Data query method and device, electronic equipment, storage medium and program

Abstract

The embodiment of the invention discloses a data query method and device, electronic equipment, a storage medium and a program. The method comprises the steps of obtaining a query statement input by a target user, and determining a query intention of the target user according to the query statement input by the target user; determining target hierarchical storage associated data from the hierarchical storage associated data of the target database according to the query intention of the target user; wherein the hierarchical storage associated data comprises at least one of message-level storage data, theme-level storage data and a dialogue flow graph; performing hierarchical retrieval on the target hierarchical storage associated data according to the query intention of the target user to obtain an associated query result of the query statement; and screening and adjusting the associated query result according to the query intention of the target user to generate a target query result of the target user. According to the technical scheme provided by the embodiment of the invention, hierarchical retrieval under different data granularity levels can be realized, so that the data query precision and efficiency are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data processing technology, and in particular to a data query method, apparatus, electronic device, storage medium and program. Background Technology

[0002] In the information age, data resources are experiencing explosive growth. Efficiently and accurately retrieving target information from massive amounts of data has become a core requirement for supporting the efficient operation of various businesses, directly determining the efficiency of information utilization and the scientific level of decision-making. Current mainstream data retrieval technologies primarily rely on keyword matching or simple indexing mechanisms to locate information.

[0003] In the process of realizing this invention, the inventors found the following defects in the prior art: (1) Keyword-based retrieval methods often return a large number of results that only contain the search keywords but do not match the user's actual search intent. Users need to manually screen the search results, which consumes a lot of time and energy and the search efficiency is low; (2) The prior art generally ignores the inherent multi-granularity structure characteristics of different types of data. Due to the lack of an adaptation mechanism for the multi-granularity structure of data, the existing retrieval technology cannot perform retrieval operations at the appropriate granularity level according to the characteristics of the data itself and the actual search needs. This makes it difficult to accurately locate the target information and further exacerbates the problem of insufficient retrieval accuracy. Summary of the Invention

[0004] This invention provides a data query method, apparatus, electronic device, storage medium, and program that can achieve hierarchical retrieval at different data granularity levels, thereby improving the accuracy and efficiency of data query.

[0005] According to one aspect of the present invention, a data query method is provided, comprising:

[0006] Obtain the query statement input by the target user, and determine the target user's query intent based on the query statement input by the target user;

[0007] Based on the query intent of the target user, target hierarchical storage associated data is determined from the hierarchical storage associated data of the target database; wherein, the hierarchical storage associated data includes at least one of message-level storage data, topic-level storage data, and dialogue flow graph;

[0008] Based on the query intent of the target user, the target hierarchical storage related data is hierarchically retrieved to obtain the related query results of the query statement;

[0009] The related query results are filtered and adjusted according to the query intent of the target user to generate the target query results for the target user.

[0010] According to another aspect of the present invention, a data query apparatus is provided, comprising:

[0011] The query intent determination module is used to obtain the query statement input by the target user and determine the query intent of the target user based on the query statement input by the target user.

[0012] The target hierarchical storage associated data acquisition module is used to determine target hierarchical storage associated data from the hierarchical storage associated data of the target database according to the query intent of the target user; wherein, the hierarchical storage associated data includes at least one of message-level storage data, topic-level storage data, and dialogue flow graph;

[0013] The related query result determination module is used to perform hierarchical retrieval of the target hierarchical storage related data according to the query intent of the target user, and obtain the related query results of the query statement;

[0014] The target query result generation module is used to filter and adjust the related query results according to the query intent of the target user, and generate the target query result for the target user.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the data query method according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the data query method described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer program product is also provided, comprising a computer program that, when executed by a processor, implements the data query method described in any embodiment of the present invention.

[0021] This invention, through its embodiments, acquires the query statement input by the target user and determines the target user's query intent based on this statement. Further, based on the target user's query intent, it determines target hierarchical storage association data from the target database, which includes message-level storage data, topic-level storage data, and dialogue flow graphs. After determining the target hierarchical storage association data, it performs a hierarchical retrieval of this data according to the target user's query intent, obtaining the associated query results. These results are then filtered and adjusted based on the target user's query intent to generate the target user's target query results. This solution addresses the problems of insufficient retrieval accuracy and the inability to perform retrieval operations at an adaptive granularity level based on data characteristics and actual retrieval needs in existing data retrieval methods. It enables hierarchical retrieval at different data granularity levels, thereby improving the accuracy and efficiency of data queries.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of a data query method provided in Embodiment 1 of the present invention;

[0025] Figure 2 This is a flowchart of a data query method provided in Embodiment 2 of the present invention;

[0026] Figure 3 This is an architecture diagram of an intelligent chat history retrieval system provided in Embodiment 2 of the present invention;

[0027] Figure 4 This is a schematic diagram of a data query device provided in Embodiment 3 of the present invention;

[0028] Figure 5 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "initial," "intermediate," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] Example 1

[0032] Figure 1 This is a flowchart of a data query method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where retrieval operations are performed at an adaptive granularity level based on the characteristics of the data itself and actual retrieval needs. This method can be executed by a data query device, which can be implemented in software and / or hardware, and is generally integrated into an electronic device. This electronic device can be a terminal device or a server device, as long as it can execute the data query method. The present invention does not limit the specific type of electronic device. Correspondingly, as... Figure 1 As shown, the method includes the following operations:

[0033] S110. Obtain the query statement input by the target user, and determine the query intent of the target user based on the query statement input by the target user.

