input corresponding to the data table; a heat score determination module, configured to determine a heat score for the data table, the heat score indicating how frequently the user accesses the data table; and a data table selection module, configured to determine a group of data tables corresponding to the target query input from the target data table set based on the first similarity score, the second similarity score and the heat score.

[0008] In a third aspect of the present disclosure, an electronic device is provided, comprising at least one processor; and a storage device for storing at least one program, wherein when the at least one program is executed by the at least one processor, the at least one processor implements the method according to the first aspect of the present disclosure.

[0009] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0010] In a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, which implements the method according to the first aspect of the present disclosure when executed by a processor.

[0011] It should be understood that the contents described in this content section are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The above and other objects, features and advantages of the present disclosure will become more apparent through a more detailed description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present disclosure.

[0013] Figure 1 A schematic diagram illustrating an example environment in which the apparatus and / or methods of some embodiments of the present disclosure may be implemented; Figure 2 A schematic diagram illustrating an example method for determining a data table according to some embodiments of the present disclosure is illustrated; Figure 3 A schematic diagram illustrating an example process of table recall with a screening process according to some embodiments of the present disclosure; Figure 4 A schematic diagram illustrating an example process of table recall without a screening process according to some embodiments of the present disclosure; Figure 5 A schematic block diagram of an apparatus for determining a data table according to some embodiments of the present disclosure is illustrated; Figure 6 A schematic block diagram of an example device suitable for implementing various embodiments of the present disclosure is illustrated.

[0014] In the various drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0015] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0016] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0017] For example, when receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0018] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0019] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0020] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0021] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0022] Generally, when using natural language for querying, it is necessary to effectively recall the tables in the database. For recalling database tables and fields, the vector matching or literal matching method described above is usually used. Among them, vector matching converts the query request in natural language into a specific query vector, and then recalls the tables and fields with the highest similarity by calculating the similarity between the query vector and the table in the database. Literal matching uses literal similarity to complete the matching of table information. However, due to the complex structure of the database, the huge amount of data, and the diversity of query requirements, accurately locating the required database table for query has become an important technical problem. At the same time, while ensuring accuracy, improving the query speed to meet the user's real-time requirements for data query is also a key technical problem that needs to be solved. The vector matching method relies on the calculation and comparison of vectors, so when processing large-scale databases, the calculation amount is huge and the efficiency is low. Secondly, since the vector matching method cannot accurately reflect the complex relationship between tables and fields, there are problems in the accuracy of recall. Finally, although the literal matching method can improve the speed of recall, it cannot handle semantic similarity, so there are also problems in the accuracy of recall. These two methods each have their significant disadvantages, resulting in a poor user experience.

[0023] To this end, an embodiment of the present disclosure proposes a method for determining a data table. In this method, after a user inputs a target query input to perform a query operation, a computing device can first determine a table description of a data table in a target data table set related to the user, and then calculate a first similarity score for the target query input and the data table based on the table description and the query input. Then, based on the historical query input corresponding to the data table, a second similarity score between the target query input and the data table is determined. The computing device can further determine the heat of the data table relative to the user. Then, the computing device uses the first similarity score, the second similarity score and the heat score to determine a group of data tables corresponding to the target query input from the target data table set. Through this method, when recalling a data table, multiple factors such as the similarity between the user query and the table description of the data table, the query similarity and the user heat are comprehensively considered, so that the recalled table is more in line with the user's intention, the recall accuracy is improved, and the required data can be obtained faster and more accurately, thereby greatly improving the user experience.

[0024] The embodiments of the present disclosure will be described in detail below in conjunction with the accompanying drawings. Figure 1 An example environment in which the device and / or method of an embodiment of the present disclosure may be implemented is shown. In environment 100, computing device 102 may recall corresponding data tables according to a target query input from a user.

