Data set intention generation method and device, equipment, medium and program product

Abstract

The invention relates to the technical field of data processing, and discloses a data set intention generation method and device, equipment, a medium and a program product. The method comprises the following steps: acquiring data table information in a data set; analyzing the data table information, and extracting a data table entity and a key index corresponding to the data table information; and generating a data table intention based on the association relationship between the data table entity and the key index. According to the technical scheme, automatic extraction of the data table intention is achieved, manual intervention is reduced, the recognition efficiency of the data table intention is improved, the generation accuracy of the data table intention is improved, and therefore the accuracy of data recall according to the data table intention is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a method, apparatus, device, medium and program product for generating the intent of a data set. Background Art

[0002] In data analysis and business decision-making, it is usually necessary to quickly and accurately extract information from a large number of data tables. However, faced with complex and diverse data table structures, manual analysis is mainly relied on to identify and define the data table intent, which is not only time-consuming and laborious, but also prone to inaccurate recognition of data table intent, resulting in inaccurate data recall and affecting decision-making quality. Summary of the invention

[0003] In view of this, the present disclosure provides a method, apparatus, device, medium and program product for generating the intent of a data set to solve the problem of inaccurate recognition of the intent of a data table.

[0004] In a first aspect, the present disclosure provides a method for generating the intent of a data set, including: obtaining data table information in the data set; parsing the data table information, extracting data table entities and key indicators corresponding to the data table information; and generating data table intent based on the association relationship between the data table entities and the key indicators.

[0005] In a second aspect, the present disclosure provides a device for generating intent of a data set, including: an acquisition module for acquiring data table information in a data set; a parsing module for parsing the data table information and extracting data table entities and key indicators corresponding to the data table information; and an intent generation module for generating data table intents based on the association between the data table entities and the key indicators.

[0006] In a third aspect, the present disclosure provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the intention generation method of the data set of the above-mentioned first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0007] In a fourth aspect, the present disclosure provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for generating an intention of a data set according to the first aspect or any corresponding embodiment thereof.

[0008] In a fifth aspect, the present disclosure provides a computer program product, comprising computer instructions for causing a computer to execute the method for generating an intention of a data set according to the first aspect or any corresponding embodiment thereof.

[0009] The method, apparatus, equipment, medium and program product for generating the intent of a data set provided by the present invention parses the acquired data table information to determine the data table entities and key indicators corresponding to the data table information, and determines the data table intent in combination with the association between the data table entities and the key indicators, thereby realizing the automatic extraction of the data table intent, reducing manual intervention, and improving the recognition efficiency of the data table intent. At the same time, it avoids the errors introduced by manual analysis, improves the generation accuracy of the data table intent, and thus improves the accuracy of data recall based on the data table intent. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0011] Figure 1 is a flowchart of a method for generating an intention of a data set according to an embodiment of the present disclosure;

[0012] Figure 2 is a flowchart of a method for generating intent of another data set according to an embodiment of the present disclosure;

[0013] Figure 3 is a flowchart of another method for generating an intent of a data set according to an embodiment of the present disclosure;

[0014] Figure 4 is a structural block diagram of a device for generating an intention of a data set according to an embodiment of the present disclosure;

[0015] Figure 5 It is a schematic diagram of the hardware structure of the computer device of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0017] 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.

[0018] For example, in response to 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] In the current data analysis field, it is usually necessary to rely on manual methods to identify and define the intent of data tables. This method mainly includes the following steps: data analysts manually extract the main intent and auxiliary intent of the table by understanding the structure and fields of the data table. However, this method has certain defects:

[0023] 1) Low efficiency. Manual analysis takes a lot of time and effort, especially when faced with complex and diverse data tables, the efficiency problem is particularly prominent.

[0024] 2) Lack of accuracy. Due to reliance on personal experience and understanding, manual analysis is prone to misjudgment, resulting in inaccurate recognition of the intent of the data table, which in turn affects the effectiveness of data recall.