[0034] In this context, the target user can be any user who is about to perform data retrieval. The query statement can be a query entered by the target user. The query intent can be the specific purpose, need, or expected result that the target user wants to achieve through the query. For example, query intent can include, but is not limited to, fact query intent, summary query intent, and relation query intent. This embodiment of the invention does not limit the specific type of query intent. A fact query intent is a query intent type where the target user initiates a query with the purpose of obtaining objective, verifiable, and definitive factual information. A summary query intent is a query intent type where the target user initiates a query with the purpose of summarizing and refining certain content to obtain concise and structured core points. A relation query intent is a query intent type where the target user initiates a query with the purpose of exploring the relational attributes between two or more things.

[0035] In a specific example, when the target user enters the query "meeting room number", the target user's query intent can be a fact query intent; when the target user enters the query "project discussion last week", the target user's query intent is a summary query intent; and when the target user enters the query "decision between A and B", the target user's query intent is a relation query intent.

[0036] In existing retrieval technology systems, the fixed-weight retrieval fusion mechanism has significant limitations. This mechanism cannot dynamically adapt to the diverse query intents of users, easily leading to issues such as fluctuating search results and insufficient accuracy. Therefore, in scenarios where target users conduct information retrieval, the first step is to obtain the query statement entered by the target user. Then, based on this query statement, the user's specific query intent can be parsed and determined, serving as the core reference for subsequent retrieval strategy formulation and result optimization.

[0037] S120. Determine target hierarchical storage association data from the hierarchical storage association data of the target database according to the query intent of the target user; wherein, the hierarchical storage association data includes at least one of message-level storage data, topic-level storage data, and dialogue flow graph.

[0038] The target database can be a database specifically designed to store various data resources and support data retrieval operations by target users. For example, the target database can be used to store dialogue data. Hierarchical storage associated data can be structured data obtained by implementing a hierarchical storage strategy based on the data differentiation in the target database. For example, hierarchical storage associated data can include, but is not limited to, message-level storage data, topic-level storage data, and dialogue flow graphs. This embodiment of the invention does not limit the specific type of hierarchical storage associated data. Message-level storage data can be storage data based on a single dialogue data entry. For example, message-level storage data can include, but is not limited to, information such as the text, sender, receiver, timestamp, and message type of a single dialogue data entry. This embodiment of the invention does not limit the specific data content included in message-level storage data. Topic-level storage data can be storage data organized by topic. The dialogue flow graph can be a directed graph structure used to present the complete logical link of interactions between various roles in the dialogue data.

[0039] Accordingly, after determining the target user's query intent based on the query statement entered by the target user, the target hierarchical storage related data can be determined from the hierarchical storage related data of the target database based on the target user's query intent, so as to perform accurate retrieval in the target hierarchical storage related data, thereby improving retrieval efficiency and retrieval accuracy.

[0040] In a specific example, hierarchical storage of related data can also include session-level storage data, time-window-level storage data, and topic evolution graphs. Session-level storage data can be session fragments segmented based on fixed time intervals and topic shifts. Session-level storage data can be used to identify semantic coherence between adjacent dialogue data. Furthermore, a large model can be used to perform topic clustering on the session-level storage data to obtain topic-level storage data, along with topic tags and topic evolution paths for each topic-level storage data. Time-window-level storage data can be a dataset formed by defining different time dimensions based on business needs, creating a dedicated index for each time window, and classifying and storing data according to its respective time window. For example, it can be divided according to time dimensions such as days, weeks, months, or project cycles. A topic evolution graph can represent the complete trajectory of a topic in dialogue data from its emergence, development, to its shift, and can be used to calculate topic timeliness scores and context-expanded retrieval.

[0041] S130. Based on the query intent of the target user, perform hierarchical retrieval of the target hierarchical storage related data to obtain the related query results of the query statement.

[0042] Among them, the related query results can be a collection of dialogue data filtered from the target database that have semantic or logical relationships with the query statement entered by the user.

[0043] Accordingly, after determining the target hierarchical storage associated data, a targeted retrieval strategy can be determined based on the target user's query intent. Then, hierarchical retrieval operations can be performed on the target hierarchical storage associated data according to the determined retrieval strategy to obtain a set of dialogue data related to the target user's query statement, and use it as the associated query result of the query statement.

[0044] S140. Filter and adjust the related query results according to the query intent of the target user to generate the target query results for the target user.

[0045] The target query result can be the result that the target user expects to obtain when initiating a query request, which is exactly matched with the query intent.

[0046] Correspondingly, after obtaining the related query results of the query statement, the related query results can be filtered and adjusted according to the query intent of the target user, so as to obtain the target query results that meet the query intent of the target user.

[0047] Therefore, the data query method provided in this embodiment of the invention, relying on the inherent multi-granularity structural features of dialogue data and combining the query intent of the target user, conducts hierarchical retrieval at the matching granularity level, effectively balancing the accuracy and efficiency of data query. Furthermore, this embodiment of the invention can adaptively adjust the data retrieval strategy based on the query intent of the target user, thereby improving the stability of the retrieval results.

