Report generation method and device based on big data intelligent interaction

Abstract

The invention specifically provides a report generation method and device based on big data intelligent interaction, and relates to the technical field of data processing of insurance services. The method comprises the following steps: acquiring natural language information and identity information of a user; calling a user intention recognition model corresponding to the user based on the identity information of the user; substituting the natural language information of the user into a user intention recognition model corresponding to the user, generating work order data corresponding to the natural language information, and recording a recognition record of the user intention recognition model; based on the work order data and a preset analysis database, generating an SQL structure corresponding to the work order data and feeding back the SQL structure to a user; and based on a feedback result of the user, selectively executing the SQL structure and visually displaying an execution result in real time. According to the method, the executed SQL structure can be ensured to meet user requirements, intelligent management and analysis of the work order data are improved, the working efficiency is improved, and the labor cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of insurance business data processing, and in particular to a report generation method and device based on big data intelligent interaction. Background Art

[0002] In today's digital age, insurance risk control operations rely on a huge data system, and the number of business tables has increased dramatically with the expansion of enterprise scale and diversification of business. These business tables not only cover multiple vertical fields such as auto insurance, life insurance, and non-auto insurance, but also the data structure and business logic in each field are extremely complex, involving many unique indicators and complex calculation logic.

[0003] The work order management method in the existing technology generally requires manual calculation of the data structure and business logic of multiple vertical fields separately, and then executes the work order processing. This requires professionals to have in-depth technical knowledge and complex operation procedures, making it difficult for ordinary users to use it directly, which in turn affects the efficiency and accuracy of work order management and increases labor costs.

[0004] Accordingly, a new report generation solution is needed in the art to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects, an embodiment of the present invention provides a report generation method and device based on big data intelligent interaction, so as to provide a solution or at least partially solve the technical problem that the work order management method in the prior art requires professionals to have in-depth technical knowledge and complex operation procedures, making it difficult for ordinary users to use it directly, thereby affecting the efficiency and accuracy of work order management and high labor costs.

[0006] In a first aspect, the present invention provides a report generation method based on big data intelligent interaction, the method is applied to a report generation device based on big data intelligent interaction, the device interacts with multiple user terminals, and the method comprises the following steps: Obtaining natural language information of a user and identity information of the user; Based on the identity information of the user, retrieve a user intention recognition model corresponding to the user, wherein the level of the user intention recognition model is one of primary, advanced and ultimate; Substituting the natural language information of the user into the user intention recognition model corresponding to the user, generating work order data corresponding to the natural language information, and recording the recognition record of the user intention recognition model; Based on the work order data and a preset parsing database, an SQL structure corresponding to the work order data is generated and fed back to the user; Based on the user's feedback result, the SQL structure is selectively executed and the execution result is visualized and displayed in real time.

[0007] In a technical solution of the report generation method based on big data intelligent interaction, before calling the user intention recognition model corresponding to the user based on the identity information of the user, the method includes: Generate a user intention recognition model corresponding to each user; The user intention recognition model corresponding to each user is trained to obtain a trained user intention recognition model corresponding to the user.

[0008] In a technical solution of the above-mentioned report generation method based on big data intelligent interaction, generating a user intention recognition model corresponding to each user includes: Obtaining a preset user basic recognition model, wherein the user basic recognition model at least includes a preset basic language comparison library and a preset basic recognition module; The identity information of each user and the preset user basic recognition model are obtained to generate a user intention recognition model corresponding to each user, wherein the level of the recognition training model is primary.

[0009] In a technical solution of the report generation method based on big data intelligent interaction, the training of the user intention recognition model corresponding to each user to obtain the trained user intention recognition model corresponding to the user includes: Select any user as a training sample, and obtain a training sample set corresponding to the training sample, wherein the training sample set includes the identity information of the training sample and the user portrait data of the training sample; Based on the training sample set, the recognition training model is trained to obtain a training result; Based on the training results, selectively adjusting the recognition training model; and selectively adjusting the level of the recognition training model or the adjusted recognition training model based on the training result of the adjusted recognition training model or the training result of the recognition training model; Based on the level of the recognition training model or the adjusted recognition training model, and the training result of the recognition training model or the adjusted recognition training model, the training result of the user intention recognition model corresponding to the user corresponding to the training sample is determined.

[0010] In a technical solution of the report generation method based on big data intelligent interaction, the selecting any one user as a training sample and obtaining a training sample set corresponding to the training sample includes: Select any user as a training sample; Obtaining a preset user portrait database, wherein the user portrait database stores identity information of each user, historical language data of each user, and user portrait data of each user; Based on the training samples and a preset user portrait database, a training sample set corresponding to the training samples is generated.