[0025] 3) Poor scalability. As the amount of data increases and business needs change, manual methods are difficult to adapt and expand quickly and cannot meet the company's requirements for real-time and flexibility.

[0026] 4) Lack of standardization. Different analysts may have different understandings of the same data table, resulting in a lack of consistency and standardization in intent definitions, affecting the uniformity of subsequent data analysis and decision-making.

[0027] At present, although there are some automated tools that can assist in data analysis, most of them are limited to data cleaning and simple statistical analysis, and have not yet effectively solved the problem of automatic generation of data table intent. Therefore, how to automatically and accurately generate data table intent to improve the efficiency and accuracy of data analysis needs to be solved urgently.

[0028] Based on this, the technical solution disclosed in the present invention automatically generates the main intention and auxiliary intention of the data table by automatically identifying the data table entities and key indicators corresponding to the data table information, and the association relationship between the data table entities and the key indicators, thereby improving the accuracy and efficiency of data table recall.

[0029] According to an embodiment of the present disclosure, an embodiment of a method for generating intent of a data set is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0030] In this embodiment, a method for generating an intention of a data set is provided, which can be used in computer devices such as mobile phones, computers, tablet computers, etc. Figure 1 is a flowchart of a method for generating an intention of a data set according to an embodiment of the present disclosure, such as Figure 1 As shown, the process includes the following steps:

[0031] Step S101, obtaining data table information in a data set.

[0032] The data table information is the attribute information of the data table, such as the table name, the index of the data table, the index value of the data table, etc. Different data analysis fields have different data sets, each of which includes one or more data tables, and each data table has corresponding data table information.

[0033] Specifically, for different data analysis scenarios, different methods can be used to obtain data table information. For example, in a database management system, SQL statements can be used to obtain data table information in a data set; a database connection library under a programming language can be used to execute SQL statements to query and obtain data table information in a data set; or a graphical interface provided by a database management tool can be used to query and obtain data table information in a data set. Of course, other methods can also be used to obtain data table information, which are not specifically limited here.

[0034] Step S102: parse the data table information to extract data table entities and key indicators corresponding to the data table information.

[0035] The data table entity is the core entity that represents the data table, such as orders, amounts, stores, etc.; key indicators are important data indicators contained in the data table, such as data indicators of specific columns, data indicators of specific rows, statistical indicators, etc. Specifically, there will be a unique data table entity for a data table. Due to the large number of data sets in the database, the intent recognition model can be used here to complete the automatic extraction of data table entities. The data table information is input into the intent recognition model, and the data table information is entity recognized through the intent recognition model to obtain the data table entity corresponding to the data table information. At the same time, the data indicators in the data table information are analyzed to filter out the key indicators related to the data table.

[0036] Step S103, generating a data table intent based on the association relationship between the data table entity and the key indicators.

[0037] The association relationship represents the association between the data table entity and the key indicator in terms of attributes. The data table intent is the descriptive information generated for the purpose and content of the data table. The intent recognition model is used to analyze the association relationship between the data table entity and the key indicator, understand the importance of each key indicator in the data table, and determine the main content represented by the data table based on the importance of each key indicator. The main content represented by the data table is used as the intent prompt information to generate the data table intent corresponding to the data table.

[0038] It should be noted that the intent recognition model can be pre-trained based on a large model architecture, or pre-trained using a deep learning model architecture, or pre-trained using a neural network model architecture. There is no specific limitation here, as long as it can achieve entity recognition and intent generation.

[0039] The method for generating the intent of a data set provided in this embodiment parses the acquired data table information, determines the data table entities and key indicators corresponding to the data table information, and determines the data table intent in combination with the association between the data table entities and key indicators, thereby realizing the automatic extraction of the data table intent, reducing manual intervention, and improving the efficiency of identifying the data table intent. At the same time, it avoids the errors introduced by manual analysis, improves the accuracy of generating the data table intent, and thus improves the accuracy of data recall based on the data table intent. Furthermore, through the automated process, the consistency and standardization of the generation of the data table intent is ensured, supporting unified data analysis and decision-making.