[0048] This invention, through its embodiments, acquires the query statement input by the target user and determines the target user's query intent based on this statement. Further, based on the target user's query intent, it determines target hierarchical storage association data from the target database, which includes message-level storage data, topic-level storage data, and dialogue flow graphs. After determining the target hierarchical storage association data, it performs a hierarchical retrieval of this data according to the target user's query intent, obtaining the associated query results. These results are then filtered and adjusted based on the target user's query intent to generate the target user's target query results. This solution addresses the problems of insufficient retrieval accuracy and the inability to perform retrieval operations at an adaptive granularity level based on data characteristics and actual retrieval needs in existing data retrieval methods. It enables hierarchical retrieval at different data granularity levels, thereby improving the accuracy and efficiency of data queries.

[0049] Example 2

[0050] Figure 2This is a flowchart of a data query method provided in Embodiment 2 of the present invention. This embodiment is based on the above embodiment and is further specified. In this embodiment, a specific optional implementation method is given to perform hierarchical retrieval of target hierarchically stored related data according to the target user's query intent to obtain the related query results of the query statement. Correspondingly, as Figure 2 As shown, the method in this embodiment may include:

[0051] S210. Obtain the query statement input by the target user, and determine the query intent of the target user based on the query statement input by the target user.

[0052] In an optional embodiment of the present invention, before obtaining the query statement input by the target user, the method may further include: generating a dialogue flow graph of the dialogue data, using the dialogue users in the dialogue data as nodes and the interaction frequency and response sequence included in the dialogue data as edges; generating a topic evolution graph of the dialogue data, using the dialogue topics in the dialogue data as nodes and the time sequence and semantic association strength included in the dialogue data as edges; generating a user profile of the target user based on the target user's historical query behavior data; generating the hierarchical storage association data based on the original dialogue data; and generating a multi-source fusion vector corresponding to the hierarchical storage association data based on the dialogue flow graph and each data record in the original dialogue data; wherein the multi-source fusion vector includes at least one of a data semantic vector, a dialogue flow graph vector, and a user role vector.

[0053] In this context, the dialogue user can be the subject of the dialogue interaction in the dialogue data; for example, the dialogue user can be the speaker in the dialogue data. Interaction frequency can be the cumulative number of effective interactive behaviors between dialogue users within a specific time range. Response sequence can be the chronological order of message sending and replying behaviors of dialogue users in the dialogue data, as well as the set of relational response relationships and time characteristics between messages. Dialogue topic can be the core content or core issue around which a dialogue revolves. Semantic association strength can be the degree of semantic association between different messages on the same topic in the dialogue data. Topic evolution graph can be used to represent the complete trajectory of a topic in the dialogue data from its generation, development, to its shift. Original dialogue data can be dialogue data obtained in real time from the target database. Historical query behavior data can be all behavioral records generated by the target user for various data query needs over a past period. User profile can be a user feature model constructed based on the target user's historical query behavior data. Multi-source fusion vector can be a vector generated by fusing multi-source data based on the dialogue flow graph and various data records in the original dialogue data. For example, the multi-source fusion vector can include, but is not limited to, data semantic vectors, dialogue flow graph vectors, and user role vectors.

[0054] In this embodiment of the invention, before obtaining the query statement input by the target user, a dialogue flow graph of the dialogue data can be generated, using the dialogue users as nodes and the interaction frequency and response sequence as edges. Further, the dialogue flow graph can be encoded using the GraphSAGE (Graph Simple and Aggregate) algorithm to generate a dialogue flow graph vector. Simultaneously, hierarchical storage associated data can be generated based on the original dialogue data, and multi-source fusion vectors corresponding to the hierarchical storage associated data can be generated based on the dialogue flow graph and each data record in the original dialogue data. For example, the multi-source fusion vector of each data record = text semantic vector (768 dimensions) ⊕ dialogue flow graph vector (128 dimensions) ⊕ user role vector (64 dimensions).

[0055] Additionally, a topic evolution graph of the dialogue data can be generated by using the dialogue topics in the dialogue data as nodes and the temporal order and semantic association strength included in the dialogue data as edges. The GraphSAGE algorithm can then be used to encode the topic evolution graph, thereby generating a topic evolution graph vector of the dialogue data.

[0056] In addition, user profiles of target users can be generated based on their historical query behavior data. This includes recording the high-frequency words used in the target user's queries, the granularity of their preferred information (detailed / general), and the categories of topics they are interested in. The results of related queries can then be filtered and adjusted based on the target user profiles, thereby making the query results more accurately match the actual query intent of the target user.

[0057] S220. Determine the target hierarchical storage association data from the hierarchical storage association data of the target database according to the query intent of the target user.

[0058] In an optional embodiment of the present invention, determining the target hierarchical storage association data from the hierarchical storage association data of the target database based on the target user's query intent may include: if the target user's query intent is a fact query intent, then determining the weight ratio of vectors and keywords based on the fact query intent, and determining the message-level storage data and the topic-level storage data as the target hierarchical storage association data; if the target user's query intent is a summary query intent, then determining the weight ratio of vectors and keywords based on the summary query intent, and determining the message-level storage data and the topic-level storage data as the target hierarchical storage association data; if the target user's query intent is a relationship query intent, then determining the weight ratio of vectors, keywords, and graph structures based on the relationship query intent, and determining the message-level storage data, the topic-level storage data, and the dialogue flow graph as the target hierarchical storage association data.

