Information display method and device, electronic device, storage medium and program product

Abstract

The present disclosure provides an information display method and device, electronic device, storage medium and program product, which relate to the fields of computer technology, big data technology and data processing technology, and in particular to data analysis and visualization. The information display method includes: in response to receiving a query request for a target subject, querying a data set for the target subject according to the target intent indicated by the query request, obtaining target mapping information, wherein the target mapping information represents a mapping relationship between a data mapping and a data item, and both the data mapping and the data item are related to the target intent; verifying an initial statement generated based on the target mapping information according to a preset verification rule to obtain a verification result, wherein the preset verification rule defines the query authority of each data mapping and data item; determining a query statement based on the initial statement according to the verification result; and displaying the query result obtained by executing the query statement.

Description

Technical Field

[0001] The present disclosure relates to the fields of computer technology, big data technology, and data processing technology, and in particular to data analysis and visualization, and more specifically to an information display method and device, electronic device, storage medium, and program product. Background Art

[0002] With the advent of the big data era, the demand for data management and analysis is becoming increasingly complex. The amount and types of data involved in various industries are constantly increasing, and the coupling complexity of data is also becoming increasingly higher. For non-R&D users, independently completing data analysis is extremely challenging. Summary of the Invention

[0003] The present disclosure provides an information display method and device, an electronic device, a storage medium, and a program product.

[0004] According to one aspect of the present disclosure, there is provided an information display method, comprising: in response to receiving a query request for a target subject, querying a data set for the target subject according to the target intent indicated by the query request to obtain target mapping information, wherein the target mapping information represents a mapping relationship between a data mapping and a data item, and both the data mapping and the data item are related to the target intent; verifying an initial statement generated based on the target mapping information according to a preset verification rule to obtain a verification result, wherein the preset verification rule defines the query authority of each of the data mapping and the data item; determining a query statement based on the initial statement according to the verification result; and displaying the query result obtained by executing the query statement.

[0005] According to another aspect of the present disclosure, an information display device is provided, including: a query module for, in response to receiving a query request for a target topic, querying a data set for the target topic according to the target intent indicated by the query request to obtain target mapping information, wherein the target mapping information represents a mapping relationship between a data mapping and a data item, and both the data mapping and the data item are related to the target intent; a verification module for verifying an initial statement generated based on the target mapping information according to preset verification rules to obtain a verification result, wherein the preset verification rules define the query permissions of the data mapping and the data item respectively; a first determination module for determining a query statement based on the initial statement according to the verification result; and a display module for displaying the query result obtained by executing the query statement.

[0006] According to another aspect of the present disclosure, an artificial intelligence agent is provided, which is configured to execute an information presentation method.

[0007] According to another aspect of the present disclosure, an electronic device is provided, comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0008] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program or instructions are stored. When the computer program or instructions are executed by a processor, the steps of the above method are implemented.

[0009] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program or instructions, which implement the steps of the above method when executed by a processor.

[0010] It should be understood that the contents described in this section are not intended to identify 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 readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0012] Figure 1 The system architecture to which the information display method according to an embodiment of the present disclosure can be applied is schematically shown;

[0013] Figure 2 The following schematically shows a flow chart of an information display method according to an embodiment of the present disclosure;

[0014] Figure 3 An example diagram schematically illustrates a process of constructing a data set for a target topic according to an embodiment of the present disclosure;

[0015] Figure 4 The following is a schematic diagram schematically illustrating an example of a process for obtaining a target intention according to an embodiment of the present disclosure;

[0016] Figure 5 Schematically illustrates an example process of querying a data set for a target subject according to a target intent indicated by a query request to obtain target mapping information according to an embodiment of the present disclosure;

[0017] Figure 6 The following schematically illustrates an example of an information display process according to an embodiment of the present disclosure;

[0018] Figure 7AAn example diagram of an information display interface for data query according to an embodiment of the present disclosure is schematically shown;

[0019] Figure 7B Schematically shows an example schematic diagram of an information display interface for data query according to another embodiment of the present disclosure;

[0020] Figure 7C An example diagram of an information display interface for topic management according to an embodiment of the present disclosure is schematically shown;

[0021] Figure 7D Schematically shows an example schematic diagram of an information display interface for topic management according to another embodiment of the present disclosure;

[0022] Figure 8 A block diagram of an information display device according to an embodiment of the present disclosure is schematically shown;

[0023] Figure 9 Schematically shows a structural block diagram of an intelligent agent of a large model according to an embodiment of the present disclosure; and

[0024] Figure 10 A block diagram schematically shows an electronic device suitable for implementing the information presentation method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present disclosure. In the detailed description below, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0026] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise," "include," etc. used herein indicate the presence of the features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0027] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0028] When expressions such as "at least one of A, B, and C, etc." are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0029] Data management platforms host a vast amount of enterprise data resources, and daily, numerous users use them to search for data, apply for permissions, and access data. During the operation of the data management platform, numerous non-R&D users, such as those in product development, operations, and sales, inquire about data query and analysis. For these non-R&D users, independently completing data analysis can be extremely challenging.

[0030] Although the NL2SQL (Natural Language to SQL) field has made significant progress in converting natural language instructions into executable SQL (Structured Query Language) scripts for data query and processing, achieving full automation in the broader data science pipeline (i.e., including data query, analysis, visualization, and reporting) remains a complex challenge. For example, current data analysis tasks still require manual participation, that is, they are completed through R&D paradigms such as NL2SQL + tool set calls + manual task splicing, which is time-consuming and labor-intensive.

[0031] In one example, the data analysis method may adopt at least one of the following: implementing data analysis based on a fine-tuned NL2SQL dedicated model and implementing data analysis based on a general large model or a data analysis agent product.

[0032] Data analysis based on fine-tuned NL2SQL specialized models involves fine-tuning proprietary models to improve the accuracy of NL2SQL tasks. However, this approach requires collecting sufficiently large, cross-domain, generalized datasets to adapt to each data domain (for example, databases for grocery stores, wholesalers, and corporate finance), which consumes significant resources. Furthermore, improper fine-tuning can lead to catastrophic forgetting, compromising the adaptability of large language models (LLMs).