[0011] In a technical solution of the report generation method based on big data intelligent interaction, the recognition training model is trained based on the training sample set to obtain the training results including: Based on the historical language data corresponding to the training samples in the preset user portrait database, multiple groups of language data sets corresponding to the training samples are generated, wherein the language data sets include at least one data level and multiple dimensions of natural language at each data level; Based on multiple language data sets, the recognition training model is trained to obtain the work order data generated corresponding to each language data set; Based on the user portrait data of the training samples in the training sample set, the work order data generated corresponding to each group of language data sets are tested to obtain the accuracy rate corresponding to each group of language data sets; Based on the accuracy rate corresponding to each group of language data sets, the training results of the recognition training model are obtained, wherein the training results include the comprehensive training accuracy rate and the proportion of the number of matches.

[0012] In a technical solution of the report generation method based on big data intelligent interaction, the training result of the user intention recognition model corresponding to the user corresponding to the training sample is determined based on the level of the recognition training model or the adjusted recognition training model, and the training result of the recognition training model or the adjusted recognition training model, including: If the level of the recognition training model is primary, and the training result of the recognition training model meets the preset additional training requirements, the recognition training model is used as the user intention recognition model corresponding to the user corresponding to the training sample, and it is determined that the user intention recognition model corresponding to the user is successfully trained; or, If the level of the recognition training model reaches the advanced level, and the training result of the recognition training model meets the preset training requirements, the recognition training model is used as the user intention recognition model corresponding to the user corresponding to the training sample, and it is determined that the user intention recognition model corresponding to the user is successfully trained; Otherwise, it is determined that the user intention recognition model training has failed, and the user intention recognition model is retrained.

[0013] In a technical solution of the report generation method based on big data intelligent interaction, the work order data at least includes identification fields that match each dimension of each data level of the natural language information, and the preset parsing database stores SQL statement structures corresponding to each identification field of each dimension of each data level and mapping logic corresponding to each SQL statement structure. The generation of the SQL structure corresponding to the work order data based on the work order data and the preset parsing database includes: Based on the work order data and a preset parsing database, determine the SQL statement structure corresponding to each identification field corresponding to the work order data and the mapping logic corresponding to each SQL structure; Based on the SQL statement structure corresponding to each identification field corresponding to the work order data and the mapping logic corresponding to each SQL statement structure, an SQL structure corresponding to the work order data is generated.

[0014] In a technical solution of the report generation method based on big data intelligent interaction, the selective execution of the SQL structure based on the user's feedback result and the visual real-time display of the execution result include: If there is no adjustment to the SQL structure in the user's feedback structure, the SQL structure is executed and the execution process and execution result of the SQL structure are obtained, and the execution process and execution result of the SQL structure are visually displayed in real time; or, If the feedback result of the user includes an adjustment to the SQL structure, and the adjustment range is lower than a preset adjustment range threshold, adjusting the SQL structure based on the feedback result of the user to obtain an adjusted SQL structure; The adjusted SQL structure is executed and the execution process and execution result of the adjusted SQL structure are obtained, and the execution process and execution result of the adjusted SQL structure are visualized and displayed in real time.

[0015] In the second aspect, an embodiment of the present invention further provides a report generation device based on intelligent interaction of big data, wherein the device interacts with multiple user terminals, and the device includes a user intention recognition model corresponding to each user, a communication device, a natural language input device, a preset parsing database and a control unit. The communication device is communicatively connected to the multiple user terminals to realize information interaction between the device and the multiple user terminals, and the natural language input device is used to collect natural language information of the user. The control unit includes a processor and a memory, and the memory is suitable for storing multiple program codes, and the program codes are suitable for being loaded and run by the processor to execute the report generation method based on intelligent interaction of big data described in the first aspect.

[0016] The above one or more technical solutions of the present invention have at least one or more of the following beneficial effects: By acquiring the identity information of the user, the user intent recognition model is retrieved, and the user intent recognition model for each user is implemented, thereby improving the accuracy of the recognition of the user's natural language information. Then, the user's natural language information is substituted into the corresponding user intent recognition model to generate the work order data corresponding to the natural language information, thereby realizing the automatic generation of the work order data, and recording the recognition record of the user intent recognition model to realize the recording of each recognition result, so as to facilitate the subsequent model optimization and comparison of the recognition results. Then, according to the work order data and the preset parsing database, the corresponding SQL structure is generated and fed back to the user. Then, combined with the user's feedback results, the SQL structure is selectively executed and the execution results are visualized and displayed in real time, thereby realizing the intelligent interaction between the user and the generated SQL structure, so as to ensure that the executed SQL structure meets the user's needs, thereby improving the intelligent management and analysis of the work order data, improving work efficiency, reducing labor costs, and avoiding the work order management method in the prior art that requires professionals to have in-depth technical knowledge and complex operation processes, making it difficult for ordinary users to directly use it, thereby affecting the efficiency and accuracy of work order management, and increasing the technical problem of labor costs.