[0059] In this embodiment of the invention, when determining the target hierarchical storage associated data from the hierarchical storage associated data of the target database according to the query intent of the target user, precise keyword matching can be performed in message-level storage data, and the format features preserved by FPE (Format-Preserving Encryption) can be used to quickly locate the data; vector retrieval can be performed in topic-level storage data, and topic cluster summaries can be used to accelerate the efficiency of retrieval and matching; graph structure traversal can be performed on the dialogue flow graph to find the interaction records of key dialogue users.

[0060] Therefore, when the target user's query intent is determined to be a factual query intent, the weight ratio of vectors and keywords can be determined based on the factual query intent. For example, in this case, the weight ratio of vectors and keywords can be set to 3:7, relying heavily on keyword exact matching. Furthermore, message-level stored data and topic-level stored data can be identified as target hierarchical stored related data. When the target user's query intent is determined to be a summary query intent, the weight ratio of vectors and keywords can be determined based on the summary query intent. For example, in this case, the weight ratio of vectors and keywords can be set to 8:2, relying heavily on semantic understanding. Furthermore, message-level stored data and topic-level stored data can be identified as target hierarchical stored related data. When the target user's query intent is determined to be a relational query intent, the weight ratio of vectors, keywords, and graph structures can be determined based on the relational query intent. For example, in this case, the weight ratio of vectors, keywords, and graph structures can be set to 5:3:2. Furthermore, message-level stored data, topic-level stored data, and dialogue flow graphs can be identified as target hierarchical stored related data. The above solution identifies the category characteristics of the target user's query intent and adaptively adjusts the weight allocation and execution strategy of the three retrieval methods—vector retrieval, keyword retrieval, and graph structure retrieval—in real time, effectively solving the technical defect of the traditional static fusion retrieval mode's insufficient adaptability to diverse query scenarios.

[0061] S230. Based on the target user's query intent and target weight ratio, perform hierarchical retrieval of the multi-source fusion vector corresponding to the target hierarchical storage associated data to obtain the initial associated query results.

[0062] The target weight ratio can be the proportion of each retrieval method determined based on the target user's query intent. The initial related query results can be multiple results directly obtained by hierarchically retrieving the multi-source fusion vector corresponding to the hierarchically stored related data of the target.

[0063] Specifically, after determining the target weight ratio and the target hierarchical storage associated data, the multi-source fusion vector corresponding to the target hierarchical storage associated data can be hierarchically retrieved according to the target user's query intent and the target weight ratio, thereby obtaining multiple initial associated query results related to the query statement entered by the target user.

[0064] Optionally, the weights of each retrieval method can be dynamically adjusted based on the relevance distribution of the initial related query results. For example, when the confidence level of the vector retrieval results is low, the weight of the keyword retrieval can be automatically increased to supplement it.

[0065] S240. Calculate the multidimensional relevance comprehensive score of each of the initial association query results.

[0066] Among them, the multidimensional relevance comprehensive score can be the overall score result obtained by weighted integration or comprehensive calculation after quantitative evaluation of each initial related query result from multiple preset dimensions.

[0067] Correspondingly, after obtaining the initial related query results, a comprehensive calculation can be performed on each initial related query result to obtain a multidimensional relevance comprehensive score for each initial related query result.

[0068] In an optional embodiment of the present invention, calculating the multidimensional relevance comprehensive score of each of the initial related query results may include: calculating the semantic similarity between the query statement input by the target user and each of the initial related query results; calculating the keyword matching degree between the query statement input by the target user and each of the initial related query results; calculating the dialogue importance score of each of the initial related query results based on the dialogue flow graph of the target knowledge base; calculating the topic timeliness score of each of the initial related query results based on the topic evolution graph; calculating the user preference matching degree of each of the initial related query results based on the user profile of the target user; determining the initial multidimensional relevance comprehensive score of each of the initial related query results based on the semantic similarity, the keyword matching degree, the dialogue importance score, the topic timeliness score, and the user preference matching degree; and propagating and correcting each of the initial multidimensional relevance comprehensive scores based on the structural relationship between the dialogue flow graph and the topic evolution graph to obtain the multidimensional relevance comprehensive score of each of the initial related query results.

[0069] Semantic similarity refers to the degree of similarity between the target user's input query and each initial related query result at the semantic content level. Keyword matching degree refers to the degree of matching between the keywords in the target user's input query and the keywords in each initial related query result in terms of overlap, relevance, and semantic association. Dialogue importance score is a score that quantifies the influence of the users in the dialogue within each initial related query result. Topic timeliness score is a score used to represent the time value of each initial related query result. User preference matching degree refers to the degree of matching between each initial related query result and user preferences. For example, the user preference matching degree of each initial related query result can be calculated based on the personalized weights corresponding to the user profile. The initial multidimensional relevance comprehensive score can be a comprehensive score obtained by weighted summation of semantic similarity, keyword matching degree, dialogue importance score, topic timeliness score, and user preference matching degree.