[0033] Data analysis based on general-purpose big models or data analysis agents involves generating executable SQL statements through natural language interaction using data analysis agents such as instruction engineering, general-purpose big models, or code interpreters. However, these general-purpose big models or code interpreters require user supervision to generate scripts or manipulate uploaded datasets for potential data analysis. Human intervention is still required between each interactive step (for example, checking the correctness of the generated script and copying and pasting the script into the local execution environment for further data manipulation), preventing full automation of the data science pipeline.

[0034] To this end, embodiments of the present disclosure propose an information display scheme. For example, in response to receiving a query request for a target topic, a query is performed on a data set for the target topic according to the target intent indicated by the query request to obtain target mapping information, wherein the target mapping information represents a mapping relationship between a data mapping and a data item, and both the data mapping and the data item are related to the target intent; an initial statement generated based on the target mapping information is verified according to preset verification rules to obtain a verification result, wherein the preset verification rules define the query permissions of the data mapping and the data item; a query statement is determined based on the initial statement according to the verification result; and the query result obtained by executing the query statement is displayed.

[0035] According to the embodiments of the present disclosure, by responding to query requests for target topics and performing precise queries within relevant data sets based on the target intent indicated by the query requests, target mapping information can be quickly located and retrieved, ensuring the accuracy and relevance of data retrieval. Furthermore, permissions are verified against the initial statement using pre-set validation rules, enhancing data access security. Query statements generated based on the validation results can be accurately executed, displaying query results that meet the user's intent, thereby improving user experience and data processing efficiency.

[0036] In the technical solution of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0037] In the technical solution of the present invention, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0038] Figure 1 The system architecture to which the information display method according to the embodiment of the present disclosure can be applied is schematically shown. It should be noted that, Figure 1 The examples shown are merely examples of system architectures to which the embodiments of the present disclosure may be applied, to help those skilled in the art understand the technical content of the present disclosure, but do not mean that the embodiments of the present disclosure may not be used in other devices, systems, environments or scenarios.

[0039] like Figure 1 As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0040] The user may use at least one of the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0041] The first terminal device 101 , the second terminal device 102 , and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0042] The server 105 may be a server that provides various services, such as a background management server (for example only) that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process received data such as user requests, and feed back processing results (e.g., web pages, information, or data obtained or generated based on user requests) to the terminal devices.

[0043] It should be noted that the information display method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the information display device provided in the embodiment of the present disclosure can generally be set in the server 105. The information display method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the information display device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0044] Alternatively, the information display method provided in the embodiment of the present disclosure may also be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103, or may also be executed by another terminal device different from the first terminal device 101, the second terminal device 102, or the third terminal device 103. Accordingly, the information display device provided in the embodiment of the present disclosure may also be provided in the first terminal device 101, the second terminal device 102, or the third terminal device 103, or may be provided in another terminal device different from the first terminal device 101, the second terminal device 102, or the third terminal device 103.

[0045] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0046] It should be noted that the sequence numbers of the operations in the following method are only used to indicate the operation for the purpose of description, and should not be regarded as indicating the order in which the operations should be performed. Unless explicitly stated, the method does not need to be performed in the order shown.

[0047] Figure 2 The flowchart of the information display method according to the embodiment of the present disclosure is schematically shown.

[0048] like Figure 2 As shown, the information display method 200 includes operations S210 to S240.

[0049] In operation S210, in response to receiving a query request for a target subject, a query is performed on the data set used for the target subject according to the target intent indicated by the query request to obtain target mapping information, wherein the target mapping information represents a mapping relationship between data mapping and data items, and both the data mapping and the data items are related to the target intent.

[0050] In operation S220 , the initial statement generated based on the target mapping information is verified according to a preset verification rule to obtain a verification result, wherein the preset verification rule defines the query authority of the data mapping and the data item.

[0051] In operation S230 , according to the verification result, a query statement is determined based on the initial statement.

[0052] In operation S240 , the query results obtained by executing the query statement are displayed.

[0053] A query request refers to a user-initiated query operation targeting a specific target topic. After receiving the query request, an intent analysis can be performed on the query request to determine the target intent indicated by the query request. The target intent refers to the specific purpose or requirement expressed in the query request, which is used to guide the direction and scope of the query. The specific method of intent analysis can be configured according to actual business needs and is not limited here. For example, the specific method of intent analysis can include at least one of the following: a rule-based method, a context-based method, a machine learning-based method, a deep learning-based method, and a large model-based method.

[0054] Rule-based methods use predefined rules to identify keywords or phrases in query requests and map them to specific intents. Context-based methods consider the context of the query request, such as previous conversation content and the user's historical behavior, to more accurately determine intent. Machine learning-based methods use labeled training datasets to extract relevant features through feature engineering, and then use machine learning algorithms (such as logistic regression, random forests, support vector machines, etc.) to classify intent. Deep learning-based methods use deep neural networks to automatically extract features and classify intent. Large model-based methods use pre-trained language models to identify the intent of query requests.

[0055] A data set is a collection of all data related to a target topic. Each data item in a data set can be used for query operations. The specific form of a data set can be configured based on actual business needs and is not limited here. For example, a data set can include at least one of the following: a distributed file system, a database, a data lake, and a cloud storage service.

[0056] In an embodiment of the present disclosure, before querying the data set for the target subject, the user's consent or authorization may be obtained. For example, before operation S210, a request to query the data set for the target subject may be issued to the user. If the user consents or authorizes the query of the data set for the target subject, operation S210 is performed.

[0057] After obtaining the target intent, you can query the data set for the target subject according to the target intent to obtain the target mapping information. The target mapping information is used to characterize the mapping relationship between the data mapping and the data item, and may include at least one data mapping related to the target intent, and at least one data item related to the target intent in each data mapping. Data mapping refers to a representation used to reflect the correspondence between data. The specific form of the data mapping can be configured according to actual business needs and is not limited here. For example, the form of the data mapping may include at least one of the following: list form, dictionary form, and array form, etc. A data item refers to a single data element or field in a data mapping, that is, the smallest data unit. For example, when the data mapping is in list form, the data item may be a row or column. Alternatively, when the data mapping is in dictionary form, the data item may be a key-value pair. Alternatively, when the data mapping is in array form, the data item may be an element.