[0025] Examples of computing device 102 include, but are not limited to, a personal computer, a server computer, a handheld or laptop device, a mobile device (such as a mobile phone, a personal digital assistant (PDA), a media player, etc.), a multiprocessor system, consumer electronics, a minicomputer, a mainframe computer, a distributed computing environment including any of the above systems or devices, and the like.

[0026] like Figure 1 As described above, when a user wants to obtain some information, a target query input 104 may be provided to a computing device 102. For example, in a natural language query data scenario, "sales of a certain brand in 7 days" is input through a search engine to query information. Additionally, a natural language query data scenario may refer to a specific application scenario in which a user interacts with a system through natural language to query data or information.

[0027] After the computing device 102 receives the user's target query input 104, it needs to further obtain a target data table set 106. The target data table set 106 is some pre-stored data tables. In one example, the computing device 102 obtains the target data table set from other computing devices, such as a server, via a network. In another example, the computing device 102 locally stores the target data table set 106.

[0028] The target data table set 106 includes multiple data tables. For example, the target data table set 106 may include production data tables, sales data tables, and material data tables of some enterprises. In addition, a corresponding table description may be generated for each data table in the target data table set 106 to describe the information of this table. For example, the table description information may consist of a table name, a main purpose of the table, and a secondary purpose of the table, wherein the main purpose of the table is used to describe information related to the main purpose of the table, and the secondary purpose may be used to describe information that is not very related to the main purpose of the table. For example, if the table is a sales table, the main purpose may be information related to the name of the product and sales, while the secondary purpose may be some information that is not very related to the sales information, such as some material information related to the salesperson and the products sold. In Figure 1 , for a data table 108 in the target data table set, it has a corresponding table description 112. The computing device 102 may calculate a first similarity score 116 between the target query input 104 and the table description 112. For example, the first similarity score 116 between the target query input 104 and the data table 108 is calculated using a vector corresponding to the target query input 104 and a vector corresponding to the table description 112.

[0029] In addition, the computing device 102 may also obtain historical query inputs associated with each data table in the target data table set. If a user has previously obtained the data table through query input, the query input may be used as the historical query input for the data table. Figure 1 As shown, the data table 108 may have a history query input 114 . Figure 1 104 and 114. The data table 108 corresponds to a historical query input, which is only used to describe the present disclosure, but not to specifically limit the present disclosure. Each data table in the target data set may have one or more historical query inputs. Therefore, the computing device 102 may also calculate a second similarity score 118 using the target query input 104 and the historical query input 114. For example, the second similarity score 118 between the target query input 104 and the data table 108 is calculated using a vector corresponding to the target query input 104 and a vector corresponding to the historical query input 114.

[0030] In addition, the computing device 102 can also obtain the popularity score of the data table in the target data table set 106. For example, the popularity score 110 of the data table 108 can be obtained, and the popularity score reflects the frequency of the user using the data table. For example, the popularity can be represented by the number of times the user has used the data table or the field in the data table by retrieving it. Additionally, the popularity score can also be determined by the popularity of the data table for the user and the popularity for the user's department. If the data table has not been used by the user's department, the popularity score can be set to a predetermined popularity.

[0031] Then, the computing device 102 can determine a target score for the similarity between the data table and the target query input based on the popularity score 110, the first similarity score 116, and the second similarity score 118 of the data table 108. Similarly, the computing device 102 can obtain a score for the similarity with the target query input for each table in the target data table set 106. Then, based on these scores, the computing device 102 selects a set of data tables 120 from the target data table set to be used as data tables for the target query input.

[0032] Additionally, the target data table set 106 in the computing device 102 is obtained by filtering from all available data table sets. In one example, the computing device filters the target data table set from all available data table sets based on the identifier of the user department or the identifier of the user. In another example, the target data table set is an unfiltered set of all available data tables. The above examples are only used to describe the present disclosure, but are not specific limitations of the present disclosure.

[0033] Through this method, the efficiency and accuracy problems encountered in recalling required data tables in natural language queries are solved, the recall accuracy is improved, the query efficiency is optimized, performance improvements are provided, and the user experience is improved.