[0070] In this embodiment of the invention, when calculating the multidimensional relevance score of each initial related query result, the semantic similarity between the query statement input by the target user and each initial related query result can be calculated; the keyword matching degree between the query statement input by the target user and each initial related query result can be calculated; the dialogue importance score of each initial related query result can be calculated based on PageRank and the dialogue flow graph; the topic timeliness score of each initial related query result can be calculated based on the time decay factor corresponding to the topic evolution path; and the user preference matching degree of each initial related query result can be calculated based on the personalized weight corresponding to the user profile. Furthermore, the semantic similarity, keyword matching degree, dialogue importance score, topic timeliness score, and user preference matching degree can be weighted and summed to determine the initial multidimensional relevance score of each initial related query result. After obtaining the initial multidimensional relevance score of each initial related query result, the score can be propagated and corrected using the structural relationship of the dialogue flow graph and the topic evolution graph to obtain the multidimensional relevance score of each initial related query result. For example, one can first locate topic nodes related to the query intent, and then, starting from these nodes, diffuse relevance scores along the path of the topic evolution graph to related topics, thereby dynamically correcting the initial multidimensional relevance comprehensive score. Simultaneously, during the above calculation and correction process, result diversity control can be used as a ranking constraint to avoid the concentrated appearance of highly similar nodes, ensuring that search results cover different speakers and topic branches. Specific implementation methods for result diversity control include using dialogue flow graphs to detect duplicate speakers or similar content, and using a differentiated filtering mechanism to ensure that the final search results present multi-perspective information dimensions.

[0071] S250. Determine the intermediate query results for the target user based on the multidimensional relevance comprehensive score of each of the initial association query results.

[0072] The intermediate query result can be a query result that best matches the query statement entered by the target user, determined by a comprehensive multi-dimensional relevance score of each initial related query result.

[0073] Specifically, after obtaining the multidimensional relevance comprehensive score of each initial related query result, the initial related query results can be sorted according to the multidimensional relevance comprehensive score, so as to filter out the intermediate query results with the highest matching degree with the query statement entered by the target user from each initial related query result.

[0074] S260. Determine the context information of the intermediate query results based on the dialogue flow graph and the topic evolution graph.

[0075] The context information can be information in the target database that is associated with the intermediate query results.

[0076] Specifically, after determining the intermediate query results for the target user, the dialogue flow path of the intermediate query results can be determined based on the dialogue flow graph. Simultaneously, the topic evolution context of the intermediate query results can be determined based on the topic evolution graph. Furthermore, the dialogue flow path and topic evolution context can be used as contextual information for the intermediate query results.

[0077] S270. Generate the associated query results of the query statement based on the user profile of the target user, the intermediate query results, and the context information of the intermediate query results.

[0078] Accordingly, after determining the contextual information of the intermediate query results, information assembly rules such as information presentation granularity, ranking of key elements, and tone and depth of explanation can be determined based on the target user's user profile. Furthermore, the intermediate query results and their contextual information can be assembled according to these information assembly rules, and the assembled result can be used as the associated query result of the query statement.

[0079] S280. Based on the query intent of the target user, filter and adjust the related query results to generate the target query results for the target user.

[0080] In an optional embodiment of the present invention, the step of filtering and adjusting the related query results according to the query intent of the target user to generate the target query result for the target user may include: extracting key information from the related query results according to the user profile of the target user to obtain summary information of the related query results; extracting initial target query results from the related query results according to the query intent of the target user; and adjusting the initial target query results according to the user role information in the user profile to obtain the target query result for the target user.

[0081] The summary information of the related query results can be obtained by extracting key information from the related query results. The initial target query results can be information directly extracted from the related query results based on the target user's query intent.

[0082] In this embodiment of the invention, when filtering and adjusting related query results according to the target user's query intent, key information can be extracted from the related query results based on the target user's user profile, thereby obtaining summary information of the related query results. In a specific example, when the target user queries "progress of project A", if the user profile shows details of the target user's preferences, a detailed summary including specific tasks, responsible persons, and time points can be generated; if the user profile shows a general overview of the target user's preferences, a high-level progress summary of project A can be generated.

[0083] Simultaneously, initial target query results can be extracted from related query results based on the target user's query intent. In a specific example, if the target user's query intent is a fact query, an extraction-based generation strategy can be used to directly extract the answer from the related query results; if the target user's query intent is a summary query, an abstract generation strategy can be used to synthesize multiple data records in the related query results to generate a summary; if the target user's query intent is a relational query, a graph path interpretation generation strategy can be used to explain the logical relationships between the data records.

[0084] Furthermore, the initial target query results can be adjusted based on user role information in the user profile to obtain the target query results for the target user. In a specific example, the formality and granularity of the initial query results can be flexibly adjusted according to the target user's level (such as management or execution level). For example, when targeting management, the focus can be on extracting core conclusions, key indicators, and decision-making information, while downplaying technical details and underlying data, using a highly generalized and formal expression. When targeting the execution level, specific execution parameters, operational details, and data sources can be added, retaining the complete logical chain, and balancing professionalism and practicality in the expression to support the implementation of specific tasks.

[0085] Optionally, after obtaining the target user's target query results, the user profile of the target user can be updated based on the target user's feedback.

[0086] Figure 3 This is an architecture diagram of a chat history intelligent retrieval system provided in Embodiment 2 of the present invention. In a specific example, such as... Figure 3 As shown, the aforementioned intelligent retrieval system includes a data acquisition layer, a context-aware layer, a storage and indexing layer, a query processing layer, and an intelligent generation layer.