[0058] After obtaining the target mapping information, it can be converted into an executable initial statement to enable automated interaction between the natural language question-answering system and the data set. For example, if the data set is in the form of a database, the initial statement can be an SQL statement. After obtaining the initial statement, it can be validated according to pre-set validation rules to obtain a validation result. The pre-set validation rules define the query permissions for the data mapping and data items, ensuring that the query operation complies with predetermined permission requirements.

[0059] After obtaining the verification result, a query statement can be determined based on the initial statement according to the verification result. The verification result indicates whether the initial statement complies with the query permissions of the data mapping and the data item. For example, if the verification result indicates that the initial statement complies with the query permissions of the data mapping and the data item, the initial statement can be determined as the query statement. Alternatively, if the verification result indicates that the initial statement does not comply with the query permissions of the data mapping and the data item, the initial statement can be processed again to obtain a query statement.

[0060] After obtaining the query statement, the query statement can be executed and the query results can be displayed. The display format of the query results can be configured according to actual business needs and is not limited here. For example, the display format of the query results may include at least one of the following: table format, chart format, and text format. The table format can be used to display structured query results. The chart format may include a bar chart, a line chart, a pie chart, and a scatter chart. The bar chart is used to show the comparison of data of different categories. The line chart is used to show the trend of data changes over time. The pie chart is used to show the proportion of each part to the whole. The scatter chart is used to show the relationship between two variables. The text format is used to summarize the query results.

[0061] This application can be applied in the field of data analysis, and can automate end-to-end data science tasks such as data query, data analysis, chart visualization, and report generation according to the target intent indicated by the query request.

[0062] According to the embodiments of the present disclosure, by responding to query requests for target topics and performing precise queries within relevant data sets based on the target intent indicated by the query requests, target mapping information can be quickly located and retrieved, ensuring the accuracy and relevance of data retrieval. Furthermore, permissions are verified against the initial statement using pre-set validation rules, enhancing data access security. Query statements generated based on the validation results can be accurately executed, displaying query results that meet the user's intent, thereby improving user experience and data processing efficiency.

[0063] Figure 3 An example diagram schematically illustrates a process of constructing a data set for a target topic according to an embodiment of the present disclosure.

[0064] like Figure 3 As shown, in 300, the data collection for the target subject can be understood as an offline stage, which is equivalent to a pre-processing link and only needs to be executed once.

[0065] The candidate set for the target topic includes at least one first candidate data mapping 301. Structural information 303 can be obtained by semantically modeling the metadata 302 of each of the at least one first candidate data mapping 301. Semantic modeling includes constructing indicators, dimensions, and views. The first candidate data mapping 301 can refer to a data table, and the metadata 302 can refer to the schema of the data table. For example, the metadata 302 can include at least one of the following: table name, table description, data source connection information, SQL engine type, database, field name, field type, Chinese name, and field description. Structural information 303 can refer to a view that represents the descriptive information of each first candidate data mapping 301 for the target topic, as well as the associations between the first candidate data mappings 301.

[0066] By performing semantic modeling on each metadata 302 to obtain structural information 303, multiple reuses can be achieved through a single definition, and the data caliber is easy to change. In addition, SQL code generation can be simplified, thereby solving the problem of large differences in results caused by different semantics or semantic ambiguity when users inquire, solidifying the data caliber definition at the semantic layer, and simplifying user semantic expression.

[0067] Initial object 304 can refer to a query-SQL example pair and can include question text 304_1 and historical statements 304_2. In one example, the specific source of initial object 304 can be configured based on actual business needs and is not limited here. For example, initial object 304 can be provided by the business side, in which case the input is a natural language question and the output is SQL. Alternatively, initial object 304 can be provided by user feedback and SQL corrections, in which case the input is a natural language question and incorrect SQL, and the output is a natural language question and correct SQL.

[0068] Alternatively, the initial object 304 can be obtained through SQL2NL. In this case, the input is SQL and the output is a natural language question. It should be noted that the seed data used to generate the initial object 304 is added through reverse engineering based on SQL2NL. SQL2NL reverse engineering refers to the process of generating natural language based on SQL and Schema information.

[0069] Compared to not providing seed data, the seed data generated by SQL2NL improves the execution accuracy of NL2SQL tasks by approximately 40%. This solution achieves the business benefit of eliminating the need for users to provide costly seed data after creating a new subject domain. We reverse-generate natural language data based on the user's SQL and then construct seed data, significantly reducing user access costs.

[0070] By performing semantic enhancement on question text 304_1, enhanced text 305_1 is obtained. This data enhancement is performed on the semantics of the user's question based on the table's schema information, and the enhanced results are added to the memory of the large model. Therefore, by adding incorrect SQL examples and their data enhancement results to the model memory, the accuracy of the NL2SQL task can be effectively improved.

[0071] By performing syntax tree parsing on historical statement 304_2, candidate mapping information 305_2 is obtained. Candidate mapping information 305_2 may include at least one second candidate data mapping and candidate data items for each second candidate data mapping. In one example, candidate mapping information 305_2 refers to schema_linking information. For example, candidate mapping information 305_2 may be [{"FIELD":["shop_id","profits","product_type","cities","year"],"TABLE":"product"}], which indicates the candidate data items "shop_id", "profits", "product_type", "cities", and "year" in the second candidate data mapping "product".

[0072] On this basis, the enhanced text 305_1, historical sentences 304_2, and candidate mapping information 305_2 can be determined as candidate objects 305, and a data set 306 for the target topic can be constructed based on the structural information 303 and the multiple candidate objects 305. In addition, the structural information 303 can also have a corresponding vectorized representation, and the enhanced text 305_1 can also have a corresponding vectorized representation.

[0073] According to the embodiments of the present disclosure, by performing semantic modeling on the metadata of each first candidate data mapping in the candidate set, it is possible to deeply understand the inherent structure and semantic relationships of the data, thereby generating accurate structural information. In addition, the generation of enhanced text is achieved by semantically enhancing the question text of the initial object within the target topic, ensuring the semantic richness and accuracy of the text. By performing syntax tree parsing on historical sentences to obtain candidate mapping information, the analysis accuracy of the sentence structure is improved, which helps to improve the accuracy and efficiency of data processing.