[0034] Combined with the above Figure 1 A schematic diagram of an example environment in which the devices and / or methods of some embodiments of the present disclosure may be implemented is described below. Figure 2 A schematic diagram depicting an example method for determining a data table according to some embodiments of the present disclosure. Figure 2 The method in can be Figure 1 Executed by computing device 102 or any suitable computing device.

[0035] In such Figure 2As shown, in the example method 200, at block 202, the computing device 102 determines a target query input from a user and a target data table set related to the user. The computing device 102 may obtain the target query input 104, such as a query statement input in a natural language. Furthermore, the computing device 102 may convert the target query input into a quantitative representation, such as converting the target query input into a vector by a vectorization method.

[0036] In some embodiments, while determining the target query input from the user, the computing device 102 may also determine a set of data tables that can be used for querying, and then filter out a target set of data tables related to the user from the set of data tables used for querying. In one example, the target set of data tables is selected from the set of data tables based on the identification of the user. At this time, the identification of the user is used to select the data tables used by the user from the set of data tables as the target set of data tables. In another example, the computing device may first determine the identification of the department to which the user belongs based on the identification of the user. For example, each user belongs to a department, and the identification of the user and the identification of the department may be associated and stored. Then, the computing device may use the identification of the department to select multiple data tables used by the department from the set of data tables as the target set of data tables. Additionally, each data table or field in the data table may be stored in the set of data tables by which users or users of which departments have used it. Therefore, the computing device may determine a set of data tables as the target set of data tables based on the identification of the user or the identification of the department. The above examples are only used to describe the present disclosure, and are not specifically limited to the present disclosure. Those skilled in the art may filter the data set in any suitable manner. Alternatively, the target data table set may also be an unfiltered data table set.

[0037] At block 204, a first similarity score between the target query input and the data table is determined based on the table descriptions for the data tables in the target data table set. In order to facilitate table lookup, a corresponding table description may be generated for each table in the target data table set or the data table set. For example, the table description information consists of a table name, a main intent, and a secondary intent. The main intent and the secondary intent of the table may exist in the form of vectors in the database.

[0038] Therefore, the computing device 102 can perform vectorization processing on the target query input to determine the first quantized representation corresponding thereto, and can also obtain the second quantized representation of the data table in the target data table set. Then, the computing device 102 can generate a first similarity score for the target query input and the data table based on the first quantized representation and the second quantized representation. For example, the computing device can calculate the similarity score of the target query input and the data table through two phasors. For example, the cosine similarity or Euclidean distance of two vectors can be calculated to determine the similarity score.

[0039] At block 206, a second similarity score for the target query input and the data table is determined based on the historical input query corresponding to the data table. In order to facilitate obtaining an accurate data table, the computing device 102 may also obtain the historical input query used to query the data table. Then, the similarity between the target query input and the data table is further determined by comparing the target query input with the historical input query.

[0040] In some embodiments, the historical query input for the data table can be converted into a quantitative representation, which can also be referred to as a third quantitative representation for ease of description. Then, the computing device 102 stores the third quantitative representation in a vector library. Therefore, the computing device 102 can calculate the quantitative representation of the target query input and the quantitative representation of the historical query input to determine a second similarity score for the target query input and the data table. For example, by storing the vector of the historical query input and the vector of the table description in the vector library, and using the vector library to recall similar vectors, the recall speed can be greatly improved, so that the system can efficiently return matching results in a large-scale data environment.

[0041] At block 208, based on the popularity score for the data table, the popularity score indicates how frequently the user accesses the data table. In order to further determine that the acquired table is the most relevant table to the user, the popularity of the data table for the user and the popularity of other users in the department of the user or the popularity of the department of the user may also be stored while storing the data table.