[0087] The data acquisition layer communicates with the instant messaging application interface and can collect chat records in real time through the multi-granularity chat record modeling module. At the same time, it can build a four-level hierarchical structure of "message level - session level - topic level - time window level" while fully preserving the semantic relevance of chat records, and establish an independent index system for each level of data to ensure the accuracy and efficiency of data retrieval.

[0088] The context-aware layer can extract contextual features from chat logs collected in real time by the multi-granularity chat log modeling module. For example, the extracted contextual features may include, but are not limited to, the speaker's role (e.g., leader or client), message response sequence, sentiment, and mention relationships in the chat log. This embodiment of the invention does not limit the specific content included in the extracted contextual features. Simultaneously, the context-aware layer can also perform targeted processing on sensitive information in the chat logs. Specifically, the FPE algorithm can be used to anonymize sensitive information such as phone numbers in the chat logs and to reversibly pseudonymize user names in the chat logs, thus achieving both desensitization protection of sensitive information and ensuring that semantic relevance is not compromised during data retrieval.

[0089] The storage and indexing layer incorporates a privacy-enhancing vector database. This database employs adversarial masking techniques to perturb sensitive information regions, achieving searchable but irreversible privacy protection. Specifically, this database can store hierarchical data at four levels: message-conversation-topic-time window, and can also persistently store sensitive information processed by the context-aware layer. Simultaneously, the storage and indexing layer also handles the storage of dialogue flow graphs, topic evolution graphs, and vectors generated by graph neural network encoders, providing data support for full-dimensional retrieval. This system leverages graph neural network technology to analyze the interaction relationships and topic development paths of dialogue participants. By quantifying node associations and path logic, it enhances the accuracy of relationship queries and contextual understanding, improving the retrieval system's adaptability to complex dialogue scenarios.

[0090] The intelligent generation layer incorporates a user profiling module, which generates user profiles based on historical query behavior data to provide decision-making references for the context assembler in the query processing layer. Simultaneously, the intelligent generation layer also includes a large-scale model answer generator. This generator, based on a pre-trained large language model, personalizes the answer by semantically parsing, filtering, removing redundant content, and logically reconstructing the input related query results, combined with user profile features.

[0091] Large Language Models (LLMs), also known as large-scale language models, are deep learning models trained on massive amounts of relevant data (such as text, speech, or combined text and image data) to process text sequences. They can generate natural language text or understand the meaning of spoken text. These models typically have billions of parameters. LLMs can handle various natural language tasks, such as text classification, question answering, and dialogue, and have wide applications. The input to a LLM is data, such as text, speech, or combined text and image data. The LLM encodes the input data to obtain corresponding word vectors, and then decodes these word vectors to automatically process the input data and obtain the corresponding output data. For example, text can be input into a LLM, which will process and predict the input text, outputting the corresponding response text.

[0092] The query processing layer receives user natural language query requests and obtains initial related query results through a dynamic hybrid retrieval engine. Then, a graph augmentation reordering unit reorders the initial results based on their matching degree to obtain the optimal intermediate query results. Finally, a context assembler integrates and processes these results to generate the final related query results, providing a basis for the large model answer generator. The dynamic hybrid retrieval engine includes an intent classifier, a retrieval strategy adaptive adjuster, and a multi-path recall fusion unit.

[0093] This invention, through its embodiments, acquires the query statement input by the target user and determines the target user's query intent based on this statement. Further, based on the target user's query intent, it determines target hierarchical storage association data from the target database, which includes message-level storage data, topic-level storage data, and dialogue flow graphs. After determining the target hierarchical storage association data, it performs a hierarchical retrieval of this data according to the target user's query intent, obtaining the associated query results. These results are then filtered and adjusted based on the target user's query intent to generate the target user's target query results. This solution addresses the problems of insufficient retrieval accuracy and the inability to perform retrieval operations at an adaptive granularity level based on data characteristics and actual retrieval needs in existing data retrieval methods. It enables hierarchical retrieval at different data granularity levels, thereby improving the accuracy and efficiency of data queries.

[0094] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information (such as voice information) in this technical solution comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0095] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant regions.

[0096] It should be noted that any arrangement or combination of the technical features in the above embodiments also falls within the protection scope of this invention.

[0097] Example 3

[0098] Figure 4 This is a schematic diagram of a data query device provided in Embodiment 3 of the present invention, as shown below. Figure 4 As shown, the device includes: a query intent determination module 310, a target hierarchical storage associated data acquisition module 320, an associated query result determination module 330, and a target query result generation module 340, wherein:

[0099] The query intent determination module 310 is used to obtain the query statement input by the target user and determine the query intent of the target user based on the query statement input by the target user.

[0100] The target hierarchical storage association data acquisition module 320 is used to determine the target hierarchical storage association data from the hierarchical storage association data of the target database according to the query intent of the target user; wherein, the hierarchical storage association data includes at least one of message-level storage data, topic-level storage data, and dialogue flow graph.

[0101] The associated query result determination module 330 is used to perform hierarchical retrieval of the target hierarchical storage associated data according to the query intent of the target user, and obtain the associated query result of the query statement.

[0102] The target query result generation module 340 is used to filter and adjust the related query results according to the query intent of the target user, and generate the target query result for the target user.