[0074] Figure 4 An example diagram schematically illustrates a process of obtaining a target intent according to an embodiment of the present disclosure.

[0075] like Figure 4 As shown in step 400, in response to receiving a query request 401, a large model 404 may be used to perform intent understanding on the query request 401 to obtain an initial intent 402. Intent understanding refers to the process of analyzing the query request 401 to determine the purpose or intent of the object. After obtaining the initial intent 402, operation S410 may be performed.

[0076] In operation S410, does the initial intent 402 satisfy the query conditions? In one example, the query conditions can be configured according to actual business needs and are not limited here. For example, the query conditions can include whether the initial intent 402 is related to the subject domain. Alternatively, the query request can also include whether the initial intent 402 includes a drawing type.

[0077] If so, the initial intent 402 can be directly determined as the target intent 403. If not, the macro model 404 can be used to conduct multiple rounds of interactive dialogues with the subject, guiding the subject to supplement the intent so that the query request 401 meets the query conditions. Multi-round interactive dialogues refer to using the macro model to conduct multiple rounds of dialogues with the subject, gradually collecting the required information and making appropriate responses based on the context.

[0078] On this basis, the target intent 403 can be determined based on the query request 401 and the dialogue information 405 obtained through multiple rounds of interactive dialogue. In another example, the subject can also supplement the intent through follow-up questions. On this basis, the target intent 403 can be determined based on the pre-question, the follow-up question, the vectorization of the pre-question and follow-up questions, and the integrated question.

[0079] According to the embodiments of the present disclosure, through the big model technology, when the initial intention does not meet the query conditions, multiple rounds of interactive dialogues can be conducted with the user to guide the user to supplement the intention, thereby ensuring that the query request meets the conditions. This not only improves the accuracy of the query, but also optimizes the user interaction experience and enhances the pertinence and effectiveness of the query.

[0080] Figure 5 An example diagram of a process of querying a data set for a target subject according to a target intent indicated by a query request and obtaining target mapping information is schematically shown according to an embodiment of the present disclosure.

[0081] like Figure 5 As shown, in 500 , in response to receiving a query request, a query may be performed in a data set 520 for a target subject according to a target intent 510 indicated by the query request to obtain target mapping information 560 .

[0082] Data set 520 refers to a collection of all data related to the target subject. Each data item in data set 520 can be used for query operations. Data set 520 may include structure information 521 and multiple candidate objects. The process of obtaining target mapping information 560 may include a multi-way recall phase and a pattern matching phase. The multi-way recall phase may be used to recall at least one data mapping 523 that matches the target intent 510. The pattern matching phase may be used to associate the at least one data mapping 523 with a specific target data mapping and the target data item of each target data mapping to obtain target mapping information 560.

[0083] In the multi-way recall phase, the structural information 521 can be recalled according to the target intent 510 to obtain at least one data mapping 523. Recall refers to the process of retrieving structural information related to the target intent from a data set. The purpose of recall is to obtain at least one data mapping 523.

[0084] According to the embodiments of the present disclosure, by dividing a data set into structural information and multiple candidate objects, the system can quickly recall relevant structural information based on the target intent, thereby obtaining at least one data mapping, thereby improving the pertinence and speed of data retrieval. Furthermore, by conducting in-depth matching of multiple candidate objects based on these data mappings to determine the final target mapping information, the scope of unnecessary data retrieval is reduced, the accuracy of matching is improved, and the relevance and reliability of the query results are ensured.

[0085] In one example, the structure information 521 represents the description information of each first candidate data mapping for the target subject, and the association relationship between the first candidate data mappings. The association relationship refers to the logical or semantic connection between the first candidate data mappings.

[0086] During the process of recalling at least one data mapping 523 from structural information 521, an indicator 522 may be extracted from target intent 510. Based on the association relationship, the description information of each first candidate data mapping may be matched according to the indicator 522 to obtain a first number of data mappings 523. Indicator 522 refers to a specific standard or parameter extracted from target intent 510 and is used to evaluate the relevance of the description information of each first candidate data mapping.

[0087] For example, if the query "What is the online bug closure rate for Department A, Group B, Business Line C in 2023?" is closer to "What is the revenue for Department A, Group B, Business Line C in 2023?" than to "What is the online bug closure rate for Department A, Group B, Business Line C in 2022?", it's easy for data mappings with key indicators to be overlooked. Therefore, by using the large model to extract indicator 522 for target intent 510, this indicator 522 is used to recall examples most similar to target intent 510.

[0088] In the process of recalling at least one data mapping 523 from the structural information 521, an intent feature 524 of the target intent 510 can be determined, and based on the association relationship, the description information of each first candidate data mapping can be matched according to the intent feature to obtain a second number of data mappings 523. The intent feature 524 refers to a specific attribute or characteristic of the target intent 510.

[0089] By merging the first number of data mappings 523 recalled based on the indicator 522 and the second number of data mappings 523 recalled based on the intent feature 524 , the accuracy of subsequent query statement generation can be improved.

[0090] It should be noted that during the multi-channel recall phase, each channel of the recall must maintain as much independence as possible to ensure that the target data mapping appears in at least one data mapping 523. In one example, after obtaining at least one data mapping 523, the effectiveness of the multi-channel recall can also be evaluated. The specific evaluation method can be configured based on actual business needs and is not limited here. For example, the evaluation method may include determining whether the target data mapping is within the at least one data mapping 523. Alternatively, the evaluation method may include determining that the target data mapping is positioned as far forward as possible within the at least one data mapping 523.

[0091] According to the embodiments of the present disclosure, by matching the description information based on the association relationship and the indicators in the target intent, a first number of data mappings can be accurately identified, ensuring the targeted and accurate data processing. By matching the description information with the intent characteristics of the target intent, a second number of data mappings is obtained, which not only optimizes the data retrieval process but also improves the matching accuracy and ensures the relevance and reliability of the query results.

[0092] When querying a database using natural language, the model input includes natural language query statements and description information. The description information can be understood as the database schema description information, which is a string text that maps all data in the database according to a certain serialization format.

[0093] Considering that actual natural language queries are often only related to partial data mappings and some data items, the remaining schema information of redundant input may have a negative impact. Therefore, a pattern matching stage can be added as a precursor to query statement generation to alleviate the problems caused by redundant schema input.