[0042] In some embodiments, the computing device may first determine whether the user has used the data table. If the user has used the data table, the personal heat for the user and the department heat for the department to which the user belongs associated with the data table may be determined. Then, the computing device 102 determines the heat score for the data table using the personal heat and the department heat. In one example, the computing device 102 determines the heat score for the data table by weighting the personal heat and the department heat. In another example, the computing device 102 may average the two heats to determine the heat score for the data table. If the user has not used the data table, the heat score may be determined as a predetermined heat score. The above examples are only used to describe the present disclosure and are not specific limitations of the present disclosure.

[0043] In some embodiments, the computing device 102 may collect a set of historical query inputs from the user for the data table. Then, the computing device 102 determines the personal popularity of the user associated with the data table based on the set of historical query inputs. For example, if the user has used the data table five times when performing retrieval queries, the popularity of the data table for the user may be determined to be 5. Similarly, the popularity of other users for the table may also be represented by the number of times the user has visited it. The popularity of the data table for the user's department may be the popularity of the department stored in the database, or the popularity of the department may be the sum of the popularity of the department's users for the data table.

[0044] At block 210, the computing device 102 determines a set of data tables corresponding to the target query input from the target data set based on the first similarity score, the second similarity score, and the heat score. After obtaining various scores of the target query input for a data table in the target data table set 106, the computing device 102 may also calculate the scores of the target query input relative to other data tables in the target data table set 106. The computing device then uses these scores to select a set of data tables from the target data table set.

[0045] In some embodiments, the computing device 102 may use the first similarity score, the second similarity score, and the heat score to calculate a target score for the data table. For example, the target score may be obtained by weighting the first similarity score, the second similarity score, and the heat score, or the target score may be calculated by a predetermined function. Then, the computing device 102 selects a set of data tables corresponding to the target query input from the target data table set according to the target score.

[0046] In some embodiments, the computing device 102 may sort the multiple data tables in the target data table set based on the target score and the scores of other data tables in the target data table set. Then, the computing device selects a group of data tables with scores greater than a threshold from the sorted data table set. For example, the first three data tables are selected.

[0047] In some embodiments, multi-threaded parallel calculations of similar question scores, table description scores, and table popularity scores can be used to enable the system to complete score calculations more quickly when faced with a large number of queries, fully utilize computing resources, and improve computing efficiency.

[0048] In some instances, user popularity-related information and question vectors can be cached, which further improves recall efficiency. By storing popularity data in the cache, the system can quickly query users' access preferences for different data tables, thereby giving priority to tables that users frequently query during recall, improving recall accuracy and response speed. Moreover, storing the vector of the target query input in the cache allows the cache to be read directly for similar or repeated queries, avoiding repeated calculations and further improving recall efficiency. In general, by caching this information, the system can respond more efficiently when faced with a large number of user queries, reducing latency and providing users with a faster query experience.

[0049] Through this method, the recall algorithm integrates three features of table description, similarity query, and user popularity, and integrates the heat calculation of department popularity and personal popularity, which greatly improves the recall accuracy, improves the recall efficiency, provides performance improvement, and improves the user experience.

[0050] Combined with the above Figure 2 A schematic diagram of an example method for determining a data table according to some disclosed embodiments is described. Figure 3 A diagram depicting an example process for table recall with a screening process according to some embodiments of the present disclosure. Figure 3 The example process can be obtained by Figure 1 The system may be executed by the computing device 102 shown in FIG. 1 or any suitable device.

[0051] In example 300, the loading process of table recall with screening process is described. At box 302, the operation starts. The user enters natural language at the specified location for target query. Then, at box 304, the input target query input is vectorized by vectorization. For example, the process of vectorization is the process of generating embedded vectors, which converts high-dimensional data (such as sentences, pictures or other types of data) into vectors. These vectors can capture the semantic information or other important features of the data. At box 306, the computing device 102 filters out the data sets that are not clicked by the user's department. Table recall also needs to construct similar query data sets, table description data sets, and user table heat data sets. Among them, the similar query data set needs to collect historical user queries, convert the queries into vectors through the embedding method, and store them in the vector library. The table description data set uses the table description information to generate vectors and store them in the vector library. The user heat data set needs to collect information such as tables and fields corresponding to the user's historical questions, and generate heat information for each user on each table.