[0040] In this embodiment, a method for generating an intention of a data set is provided, which can be used in computer devices such as mobile phones, computers, tablet computers, etc. Figure 2 is a flowchart of a method for generating an intention of a data set according to an embodiment of the present disclosure, such as Figure 2 As shown, the process includes the following steps:

[0041] Step S201, obtaining data table information in a data set. Please refer to the relevant description of the corresponding steps in the above embodiment for details, which will not be repeated here.

[0042] Step S202: parse the data table information to extract data table entities and key indicators corresponding to the data table information.

[0043] Specifically, the above step S202 includes:

[0044] Step S2021, performing entity recognition on the data table name in the data table information to obtain the data table entity.

[0045] Each data table has a unique data table name, and each data table name has a unique data table entity. There are multiple data tables in the dataset. The data table names of each data table are recognized by the intent recognition model, and the entities in the data table names are extracted to obtain the data table entities. For example, the data table name "Closed-loop order multidimensional statistics table" is input into the intent recognition model, and the data table name is recognized by the intent recognition model to obtain the data table entity "Order"; for example, the data table name "Merchant information statistics wide table" is input into the intent recognition model to obtain the data table entity "Merchant".

[0046] In some optional implementations, the above step S2021 includes:

[0047] Step a1, constructing prompt description information based on preset entity recognition examples.

[0048] Step a2: Instruct the entity recognition of the data table name according to the prompt description information, and generate the data table entity corresponding to the data table name.

[0049] The preset entity recognition example is a pre-set prompt example, which is used to instruct the intent recognition model to perform entity recognition on the table name of the data table to extract accurate entity information from it. The prompt description information is used to guide the intent recognition model to perform entity recognition.

[0050] Specifically, the system prompt information corresponding to the intent recognition model and the preset entity recognition examples are integrated to obtain the prompt description information corresponding to the intent recognition model. According to the prompt description information, the intent recognition model is guided to automatically identify the core entities of each data table name to obtain the data table entities corresponding to each data table name. Among them, the system prompt information includes role prompt information, recognition step prompt information and constraint prompt information. The system prompt information is integrated with the preset entity recognition example to obtain the prompt description information as follows:

[0051] "##Role

[0052] You are an expert in intelligent problem solving and are good at extracting entities based on data table names.

[0053] ##step

[0054] - Understand user questions and extract the data table names in user questions

[0055] -Based on the table name, determine the most important entity stored in the table

[0056] -Returns the data table entity according to the specified format

[0057] ##Example

[0058] -Question: E-commerce order form

[0059] -Answer: Order

[0060] -Question: Merchant information table

[0061] -Answer: Merchant

[0062] ##constraint

[0063] - Do not answer questions that are not related to extracting data table entities

[0064] -Answer can only contain unique entities from a table."

[0065] In the above implementation, the preset entity recognition examples are used to construct prompt description information, which enhances the recognition ability of the intent recognition model for data table entities and improves the recognition accuracy and consistency of data table intent. The data table entities are automatically extracted through the intent recognition model, which reduces manual intervention and improves recognition efficiency.

[0066] Step S2022, parsing the data indicators in the data table information to obtain key indicators of the data table.

[0067] The data table information contains all the data indicators that constitute the data table. Each data indicator in the data table information is analyzed to determine the query information generated for each data indicator in the historical period. The query information is used to determine the importance of each data indicator to the data table. All data indicators in the data table are filtered based on the importance of the data indicators to obtain the key indicators of the data table.

[0068] In some optional implementations, the above step S2022 includes:

[0069] Step b1, obtaining the query popularity for the data indicator.

[0070] Step b2: filter the data indicators based on the query popularity to obtain key indicators.