[0094] In the pattern matching phase, multiple candidate objects can be matched based on at least one data mapping 523 to determine target mapping information 560. For example, the capabilities of the large model interface can be used to complete pattern link matching relationships from natural language to partial data mapping and partial data items.

[0095] In one example, a candidate object may include enhanced text and candidate mapping information 530. The candidate mapping information 530 may include at least one second candidate data mapping 540 and a candidate data item 550 for each second candidate data mapping 540. The second candidate data mapping 540 refers to a data mapping in the candidate mapping information, and each second candidate data mapping 540 has at least one candidate data item. A candidate data item refers to a data item associated with each second candidate data mapping 540.

[0096] After obtaining at least one data mapping 523, multiple enhanced texts can be matched based on the target intent 510 to obtain at least one target object. Based on the candidate mapping information for each of the at least one data mapping and the at least one target object, M target data mappings and N target data items for each target data mapping are determined. For example, the mapping relationships between target data mapping 540_1 and data item 550_2, data item 550_3 and data item 550_X-1, and the mapping relationships between target data mapping 540_2 and data items 550_1 and 550_2 can be determined. Based on this, target mapping information 560 can be determined based on the M target data mappings and the N target data items for each target data mapping.

[0097] According to the embodiments of the present disclosure, by matching augmented text according to target intent and rapidly identifying at least one target object, the accuracy of data retrieval is improved while also optimizing data processing efficiency. Furthermore, by determining the target data mapping and its respective target data item based on candidate mapping information for at least one data mapping and the target object, this multi-level matching and determination mechanism enhances the accuracy and flexibility of data mappings, providing strong technical support for the efficient management and retrieval of complex datasets.

[0098] Figure 6 An example diagram of an information display process according to an embodiment of the present disclosure is schematically shown.

[0099] like Figure 6As shown, in 600, in response to receiving a query request for a target subject from object 610, an intent analysis is performed on the query request to obtain an initial intent, and operation S610 is performed.

[0100] In operation S610, it is determined whether the initial intent satisfies the query conditions. If so, the initial intent can be determined as the target intent 620. If not, the large model can be used to conduct multiple rounds of interactive dialogue with the subject 610 to guide the subject 610 to supplement the intent so that the query request satisfies the query conditions. The target intent 620 is then determined based on the query request and the dialogue information obtained through the multiple rounds of interactive dialogue.

[0101] After obtaining the target intent 620, a multi-way recall phase may be performed, i.e., recalling structural information in the data set based on the target intent 620 to obtain at least one data mapping 630. After obtaining the at least one data mapping 630, a pattern matching phase may be performed, i.e., matching multiple candidate objects in the data set based on the at least one data mapping 630 to determine target mapping information 640.

[0102] In one example, target mapping information 640 may include M target data mappings and N target data items for each target data mapping. Based on the M target data mappings and the N target data items for each target data mapping, an initial statement 650 may be translated. The initial statement 650 is a preliminary query statement obtained by integrating the M target data mappings and the N target data items for each target data mapping. To mitigate potential model illusions, the data analysis agent may perform self-reflection and error correction on the generated initial statement 650.

[0103] In response to the initial statement 650 passing the syntax tree check, a permission check is performed on the initial statement 650 based on the query permissions of each target data mapping and each target data item, resulting in a verification result. Syntax tree checking refers to performing grammatical analysis on the initial statement 650 to ensure the correctness of its grammatical structure. Each target data mapping and each target data item has preset query permissions, which are used to determine whether the initial statement 650 meets predetermined security and permission requirements. Based on this, a query statement 660 can be determined based on the initial statement 650 according to the verification result.

[0104] In one example, if the verification result indicates that the initial statement 650 passes verification, the initial statement 650 is determined as the query statement 660. Alternatively, if the verification result indicates that the initial statement 650 fails verification, the initial statement 650 is corrected to obtain the query statement 660.

[0105] For example, problems with the initial statement 650 may include a mismatch between the initial statement 650 and the target mapping information 640 due to model illusions. To address this, a self-reflection instruction can be set. That is, when a mismatch between the generated initial statement 650 and the target mapping information 640 is detected, the large model can be introspected and regenerated into a query statement 660 to improve execution accuracy.

[0106] Alternatively, the problem with the initial statement 650 may include syntax errors that prevent the generated initial statement 650 from being executed correctly. To address this, table-level authentication can be performed for each target data mapping, and field-level authentication can be performed for each target data item. After authentication, different SQL engines can be invoked for execution, ensuring a secure and controllable query process.

[0107] According to the embodiments of the present disclosure, by generating an initial statement based on M target data mappings and their respective N target data items, the system can ensure the initial accuracy and relevance of the query statement. Through syntax tree checking and permission verification, not only is the syntactic correctness of the initial statement verified, but the query permissions for each data mapping and data item are also ensured to comply with pre-set rules, thereby enhancing data access security. Furthermore, by determining the query statement based on the initial statement according to the verification results, the efficiency and accuracy of query statement generation are improved.

[0108] After obtaining the query statement 660, the query results 670 obtained by executing the query statement 660 can be displayed. The display format of the query results 670 can be configured according to actual business needs and is not limited here. For example, the display format of the query results 670 can include at least one of the following: table format, chart format, and text format.

[0109] In one example, the chart type for the query chart can be determined based on the target intent 620 of the query request and the execution result obtained by executing the query statement 660. The script for generating a chart of the chart type is executed to display the query chart. For example, a table format 671 can be used to display structured query results. Chart formats can include bar charts, line charts, pie charts, and scatter charts. A bar chart 672 can be used to display comparisons of data of different categories. A line chart 673 can be used to display trends of data changes over time. A pie chart 674 can be used to display the proportion of each part to the whole. A scatter chart is used to display the relationship between two variables.

[0110] In another example, a query text may be generated and displayed based on the execution result. The text form 675 may be used to summarize the query result.

[0111] According to an embodiment of the present disclosure, based on the target intent of the query request and the execution result obtained by executing the query statement, the system intelligently determines the chart type used for the query chart, ensures that the chart form matches the data content, and improves the efficiency of information communication. By executing the script that generates the chart type, the system can automatically display the query chart, enhancing the intuitiveness and comprehensibility of the data. In addition, by generating and displaying the query text based on the execution result, the system ensures a detailed description and explanation of the data, meets the user's query needs for data details, and not only optimizes the diversity and accuracy of data display, but also improves the user experience and the flexibility of data processing.