[0103] This invention, through its embodiments, acquires the query statement input by the target user and determines the target user's query intent based on this statement. Further, based on the target user's query intent, it determines target hierarchical storage association data from the target database, which includes message-level storage data, topic-level storage data, and dialogue flow graphs. After determining the target hierarchical storage association data, it performs a hierarchical retrieval of this data according to the target user's query intent, obtaining the associated query results. These results are then filtered and adjusted based on the target user's query intent to generate the target user's target query results. This solution addresses the problems of insufficient retrieval accuracy and the inability to perform retrieval operations at an adaptive granularity level based on data characteristics and actual retrieval needs in existing data retrieval methods. It enables hierarchical retrieval at different data granularity levels, thereby improving the accuracy and efficiency of data queries.

[0104] Optionally, the above-mentioned device may further include a data processing module, configured to generate a dialogue flow graph of the dialogue data, using the dialogue users in the dialogue data as nodes and the interaction frequency and response sequence included in the dialogue data as edges; generate a topic evolution graph of the dialogue data, using the dialogue topics in the dialogue data as nodes and the time sequence and semantic association strength included in the dialogue data as edges; generate a user profile of the target user based on the target user's historical query behavior data; generate the hierarchical storage association data based on the original dialogue data; and generate a multi-source fusion vector corresponding to the hierarchical storage association data based on the dialogue flow graph and each data record in the original dialogue data; wherein the multi-source fusion vector includes at least one of a data semantic vector, a dialogue flow graph vector, and a user role vector.

[0105] Optionally, the target user's query intent includes fact query intent, summary query intent, and relationship query intent; the target hierarchical storage association data acquisition module 320 is specifically used for: if the target user's query intent is a fact query intent, then determining the weight ratio of vectors and keywords based on the fact query intent, and determining the message-level storage data and the topic-level storage data as the target hierarchical storage association data; if the target user's query intent is a summary query intent, then determining the weight ratio of vectors and keywords based on the summary query intent, and determining the message-level storage data and the topic-level storage data as the target hierarchical storage association data; if the target user's query intent is a relationship query intent, then determining the weight ratio of vectors, keywords, and graph structures based on the relationship query intent, and determining the message-level storage data, the topic-level storage data, and the dialogue flow graph as the target hierarchical storage association data.

[0106] Optionally, the related query result determination module 330 is specifically used for: performing hierarchical retrieval of the multi-source fusion vector corresponding to the hierarchical storage related data of the target user according to the query intent and target weight ratio of the target user, to obtain initial related query results; calculating the multidimensional relevance comprehensive score of each initial related query result; determining the intermediate query results of the target user according to the multidimensional relevance comprehensive score of each initial related query result; determining the context information of the intermediate query results according to the dialogue flow graph and the topic evolution graph; and generating the related query results of the query statement according to the user profile of the target user, the intermediate query results and the context information of the intermediate query results.

[0107] Optionally, the related query result determination module 330 is further configured to: calculate the semantic similarity between the query statement input by the target user and each of the initial related query results; calculate the keyword matching degree between the query statement input by the target user and each of the initial related query results; calculate the dialogue importance score of each of the initial related query results based on the dialogue flow graph of the target knowledge base; calculate the topic timeliness score of each of the initial related query results based on the topic evolution graph; calculate the user preference matching degree of each of the initial related query results based on the user profile of the target user; determine the initial multidimensional relevance comprehensive score of each of the initial related query results based on the semantic similarity, the keyword matching degree, the dialogue importance score, the topic timeliness score, and the user preference matching degree; and propagate and correct each of the initial multidimensional relevance comprehensive scores based on the structural relationship between the dialogue flow graph and the topic evolution graph to obtain the multidimensional relevance comprehensive score of each of the initial related query results.

[0108] Optionally, the target query result generation module 340 is specifically used to: extract key information from the related query results based on the user profile of the target user to obtain summary information of the related query results; extract initial target query results from the related query results based on the query intent of the target user; and adjust the initial target query results based on the user role information in the user profile to obtain the target query results of the target user.

[0109] The data query device described above can execute the data query method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method. Technical details not described in detail in this embodiment can be found in the data query method provided in any embodiment of the present invention.

[0110] Since the data query device described above is capable of executing the data query method in the embodiments of the present invention, those skilled in the art can understand the specific implementation and various variations of the data query device in this embodiment based on the data query method described in the embodiments of the present invention. Therefore, how the data query device implements the data query method in the embodiments of the present invention will not be described in detail here. Any device used by those skilled in the art to implement the data query method in the embodiments of the present invention falls within the scope of protection of this application.

[0111] Example 4

[0112] Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is 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 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0113] like Figure 5As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0114] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0115] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as data querying methods.

[0116] In some embodiments, the data query method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the data query method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to execute the data query method by any other suitable means (e.g., by means of firmware).

[0117] Optionally, the data query method may include: obtaining a query statement input by a target user, and determining the target user's query intent based on the query statement input by the target user; determining target hierarchical storage related data from hierarchical storage related data in the target database based on the target user's query intent; wherein, the hierarchical storage related data includes at least one of message-level storage data, topic-level storage data, and dialogue flow graphs; performing hierarchical retrieval on the target hierarchical storage related data according to the target user's query intent to obtain the related query results of the query statement; filtering and adjusting the related query results according to the target user's query intent to generate the target user's target query results.