[0071] The query heat is characterized by the query frequency of the data indicator, and the query frequency can be determined by the number of queries, that is, the more queries there are, the higher the query frequency and the higher the query heat. In the process of obtaining key indicators, data indicators related to the number of queries are extracted from the data table, and the query heat of the data indicator is determined in combination with the number of queries. The query heat is compared with the preset heat threshold, and the data indicators with query heat greater than the preset heat threshold are screened out, and the data indicators with query heat greater than the preset heat threshold are determined as key indicators that characterize the main content of the data table. Among them, the preset heat threshold is the pre-set minimum query heat that characterizes the data indicator as a key indicator.

[0072] In the above implementation, key indicators in the data table are identified by querying popularity to determine the most valuable indicators in the data table, which provides a solid foundation for subsequent data table intent generation and improves the accuracy of data table intent generation.

[0073] In some optional implementations, the above method further includes:

[0074] Step c1, sorting the key indicators according to the query popularity and generating the indicator sorting results;

[0075] Step c2: selecting a preset number of key indicators from the data indicators based on the indicator sorting results.

[0076] The indicators are sorted according to the query popularity corresponding to each key indicator to obtain the indicator sorting results of the key indicators. A preset number of common data indicators are selected based on the indicator sorting results, and the selected common data indicators are used as key indicators to characterize the main content of the data table. Among them, the preset number is set according to the experience value, for example, 20, and the specific value of the preset number is not specifically limited here.

[0077] In the above implementation, data indicators are sorted according to query popularity, and key indicators are selected based on the indicator sorting results to determine the most valuable data indicators in the data table, ensuring the accurate classification of primary attribute indicators and derived attribute indicators in the subsequent determination.

[0078] Step S203: Generate a data table intent based on the association relationship between the data table entity and the key indicator. Please refer to the description of the corresponding steps in the above embodiment for details, which will not be repeated here.

[0079] The method for generating the intent of a data set provided in this embodiment performs entity recognition by analyzing the data table name to realize automatic extraction of data table entities, which is beneficial to improving the recognition efficiency of the data table intent; it performs parsing on data indicators in the data table information to obtain the key indicators of the data table, thereby realizing automatic extraction of key indicators, avoiding the problem of inaccurate recognition of the data table intent due to misjudgment in manual analysis, and enhancing the recognition accuracy of the data table intent.

[0080] In this embodiment, a method for generating an intention of a data set is provided, which can be used in computer devices such as mobile phones, computers, tablet computers, etc. Figure 3 is a flowchart of a method for generating an intention of a data set according to an embodiment of the present disclosure, such as Figure 3 As shown, the process includes the following steps:

[0081] Step S301, obtaining data table information in a data set. Please refer to the relevant description of the corresponding steps in the above embodiment for details, which will not be repeated here.

[0082] Step S302: parse the data table information to extract the data table entities and key indicators corresponding to the data table information. Please refer to the relevant description of the corresponding steps in the above embodiment for details, which will not be repeated here.

[0083] Step S303, generating a data table intent based on the association relationship between the data table entity and the key indicators.

[0084] Specifically, the above step S303 includes:

[0085] Step S3031, based on the association relationship between the data table entity and the key indicator, the key indicator is classified to obtain the main attribute indicator and the derived attribute indicator.

[0086] The primary attribute indicator is the main attribute indicator of the data table; the derived attribute indicator is the secondary attribute indicator of the data table, which is the attribute indicator derived from the primary attribute indicator. Combining the association relationship between the data table entity and the key indicator, the association degree between the data table entity and each key indicator can be determined. According to the size of the association degree, the key indicator can be divided into the primary attribute indicator and the derived attribute indicator.

[0087] In a specific example, the key indicator with the greatest correlation is determined as the primary attribute indicator, and the remaining key indicators are determined as derived attribute indicators. For example, in the "order table", "payment time" is the primary attribute indicator of the data table, while the user, merchant, etc. associated with the order are derived attribute indicators.