[0052] Then, at box 308, the data table description is recalled. The data table description is recalled by matching the vector with the description of the data set. The table description information consists of the table name, the main intent of the table and the secondary intent of the table, where the main intent and the secondary intent of the table can be manually annotated to represent the problems that this table is mainly used to solve. At box 310, similar query recall is performed. The similarity of the two queries is calculated using the target query and the vector of the query in the feedback user. At box 314, it is determined whether the user has used the data table.

[0053] If it is determined that the user has used the data table, the fusion heat 320 is calculated according to the user's personal heat 312 and the department heat 316 using formula (1):

[0054] Where user_hot_score is the user hot score, department_hot_score is the department hot score, and alpha is a weight coefficient used to adjust the influence of the user hot score and the department hot score. Next, the recall algorithm of the three features of table description, similarity query, and user hotness is also integrated. In box 322, the total score of each data table is obtained by weighting through formula (2):

[0055] in and is a weight parameter, dataset_desc_score is the similarity score between the table description recall vector and the user question vector, sim_query_score is the similarity score between the recall similar question and the user question vector, and hot_score is the hotness score of the recall table. In addition, hot_score can also be calculated using the following formula (3):

[0056] Among them, pv(i) is the page views of the i-th data table. If data table i is not in the mapping, pv(i) is equal to 1. median is the median of all pv values, and max(pv) is the maximum value of all pv values.

[0057] If it is determined that the user has not used the data set, the default popularity is used at box 318, and the recall algorithm that integrates the three features of table description, similarity query, and user popularity is used to obtain the total score of each data table by weighting according to formula (1). Based on the total score obtained at box 322, the table selection result is returned at box 324 according to the sorting of the total score. The goal of the above table recall is to perform a rough recall and provide a set of relatively accurate tables for subsequent field recall.

[0058] Combined with the above Figure 3 A schematic diagram of an example process of table recall with a screening process according to some embodiments of the present disclosure is described. Figure 4 A diagram depicting an example process for table recall without a filtering process according to some embodiments of the present disclosure. Figure 4 The example process can be obtained by Figure 1 The system may be executed by the computing device 102 shown in FIG. 1 or any suitable device.

[0059] like Figure 4 In the example 400 shown, the loading process of table recall without a screening process is described. At box 402, the user enters natural language at a specified location to perform a target query. Then, at box 404, the query request in the input natural language is vectorized by vectorization. At box 406, all data sets in the database are directly used. Then, at box 408, the data set description is recalled. At box 410, a similar query is recalled. At box 414, it is determined whether the user has used the data set.

[0060] If it is determined that the user has used the data set, the fusion heat 420 is calculated according to the user's personal heat 412 and the department heat 416 by formula (1), and the recall algorithm of the three-way features of the fusion table description, similar questions and user heat is used. At box 422, the total score of each data set is obtained by formula (2).

[0061] If it is determined that the user has not used the data set, the default popularity is used at box 418, and the recall algorithm of the three features of table description, similar questions, and user popularity is integrated, and the total score of each data set is obtained by weighting through formula (2). Based on the total score obtained at box 422, the table selection result is returned at box 424 according to the ranking of the total score.

[0062] Figure 5 FIG. 1 is a schematic block diagram of an apparatus for determining a data table according to some embodiments of the present disclosure. Figure 5 As shown, the device 500 includes a query input and data table determination module 502, which is configured to determine a target query input from a user and a target data table set related to the user; a first similarity score determination module 504, which is configured to determine a first similarity score for the target query input and the data table based on a table description for a data table in the target data table set; a second similarity score determination module 506, which is configured to determine a second similarity score for the target query input and the data table based on a historical query input corresponding to the data table; a heat score determination module 508, which is configured to determine a heat score for the data table, the heat score indicating how frequently the user accesses the data table; and a data table selection module 510, which is configured to determine a group of data tables corresponding to the target query input from the target data table set based on the first similarity score, the second similarity score and the heat score.