[0112] Figure 7A An example diagram of an information display interface for data query according to an embodiment of the present disclosure is schematically shown.

[0113] like Figure 7A As shown, in 700A, the process of data query is described by taking the target subject as a chain supermarket 702 as an example.

[0114] During the data query process, the user can select the data query control 701 on the information display interface, enter a natural language question for the target topic in the input box 704, and click the query button 705 to execute the data query operation.

[0115] It should be noted that the display box 703 can display data dimensions for the target topic to help the user select any dimension to input a natural language question.

[0116] Figure 7B An example schematic diagram of an information display interface for data query according to another embodiment of the present disclosure is schematically shown.

[0117] like Figure 7B As shown in 700B, the process of information display is described by taking the target subject as a chain supermarket 702 as an example.

[0118] The natural language question 706 input by the user may be "Which store has the highest total profit in 'daily necessities' among all stores in 2022?" The agent may execute the information display method provided by the present disclosure to obtain a query statement and display the query result 704 obtained by executing the query statement.

[0119] Figure 7C An example diagram of an information display interface for topic management according to an embodiment of the present disclosure is schematically shown.

[0120] like Figure 7C As shown, in 700C, the process of theme management is explained by taking the target theme of chain supermarket 702 as an example.

[0121] During theme management, the user may select theme management control 708 on the information display interface and then select data management control 709 on the information display interface to manage data mapping 710. Management may include deletion, addition, and modification.

[0122] Figure 7D An example schematic diagram of an information display interface for topic management according to another embodiment of the present disclosure is schematically shown.

[0123] like Figure 7D As shown in 700D, the sample management control 711 in the information display interface can be selected to implement management of candidate objects.

[0124] The above are merely exemplary embodiments, but are not limited thereto. Other information display methods known in the art may also be included, as long as they can display query results that meet the user's intent, thereby improving data processing efficiency.

[0125] Figure 8 A block diagram of an information display device according to an embodiment of the present disclosure is schematically shown.

[0126] like Figure 8 As shown, the information display device 800 may include a query module 810 , a verification module 820 , a first determination module 830 and a display module 840 .

[0127] The query module 810 is used to respond to receiving a query request for a target topic, query the data set used for the target topic according to the target intent indicated by the query request, and obtain target mapping information, wherein the target mapping information represents the mapping relationship between the data mapping and the data item, and both the data mapping and the data item are related to the target intent.

[0128] The verification module 820 is used to verify the initial statement generated based on the target mapping information according to preset verification rules to obtain a verification result, wherein the preset verification rules define the query permissions of the data mapping and the data items.

[0129] The first determination module 830 is configured to determine a query statement based on the initial statement according to the verification result.

[0130] The display module 840 is used to display the query results obtained by executing the query statement.

[0131] According to an embodiment of the present disclosure, a data set includes structural information and a plurality of candidate objects.

[0132] According to an embodiment of the present disclosure, the query module 810 may include a recall submodule and a matching submodule.

[0133] The recall submodule is used to recall the structural information according to the target intention and obtain at least one data mapping.

[0134] The matching submodule is configured to match multiple candidate objects based on at least one data mapping to determine target mapping information.

[0135] According to an embodiment of the present disclosure, the structural information represents the description information of each first candidate data mapping for the target subject, and the association relationship between the first candidate data mappings.

[0136] According to an embodiment of the present disclosure, the recall submodule may include a first matching unit and a second matching unit.

[0137] The first matching unit is configured to match the description information of each first candidate data mapping based on the association relationship and the indicator extracted from the target intent to obtain a first number of data mappings.

[0138] The second matching unit is used to match the description information of each first candidate data mapping based on the association relationship and according to the intention characteristics of the target intention to obtain a second number of data mappings.

[0139] According to an embodiment of the present disclosure, the candidate object includes enhanced text and candidate mapping information, and the candidate mapping information includes at least one second candidate data mapping and a respective candidate data item of each second candidate data mapping.

[0140] According to an embodiment of the present disclosure, the matching submodule may include a third matching unit, a first determining unit, and a second determining unit.

[0141] The third matching unit is used to match the multiple enhanced texts according to the target intent to obtain at least one target object.

[0142] The first determining unit is configured to determine M target data mappings and N target data items for each target data mapping according to candidate mapping information of at least one data mapping and at least one target object, wherein M and N are both positive integers.

[0143] The second determining unit is configured to determine target mapping information according to the M target data mappings and the N target data items of each target data mapping.

[0144] According to an embodiment of the present disclosure, the verification module 820 may include a generation submodule and a verification submodule.

[0145] The generating submodule is configured to generate an initial statement according to the M target data mappings and the respective N target data items of each target data mapping.

[0146] The verification submodule is used to respond to the initial statement through the syntax tree check, perform permission verification on the initial statement according to the query permission of each target data mapping and each target data item, and obtain a verification result.

[0147] According to an embodiment of the present disclosure, the first determining module 830 may include a first determining submodule and a second determining submodule.

[0148] The first determination submodule is configured to determine the initial statement as a query statement if the verification result indicates that the initial statement passes the verification.

[0149] The second determination submodule is configured to correct the initial statement to obtain a query statement if the verification result indicates that the initial statement fails the verification.

[0150] According to an embodiment of the present disclosure, the query result includes a query graph and a query text.

[0151] According to an embodiment of the present disclosure, the presentation module 840 may include a third determination submodule, a first presentation submodule, and a second presentation submodule.

[0152] The third determination submodule is configured to determine a chart type for querying a chart according to a target intent of the query request and an execution result obtained by executing the query statement.

[0153] The first display submodule is configured to execute a script for generating a chart of the chart type to display a query chart.

[0154] The second display submodule is used to generate and display the query text according to the execution result.

[0155] According to an embodiment of the present disclosure, structural information is obtained by semantically modeling the metadata of each first candidate data mapping in a dataset for a target topic. The candidate objects include enhanced text, historical sentences, and candidate mapping information. The enhanced text is obtained by semantically enhancing the question text in the initial object within the target topic, and the candidate mapping information is obtained by parsing the historical sentences through a syntax tree.