[0088] In a specific example, the intent recognition model is used to judge the relationship between the data table entity and the key indicator, and output each key indicator and its attributes. The intent recognition model can distinguish the main attributes and derived attributes of the key indicator according to the prompt description information. Among them, the prompt description information includes system prompt information and preset example prompt information. The system prompt information includes role prompt information, step prompt information, and constraint prompt information. The system prompt information is integrated with the example prompt information to obtain the prompt description information as follows:

[0089] "##Role

[0090] You are an expert in intelligent questions, and are good at judging the relationship between data table entities and key indicators based on data table names and their data table entities.

[0091] ##step

[0092] - Understand user questions and extract data table names and their data table entities

[0093] - Based on the relationship between key indicators and data table entities, distinguish whether the indicator the user is asking about is a primary attribute indicator or a derived attribute indicator

[0094] - Returns the indicator and its properties according to the specified format.

[0095] ##Example

[0096] - Question: {"dataset_name":"Order table","main_entity":"Order","metric":"Order amount"}

[0097] -Answer:{"metric":"order amount","relation":"primary attribute"}

[0098] - Question: {"dataset_name":"Order table","main_entity":"Order","metric":"Merchant"}

[0099] -Answer:{"metric":"merchant","relation":"derived attribute"}

[0100] ##constraint

[0101] - Do not answer questions that are not relevant to extracting dataset entities

[0102] -The answer can only include this indicator and its attributes

[0103] - Answer questions using JSON".

[0104] Step S3032, generating a main intention using the main attribute index, and generating a secondary intention using the derived attributes.

[0105] The main intent represents the main content of the data table and is used to represent the main purpose of the data table, that is, the main query request initiated by the user for the data table. The auxiliary intent represents the query request of the derived attribute index associated with the main attribute index of the data table. The main intent and auxiliary intent of the data table are generated using the generated data table entities, main attribute indexes and derived attribute indexes.

[0106] In a specific example, the data table entity of the "Order Table" is "Order", and the main attribute indicators are indicators associated with "Order", such as payment GMV, live broadcast genre GMV, on-time delivery order volume, refund rate after delivery (order volume), sales volume, delivery order volume, sales in the last 60 days, etc., and the corresponding main graph can be generated in combination with the main attribute indicators as "Order-type indicators (GMV, order volume, delivery, etc.)", that is, the main graph is a query request for the main attribute indicators associated with "Order". Derived attribute indicators may include user name, number of users, merchant name, store name, purchasing user ID, store ID, etc., and the corresponding auxiliary graph generated by the derived attribute indicators is "Indicators associated with orders (users, goods, stores, etc.)".

[0107] In a specific example, the abstract induction ability of the intent recognition model is used to generate the main intent according to the main attribute index, and the auxiliary intent is generated according to the derived attribute index. The generation of the main intent and the auxiliary intent needs to correspond. The prompt description information of the main intent and the auxiliary intent generated by the intent recognition model includes system prompt information and example prompt information, wherein the system prompt information includes role prompt information, step prompt information, knowledge prompt information and constraint prompt information. The specific prompt description information is as follows:

[0108] "##Role

[0109] You are an expert in intelligent problems and are good at abstractly summarizing the main and secondary intentions of data tables.

[0110] ##step

[0111] - Understand user questions, extract data table names and their data table entities, primary attribute indicators, and derived attribute indicators

[0112] -Abstract and summarize the main purpose of the data table based on the data table entities and main attribute indicators

[0113] -Abstract and summarize the auxiliary intent of the data table based on the data table entity and derived attribute indicators

[0114] ##Knowledge

[0115] -The main idea indicates the main purpose of the user using this dataset. For example, the main idea of ​​the "order table" is the query request for the main entity of the order table, that is, the main attribute of "order".

[0116] - The secondary intent is a query request for the derived attributes associated with the primary attribute of this data table.