[0118] Various embodiments of the systems and techniques described above herein 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), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0119] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0120] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0121] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0122] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0123] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0124] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0125] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A data query method, characterized in that, include: Obtain the query statement input by the target user, and determine the target user's query intent based on the query statement input by the target user; Based on the query intent of the target user, target hierarchical storage associated data is determined from the hierarchical storage associated data of the target database; wherein, the hierarchical storage associated data includes at least one of message-level storage data, topic-level storage data, and dialogue flow graph; Based on the query intent of the target user, the target hierarchical storage related data is hierarchically retrieved to obtain the related query results of the query statement; The related query results are filtered and adjusted according to the query intent of the target user to generate the target query results for the target user.

2. The method according to claim 1, characterized in that, Before obtaining the query statement input by the target user, the following is also included: Using the users in the dialogue data as nodes and the interaction frequency and response sequence included in the dialogue data as edges, a dialogue flow graph of the dialogue data is generated. Using the dialogue topics in the dialogue data as nodes and the temporal order and semantic association strength included in the dialogue data as edges, a topic evolution graph of the dialogue data is generated. A user profile of the target user is generated based on the target user's historical query behavior data; The hierarchical storage associated data is generated based on the original dialogue data, and the multi-source fusion vector corresponding to the hierarchical storage associated data is generated based on the dialogue flow graph and each data record in the original dialogue data; wherein, the multi-source fusion vector includes at least one of data semantic vector, dialogue flow graph vector and user role vector.

3. The method according to claim 2, characterized in that, The target user's query intent includes fact query intent, summary query intent, and relationship query intent; The step of determining the target hierarchical storage association data from the hierarchical storage association data of the target database based on the query intent of the target user includes: If the target user's query intent is a factual query intent, then the weight ratio of the vector and keywords is determined according to the factual query intent, and the message-level storage data and the topic-level storage data are determined as the target hierarchical storage association data; If the target user's query intent is a summary query intent, then the weight ratio of the vector and keywords is determined according to the summary query intent, and the message-level storage data and the topic-level storage data are determined as the target hierarchical storage association data; If the target user's query intent is a relationship query intent, then the weight ratio of vectors, keywords, and graph structures is determined according to the relationship query intent, and the message-level storage data, the topic-level storage data, and the dialogue flow graph are determined as the target hierarchical storage association data.

4. The method according to claim 3, characterized in that, The step of hierarchically retrieving the target hierarchical storage related data according to the target user's query intent to obtain the related query results of the query statement includes: Based on the target user's query intent and target weight ratio, hierarchical retrieval is performed on the multi-source fusion vector corresponding to the target hierarchical storage associated data to obtain the initial associated query results; Calculate the multidimensional relevance score of each of the initial related query results; The intermediate query results for the target user are determined based on the multidimensional relevance comprehensive score of each initial related query result. The contextual information of the intermediate query results is determined based on the dialogue flow graph and the topic evolution graph; The associated query results of the query statement are generated based on the user profile of the target user, the intermediate query results, and the context information of the intermediate query results.

5. The method according to claim 4, characterized in that, The calculation of the multidimensional relevance comprehensive score for each of the initial association query results includes: Calculate the semantic similarity between the query statement input by the target user and each of the initial associated query results; Calculate the keyword matching degree between the query statement input by the target user and each of the initial associated query results; Calculate the dialogue importance score of each of the initial related query results based on the dialogue flow graph of the target knowledge base; Calculate the topic timeliness score of each initial related query result based on the topic evolution graph; Calculate the user preference matching degree of each of the initial related query results based on the user profile of the target user; The initial multidimensional relevance comprehensive score of each of the initial associated query results is determined based on the semantic similarity, the keyword matching degree, the dialogue importance score, the topic timeliness score, and the user preference matching degree. Based on the structural relationship between the dialogue flow graph and the topic evolution graph, the initial multidimensional relevance comprehensive scores are propagated and corrected to obtain the multidimensional relevance comprehensive scores of the initial associated query results.

6. The method according to claim 4, characterized in that, The step of filtering and adjusting the related query results according to the query intent of the target user to generate the target query results for the target user includes: Based on the user profile of the target user, key information is extracted from the related query results to obtain summary information of the related query results; Extract initial target query results from the related query results based on the target user's query intent; The initial target query results are adjusted based on the user role information in the user profile to obtain the target query results for the target user.

7. A data query device, characterized in that, include: The query intent determination module is used to obtain the query statement input by the target user and determine the query intent of the target user based on the query statement input by the target user. The target hierarchical storage associated data acquisition module is used to determine target hierarchical storage associated data from the hierarchical storage associated data of the target database according to the query intent of the target user; wherein, the hierarchical storage associated data includes at least one of message-level storage data, topic-level storage data, and dialogue flow graph; The related query result determination module is used to perform hierarchical retrieval of the target hierarchical storage related data according to the query intent of the target user, and obtain the related query results of the query statement; The target query result generation module is used to filter and adjust the related query results according to the query intent of the target user, and generate the target query result for the target user.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that is executed by the at least one processor to enable the at least one processor to perform the data query method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the data query method according to any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, wherein, When the computer program / instruction is executed by the processor, it implements the data query method according to any one of claims 1-6.

Images