[0063] In some embodiments, the query input and data table determination module 502 includes: a data table determination module configured to determine a set of data tables that can be used for query; and a first data table set selection module configured to select a target data table set from the data table set based on a user's identification.

[0064] In some embodiments, the data table set selection module includes: a department identification determination module, configured to determine the identification of the department to which the user belongs based on the identification of the user; and a first data table set selection module, configured to select multiple data tables used by the department from the data table set as a target data table set based on the identification of the department.

[0065] In some embodiments, the first similarity score determination module 504 includes: a quantitative representation determination module, configured to determine a first quantitative representation for a target query input and a second quantitative representation for a table description; a first score determination module, configured to determine a first similarity score for the target query input and the data table based on the first quantitative representation and the second quantitative representation.

[0066] In some embodiments, the second similarity score determination module 506 includes: a third quantitative representation determination module, configured to determine a third quantitative representation for the historical query input; and a second score determination module, configured to determine a second similarity score for the target query input and the data table based on the first quantitative representation and the third quantitative representation.

[0067] In some embodiments, the apparatus 500 further includes a quantized representation generating module configured to generate a third quantized representation by converting the historical query input; and a storage module configured to store the third quantized representation in a vector library.

[0068] In some embodiments, the popularity score determination module 508 includes: a data table determination module, configured to determine whether a user has used a data table; a popularity determination module, configured to determine the personal popularity of the user and the department popularity of the user's department associated with the data table in response to the user having used the data table; and a third score determination module, configured to determine the popularity score for the data table based on the personal popularity and the department popularity.

[0069] In some embodiments, the heat score determination module 508 further includes: a predetermined heat score determination module configured to determine the heat score as a predetermined heat score in response to the user not having used the data table.

[0070] In some embodiments, the device 500 also includes: a query input collection module, configured to collect a set of historical query inputs of the user for the data table; and a personal popularity determination module, configured to determine the personal popularity of the user associated with the data table based on the set of historical query inputs.

[0071] In some embodiments, the data table selection module 510 includes: a target score determination module, configured to determine a target score for a data table based on a first similarity score, a second similarity score, and a heat score; and a data table selection module, configured to select a set of data tables corresponding to a target query input from a target data table set based on the target score.

[0072] In some embodiments, the target score determination module includes: a weighted operation module configured to obtain the target score by performing a weighted operation on the first similarity score, the second similarity score, and the heat score.

[0073] In some embodiments, the data table selection module includes: a sorting module, configured to sort multiple data tables in the target data table set based on the target score and the scores of other data tables in the target data table set; and a group of data table selection modules, configured to select a group of data tables with scores greater than a threshold from the sorted data table set.

[0074] In some embodiments, the table description includes a table name and table intent information, where the intent information is used to describe content information of the table.

[0075] Figure 6 A schematic block diagram of an example device 600 that may be used to implement embodiments of the present disclosure is shown. Figure 1 The computing device 102 in the embodiment can be implemented by using the device 600. As shown in the figure, the device 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 602 or computer program instructions loaded from a storage unit 608 to a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0076] A number of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0077] The various processes and processing described above, such as method 200 and examples 300 to 400, may be performed by processing unit 601. For example, in some embodiments, method 200 and examples 300 to 400 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by CPU 601, one or more actions of the example method 200 and examples 300 to 400 described above may be performed.

[0078] The present disclosure may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present disclosure.