[0156] According to an embodiment of the present disclosure, the information display device 800 may further include an interactive dialogue module and a second determination module.

[0157] The interactive dialogue module is used to use a large model to conduct multiple rounds of interactive dialogues with the object in response to the initial intention obtained by performing intent analysis on the query request not meeting the query conditions, so as to guide the object to supplement the intention so that the query request meets the query conditions.

[0158] The second determination module is used to determine the target intention based on the query request and the dialogue information obtained through multiple rounds of interactive dialogue.

[0159] Figure 9 The structural block diagram of the intelligent agent of the large model according to the embodiment of the present disclosure is schematically shown.

[0160] In the embodiments of the present disclosure, inspired by the von Neumann structure in modern computer theory, such as Figure 9 As shown, the AI ​​agent 900 may include five core modules: an input module 910 , a control module 920 , a storage module 930 , a calculation module 940 and an output module 950 .

[0161] Input module 910 is responsible for receiving or perceiving information such as queries, requests, instructions, signals, or data from the outside world (e.g., users or the external environment) and converting it into a format that AI agent 900 can understand and process. Input module 910 is the primary link for AI agent 900 to interact with the outside world. It enables AI agent 900 to efficiently and accurately obtain necessary "sensory" information from the outside world and respond to this information.

[0162] In an example, the input module 910 may input the query request described above.

[0163] The control module 920 is the core support for the AI ​​agent 900's ability to handle complex tasks. The control module 920 can execute the information display method described above.

[0164] In the example, the control module 920 will continuously interact with the storage module 930, the computing module 940, and / or the output module 950 during operation. However, it should be noted that in the embodiment of the present disclosure, the control module 920 acts as a single initiator to initiate communication with the storage module 930, the computing module 940, and / or the output module 950, and there is no communication coupling between the storage module 930, the computing module 940, and the output module 950.

[0165] In this example, the performance of control module 920 may be closely related to the large model underlying AI agent 900. To fully leverage the capabilities of the large language model, the internal structure of control module 920 may be designed to be highly configurable and extensible to handle a variety of different tasks and requirements in real-world scenarios.

[0166] The storage module 930 may be responsible for memorizing information such as historical conversations, event streams, etc. The candidate set as described above may be included in the storage module 930 .

[0167] In this example, after receiving a query request, AI agent 900 can query the data set for the target topic based on the target intent indicated in the query request to obtain target mapping information. The target mapping information can be stored in storage module 930. AI agent 900 can obtain the target mapping information from storage module 930 and feed it back to control module 920. Control module 920 can then further process the fed-back target mapping information to obtain a query result, and pass the query result to output module 950.

[0168] The operation module 940 can be viewed as a predefined tool library, and tools for semantic modeling, semantic enhancement, and syntax tree parsing, etc., as described above, can be included in the operation module 940 .

[0169] In this example, when the AI ​​agent 900 needs to perform semantic modeling, it can call the relevant semantic modeling tools from the operation module 940 and feed them back to the control module 920. The control module 920 can then use the feedback for semantic modeling tools to perform semantic modeling on the metadata of each first candidate data mapping, obtain structural information, and pass this structural information to the output module 950. It can be understood that although large language models have excellent language understanding and generation capabilities, they are similar to humans and can only solve very limited tasks without the help of any tools. When the AI ​​agent 900 is given the ability to call tools, it can achieve tasks such as using tools for semantic modeling.

[0170] The output module 950 can output the query results described above.

[0171] The AI ​​agent 900 according to the embodiment of the present disclosure can simply and effectively improve the level of intelligence, and enhance flexibility and versatility.

[0172] Figure 10 A block diagram of an electronic device suitable for implementing an information display method according to an embodiment of the present disclosure is schematically shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0173] like Figure 10As shown, device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. RAM 1003 may also store various programs and data required for the operation of device 1000. Computing unit 1001, ROM 1002, and RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to bus 1004.

[0174] Various components in device 1000 are connected to I / O interface 1005, including an input unit 1006, such as a keyboard, mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a magnetic disk, optical disk, etc.; and a communication unit 1009, such as a network card, modem, wireless communication transceiver, etc. The communication unit 1009 allows device 1000 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0175] Computing unit 1001 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Computing unit 1001 performs the various methods and processes described above, such as the information presentation method. For example, in some embodiments, the information presentation method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed onto device 1000 via ROM 1002 and / or communication unit 1009. When the computer program is loaded into RAM 1003 and executed by computing unit 1001, one or more steps of the information presentation method described above may be performed. Alternatively, in other embodiments, computing unit 1001 may be configured to perform the information presentation method via any other suitable means (e.g., via firmware).

[0176] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0177] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0178] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, 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), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0179] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0180] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0181] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0182] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0183] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. An information display method, comprising: In response to receiving a query request for a target topic, a query is performed on a data set for the target topic according to a target intent indicated by the query request, wherein the data set includes structural information and multiple candidate objects, the structural information represents the description information of each first candidate data mapping for the target topic, and the association relationship between the first candidate data mappings, the association relationship refers to the logical or semantic connection between the first candidate data mappings; the method includes: recalling the structural information according to the target intent to obtain at least one data mapping; matching the multiple candidate objects based on the at least one data mapping to determine target mapping information, the target mapping information represents a mapping relationship between a data mapping and a data item, and both the data mapping and the data item are related to the target intent; Verifying the initial statement generated based on the target mapping information according to preset verification rules to obtain a verification result, wherein the preset verification rules define the query permissions of the data mapping and the data item; Determining a query statement based on the initial statement according to the verification result; and Display the query results obtained by executing the query statement.

2. The method according to claim 1, wherein The structural information represents the description information of each first candidate data mapping for the target subject, and the association relationship between the first candidate data mappings; The recalling of the structural information according to the target intent to obtain at least one data mapping includes: Based on the association relationship, and according to the indicators extracted from the target intent, matching the description information of each first candidate data mapping to obtain a first number of the data mappings; as well as Based on the association relationship and according to the intention characteristics of the target intention, the description information of each first candidate data mapping is matched to obtain a second number of the data mappings.