[0117] ##Example

[0118] - Question: {"dataset_name":"Order Table","main_entity":"Order","primary_metrics":["Payment GMV","Live Genre GMV","On-time Delivery Orders","Refund Rate after Delivery (Order Volume)","Sales Volume","Delivery Order Volume","Sales in the Last 60 Days"],"secondary_metrics":["User Name","Number of Users","Merchant Name","Store Name","Purchasing User ID","Store ID"]}

[0119] -Answer:{"primaryIntention":"Order indicators (GMV, order volume, merchants, shipments, etc.)","secondaryIntention":"Order-related: users, products, merchants (stores), etc."}

[0120] ##constraint

[0121] -Don't answer and abstractly summarize the main and secondary intentions of the data set

[0122] -Answer can only include main and auxiliary images

[0123] - Answer questions using JSON".

[0124] Step S3033, determining the data table intent based on the fusion result of the main intent and the auxiliary intent.

[0125] The main intent and auxiliary intent corresponding to the data table are merged to generate a complete intent description information, which is converted into the data table intent so that the corresponding SQL query statement or data access request can be constructed according to the data table intent to recall the data table.

[0126] Step S304: obtaining a data table query request.

[0127] The data table query request is a query request initiated by the user based on the data indicators required by the actual data demand. Specifically, the computer device can provide an interactive interface, and the user can input the data table query request for the data indicator through the interactive interface. Correspondingly, the computer device can respond to the user's data indicator input operation and obtain the data table query request corresponding to the input operation.

[0128] Step S305: semantically match the data table query request with the data table intent, and recall data based on the semantic matching result.

[0129] Data table intent is mainly used for data table recall of conversational data queries. It semantically matches the data table query request initiated by the user with the primary and secondary intents in the data table intent, and determines the data table corresponding to the data table query request based on the semantic matching results, thereby completing the accurate recall of the data table.

[0130] The method for generating the intent of a data set provided in this embodiment combines the association between data table entities and key indicators to determine the primary attribute indicators and derived attribute indicators associated with the data table entities, thereby generating the primary intent and secondary intent of the data table, ensuring that the generated primary intent and secondary intent can accurately reflect the core content of the data table, thereby improving the accuracy of data recall. At the same time, it is convenient to quickly adapt to the increase in data volume and changes in business needs, provide flexible expansion capabilities, meet dynamic data needs, and enhance the scalability of data table intent generation. Semantic matching is performed on data table query requests and data table intents, and data recall is performed using the semantic matching results, which facilitates the rapid location of data table information required by users and optimizes the query experience.

[0131] In this embodiment, a device for generating the intent of a data set is also provided, and the device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0132] This embodiment provides a device for generating an intention of a data set, such as Figure 4 As shown, including:

[0133] The acquisition module 401 is used to acquire data table information in the data set.

[0134] The parsing module 402 is used to parse the data table information and extract the data table entities and key indicators corresponding to the data table information.

[0135] The intention generation module 403 is used to generate data table intentions based on the association relationship between data table entities and key indicators.

[0136] In some optional implementations, the parsing module 402 includes:

[0137] The entity recognition unit is used to perform entity recognition on the data table name in the data table information to obtain the data table entity.

[0138] The indicator parsing unit is used to parse the data indicators in the data table information to obtain the key indicators of the data table.

[0139] In some optional implementations, the entity identification unit includes:

[0140] The prompt construction subunit is used to construct prompt description information based on preset entity recognition examples.

[0141] The recognition subunit is used to guide the entity recognition of the data table name according to the prompt description information, and generate the data table entity corresponding to the data table name.

[0142] In some optional implementations, the indicator parsing unit includes:

[0143] The query heat acquisition subunit is used to obtain the query heat for the data indicator.

[0144] The filtering subunit is used to filter data indicators based on query popularity to obtain key indicators.

[0145] In some optional embodiments, the above device further includes:

[0146] The sorting module is used to sort the key indicators according to the query popularity and generate the indicator sorting results;

[0147] The selection module is used to select a preset number of key indicators from the data indicators based on the indicator sorting results.

[0148] In some optional implementations, the intention generation module 403 includes:

[0149] The attribute classification unit is used to classify the key indicators based on the association relationship between the data table entities and the key indicators to obtain the main attribute indicators and derived attribute indicators.

[0150] The main and auxiliary intention generating unit is used to generate the main intention using the main attribute index and generate the auxiliary intention using the derived attributes.