[0079] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (non-exhaustive lists) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, for example, a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination thereof. The computer-readable storage medium used herein is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (e.g., a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0080] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0081] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of the computer-readable program instructions to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit may execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0082] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0083] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0084] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0085] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0086] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for determining a data table, comprising: Determine a target query input from a user and a target data table set related to the user; Determining a first similarity score between the target query input and the data table based on a table description for the data table in the target data table set; Determining a second similarity score for the target query input and the data table based on the historical query input corresponding to the data table; Determining a popularity score for the data table, the popularity score indicating how frequently the user accesses the data table; as well as Based on the first similarity score, the second similarity score, and the heat score, a group of data tables corresponding to the target query input is determined from the target data table set.

2. The method according to claim 1, wherein determining a target query input from a user and a target data table set related to the user comprises: Determine the set of data tables that can be used for querying; as well as The target data table set is selected from the data table set based on the identifier of the user.

3. The method according to claim 2, wherein selecting the target data table set from the data table set comprises: Based on the identifier of the user, determining the identifier of the department to which the user belongs; as well as Based on the identification of the department, a plurality of data tables used by the department are selected from the data table set as the target data table set.

4. The method of claim 1 , wherein determining a first similarity score for the target query input and the data table comprises: Determining a first quantized representation for the target query input and a second quantized representation for the table description; Based on the first quantized representation and the second quantized representation, a first similarity score for the target query input and the data table is determined.

5. The method of claim 4, wherein determining a second similarity score for the target query input and the data table comprises: determining a third quantized representation for the historical query input; as well as Based on the first quantized representation and the third quantized representation, a second similarity score for the target query input and the data table is determined.

6. The method according to claim 5, further comprising: generating the third quantized representation by transforming the historical query input; as well as The third quantized representation is stored in a vector library.

7. The method according to claim 1, wherein determining a heat score for the data table comprises: determining whether the user has used the data table; In response to the user having used the data table, determining a personal popularity for the user and a department popularity for the department to which the user belongs, which are associated with the data table; as well as A popularity score for the data table is determined based on the personal popularity and the department popularity.

8. The method according to claim 7, wherein determining the heat score for the data table further comprises: In response to the user not having used the data table, the popularity score is determined to be a predetermined popularity score.

9. The method according to claim 7, further comprising: Collecting a set of historical query inputs of the user for the data table; as well as Based on the set of historical query inputs, a personal popularity associated with the data table for the user is determined.

10. The method according to claim 1, wherein determining a set of data tables corresponding to the target query input from the target data table set based on the first similarity score, the second similarity score and the heat score comprises: determining a target score for the data table based on the first similarity score, the second similarity score, and the heat score; as well as Based on the target score, a group of data tables corresponding to the target query input is selected from the target data table set.

11. The method according to claim 10, wherein determining a target score for the data table based on the first similarity score, the second similarity score, and the heat score comprises: The target score is obtained by performing a weighted operation on the first similarity score, the second similarity score, and the heat score.

12. The method according to claim 10, wherein based on the target score, determining a group of data tables corresponding to the target query input from the target data table set comprises: sorting the multiple data tables in the target data table set based on the target score and scores of other data tables in the target data table set; as well as A group of data tables with scores greater than a threshold are selected from the sorted data table set.

13. The method according to claim 1, wherein the table description includes a table name and table intent information, and the table intent information is used to describe content information of the table.

14. A device for determining a data table, comprising: An input and data table determination module, configured to determine a target query input from a user and a target data table set related to the user; A first similarity score determination module is configured to determine a first similarity score for the target query input and the data table based on a table description for a data table in the target data table set; A second similarity score determination module, configured to determine a second similarity score for the target query input and the data table based on the historical query input corresponding to the data table; a heat score determination module, configured to determine a heat score for the data table, the heat score indicating how frequently the user accesses the data table; as well as The data table selection module is configured to determine a group of data tables corresponding to the target query input from the target data table set based on the first similarity score, the second similarity score and the heat score.

15. An electronic device, comprising: at least one processor; as well as A storage device, used to store at least one program, when the at least one program is executed by the at least one processor, so that the at least one processor implements the method according to any one of claims 1-13.

16. A computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method according to any one of claims 1 to 13 when executed by a processor.

17. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 13.

Images