3. The method according to claim 1, wherein The candidate object includes enhanced text and candidate mapping information, wherein the candidate mapping information includes at least one second candidate data mapping and a candidate data item of each second candidate data mapping; The matching the plurality of candidate objects based on the at least one data mapping to determine the target mapping information includes: Matching the multiple enhanced texts according to the target intent to obtain at least one target object; Determining, based on the candidate mapping information of each of the at least one data mapping and the at least one target object, M target data mappings and N target data items for each of the target data mappings, where M and N are both positive integers; and The target mapping information is determined based on the M target data mappings and the N target data items of each target data mapping.

4. The method according to claim 3, wherein: The initial statement generated based on the target mapping information is verified according to the preset verification rules to obtain the verification result, which includes: generating an initial statement according to the M target data mappings and the N target data items of each target data mapping; and In response to the initial statement passing the syntax tree check, performing permission verification on the initial statement according to the query permission of each target data mapping and each target data item, to obtain the verification result; Determining the query statement based on the initial statement according to the verification result includes: If the verification result indicates that the initial statement passes the verification, determining the initial statement as the query statement; and When the verification result indicates that the initial statement fails the verification, the initial statement is corrected to obtain the query statement.

5. The method according to any one of claims 1 to 4, wherein The query results include a query graph and a query text; The display of the query results obtained by executing the query statement includes: determining a chart type for the query chart based on a target intent of the query request and an execution result obtained by executing the query statement; executing a script for generating a chart of the chart type to display the query chart; and According to the execution result, the query text is generated and displayed.

6. The method according to claim 1, wherein The structural information is obtained by semantic modeling metadata of each first candidate data mapping in the candidate set for the target subject; The candidate objects include enhanced text, historical sentences and candidate mapping information. The enhanced text is obtained by semantically enhancing the question text in the initial object within the target subject. The candidate mapping information is obtained by parsing the historical sentences through a syntax tree.

7. The method according to claim 1 , further comprising, before searching the data set for the target subject according to the target intent indicated by the query request: In response to an initial intent obtained by performing intent analysis on the query request not satisfying a query condition, using a large model, conducting multiple rounds of interactive dialogues with the subject to guide the subject to supplement the intent so that the query request satisfies the query condition; and The target intention is determined based on the query request and dialogue information obtained through multiple rounds of interactive dialogues.

8. An information display device comprising: A query module for, in response to receiving a query request for a target subject, querying a data set for the target subject according to a target intent indicated by the query request, wherein the data set includes structural information and multiple candidate objects, the structural information characterizing the respective descriptive information of each first candidate data mapping for the target subject, and the association relationship between the first candidate data mappings, the association relationship being a logical or semantic connection between the first candidate data mappings; comprising: a recall submodule for recalling the structural information according to the target intent to obtain at least one data mapping; a matching submodule for matching the multiple candidate objects based on the at least one data mapping to determine target mapping information, the target mapping information characterizing the mapping relationship between the data mapping and the data item, the data mapping and the data item both being related to the target intent; a verification module, configured to verify the initial statement generated based on the target mapping information according to preset verification rules to obtain a verification result, wherein the preset verification rules define the query permissions of the data mapping and the data item; A first determining module is configured to determine a query statement based on the initial statement according to the verification result; and The display module is used to display the query results obtained by executing the query statement.

9. The device according to claim 8, wherein The structural information represents the description information of each first candidate data mapping for the target subject, and the association relationship between the first candidate data mappings; The recall submodule includes: a first matching unit, configured to match the description information of each of the first candidate data mappings based on the association relationship and according to the indicator extracted from the target intent, to obtain a first number of the data mappings; as well as The second matching unit is used to match the description information of each of the first candidate data mappings based on the association relationship and according to the intention characteristics of the target intention, so as to obtain a second number of the data mappings.

10. The device according to claim 8, wherein The candidate object includes enhanced text and candidate mapping information, wherein the candidate mapping information includes at least one second candidate data mapping and a candidate data item of each second candidate data mapping; The matching submodule includes: a third matching unit, configured to match the plurality of enhanced texts according to the target intent to obtain at least one target object; a first determining unit, configured to determine M target data mappings and N target data items for each target data mapping based on the candidate mapping information of each of the at least one data mapping and the at least one target object, wherein M and N are both positive integers; and The second determining unit is configured to determine the target mapping information according to the M target data mappings and the N target data items of each target data mapping.

11. The device according to claim 10, wherein The verification module includes: a generating submodule, configured to generate an initial statement based on the M target data mappings and the N target data items of each target data mapping; and A check submodule, configured to, in response to the initial statement passing the syntax tree check, perform permission check on the initial statement according to the query permission of each target data mapping and each target data item, and obtain the verification result; The first determining module includes: A first determining submodule is configured to determine the initial statement as the query statement if the verification result indicates that the initial statement passes verification; and The second determining submodule is configured to, if the verification result indicates that the initial statement fails verification, perform error correction on the initial statement to obtain the query statement.

12. The device according to any one of claims 8 to 11, wherein The query results include a query graph and a query text; The display module includes: a third determining submodule, configured to determine a chart type for the query chart according to a target intent of the query request and an execution result obtained by executing the query statement; a first display submodule, configured to execute a script for generating a chart of the chart type to display the query chart; and The second display submodule is used to generate and display the query text according to the execution result.

13. The device according to claim 8, wherein The structural information is obtained by semantic modeling metadata of each first candidate data mapping in the candidate set for the target subject; The candidate objects include enhanced text, historical sentences and candidate mapping information. The enhanced text is obtained by semantically enhancing the question text in the initial object within the target subject. The candidate mapping information is obtained by parsing the historical sentences through a syntax tree.

14. The apparatus according to claim 8, further comprising: an interactive dialogue module for, in response to an initial intent obtained through intent analysis of the query request not satisfying a query condition, conducting multiple rounds of interactive dialogues with the subject using a large model to guide the subject to supplement the intent so that the query request satisfies the query condition; as well as The second determination module is configured to determine the target intention based on the query request and dialogue information obtained through multiple rounds of interactive dialogues.

15. An artificial intelligence agent configured to execute the method according to any one of claims 1 to 7.

16. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 7.

17. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

18. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images