[0151] The intent fusion unit is used to determine the intent of the data table based on the fusion results of the main intent and the auxiliary intent.

[0152] In some optional embodiments, the above device further includes:

[0153] The query request acquisition module is used to obtain data table query requests.

[0154] The recall module is used to perform semantic matching between data table query requests and data table intents, and to recall data based on the semantic matching results.

[0155] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0156] The intention generation device of the data set in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above-mentioned functions.

[0157] The device for generating the intent of a data set provided in the present embodiment parses the acquired data table information to determine the data table entities and key indicators corresponding to the data table information, and determines the data table intent in combination with the association between the data table entities and the key indicators, thereby realizing the automatic extraction of the data table intent, reducing manual intervention, and improving the recognition efficiency of the data table intent. At the same time, it avoids the errors introduced by manual analysis, improves the generation accuracy of the data table intent, and thus improves the accuracy of data recall based on the data table intent.

[0158] The present disclosure also provides a computer device having the above Figure 4 Intent generation device for the dataset shown.

[0159] See also Figure 5 , Figure 5 is a schematic diagram of a computer device provided by an optional embodiment of the present disclosure, such as Figure 5 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 10 is taken as an example.

[0160] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0161] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0162] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0163] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0164] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 5 The example of connecting through bus is taken in the following.

[0165] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0166] The computer device also includes a communication interface, which is used for the computer device to communicate with other devices or a communication network.

[0167] The embodiments of the present disclosure also provide a computer-readable storage medium. The above-mentioned method according to the embodiments of the present disclosure can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium and downloaded through a network, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0168] A part of the present disclosure may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present disclosure through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0169] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for generating intent from a data set, characterized in that: The method comprises: Get the data table information in the data set; Parsing the data table information to extract data table entities and key indicators corresponding to the data table information; Based on the association relationship between the data table entity and the key indicator, a data table intent is generated.

2. The method according to claim 1, characterized in that The parsing of the data table information to extract data table entities and key indicators corresponding to the data table information includes: Perform entity recognition on the data table name in the data table information to obtain a data table entity; The data indicators in the data table information are parsed to obtain key indicators of the data table.

3. The method according to claim 2, characterized in that The step of parsing the data indicators in the data table information to obtain key indicators of the data table includes: Obtaining query popularity for the data indicator; The data indicators are filtered based on the query popularity to obtain the key indicators.

4. The method according to claim 3, characterized in that Also includes: Sorting the data indicators according to the query popularity to generate indicator sorting results; A preset number of the key indicators are selected from the data indicators based on the indicator sorting result.

5. The method according to claim 2, characterized in that: The performing entity recognition on the data table name in the data table information to obtain the data table entity includes: Construct prompt description information based on preset entity recognition examples; The entity recognition of the data table name is guided according to the prompt description information, and the data table entity corresponding to the data table name is generated.

6. The method according to claim 1, characterized in that The generating of the data table intent based on the association relationship between the data table entity and the key indicator includes: Based on the association relationship between the data table entity and the key indicator, the key indicator is classified to obtain a primary attribute indicator and a derived attribute indicator; Generate a main intention using the main attribute index, and generate a secondary intention using the derived attribute; Based on the fusion result of the main intent and the auxiliary intent, the data table intent is determined.

7. The method according to claim 1 or 6, characterized in that: Also includes: Get data table query request; Semantic matching is performed between the data table query request and the data table intent, and data is recalled based on the semantic matching result.

8. A device for generating an intention of a data set, characterized in that: The device comprises: The acquisition module is used to obtain the data table information in the data set; A parsing module, used to parse the data table information and extract data table entities and key indicators corresponding to the data table information; The intent generation module is used to generate a data table intent based on the association relationship between the data table entity and the key indicator.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for generating intent of a data set according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for generating an intention of a data set according to any one of claims 1 to 7.

11. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the method for generating an intention of a data set according to any one of claims 1 to 7.

Images