[0017] In the technical solution of the present invention, by generating a user intent recognition model for each user, the automatic construction of a targeted user intent recognition model corresponding to each user is achieved. Then, by training each user intent recognition model, the trained user intent recognition model can accurately understand the user's needs, thereby generating more accurate work order data and improving the matching degree between the user and the user intent recognition model.

[0018] In the technical solution of the present invention, the recognition training model corresponding to each training sample is trained through the training sample set corresponding to it, and according to the training results, the recognition training model is selectively adjusted, and the recognition training model is selectively adjusted again to ensure that the intention recognition model corresponding to each user matches the user, and the intelligence and accuracy of each intention recognition model are further optimized, thereby improving the accuracy of the intention recognition model and its user. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The disclosure of the present invention will become more easily understood with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. In addition, similar numbers in the figures are used to represent similar components, among which: Figure 1 It is a flowchart of the main steps of a report generation method based on big data intelligent interaction according to an embodiment of the present invention; Figure 2 It is a schematic diagram of the main structural block diagram of a report generation device based on big data intelligent interaction according to an embodiment of the present invention.

[0020] Description of Reference Numerals : 200: report generation device based on big data intelligent interaction; 201: control unit; 2011: processor; 2012: memory; 2013: program code; 202: communication equipment; 203: parsing database; 204: user intention recognition model; 205: natural language input device. DETAILED DESCRIPTION

[0021] Some embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0022] In the description of the present invention, "module" and "processor" may include hardware, software or a combination of the two. A module may include hardware circuits, various suitable sensors, communication ports, and memories, and may also include software parts, such as program codes, or a combination of software and hardware. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware, or a combination of the two. Non-temporary computer-readable storage media include any suitable medium that can store program codes, such as a magnetic disk, a hard disk, an optical disk, a flash memory, a read-only memory, a random access memory, and the like. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The term "at least one A or B" or "at least one of A and B" has a similar meaning to "A and / or B", and may include only A, only B, or A and B. The singular terms "one" and "the" may also include plural forms.

[0023] See attached Figure 1 , Figure 1 FIG. 1 is a flow chart of the main steps of a report generation method based on big data intelligent interaction according to an embodiment of the present invention. Figure 1 As shown, the report generation method based on big data intelligent interaction in the embodiment of the present invention mainly includes the following steps S101 to S105.

[0024] Step S101: Acquire the user's natural language information and the user's identity information; Step S102: based on the identity information of the user, retrieve a user intention recognition model corresponding to the user, wherein the level of the user intention recognition model is one of primary, advanced and ultimate; Furthermore, before calling the user intention recognition model corresponding to the user based on the identity information of the user, the method includes: Generate a user intention recognition model corresponding to each user; The user intention recognition model corresponding to each user is trained to obtain a trained user intention recognition model corresponding to the user.

[0025] In the above embodiment, by generating a user intent recognition model for each user, the automatic construction of a targeted user intent recognition model corresponding to each user is achieved. Then, by training each user intent recognition model, the trained user intent recognition model can accurately understand the user's needs, thereby generating more accurate work order data and improving the matching degree between the user and the user intent recognition model.

[0026] Specifically, generating a user intention recognition model corresponding to each user includes: Obtaining a preset user basic recognition model, wherein the user basic recognition model at least includes a preset basic language comparison library and a preset basic recognition module; The identity information of each user and the preset user basic recognition model are obtained to generate a user intention recognition model corresponding to each user, wherein the level of the recognition training model is primary.

[0027] Specifically, the training of the user intention recognition model corresponding to each user to obtain the trained user intention recognition model corresponding to the user includes: Select any user as a training sample, and obtain a training sample set corresponding to the training sample, wherein the training sample set includes the identity information of the training sample and the user portrait data of the training sample; Based on the training sample set, the recognition training model is trained to obtain a training result; Based on the training results, selectively adjusting the recognition training model; and selectively adjusting the level of the recognition training model or the adjusted recognition training model based on the training result of the adjusted recognition training model or the training result of the recognition training model; Based on the level of the recognition training model or the adjusted recognition training model, and the training result of the recognition training model or the adjusted recognition training model, the training result of the user intention recognition model corresponding to the user corresponding to the training sample is determined.

[0028] In the above embodiment, the recognition training model corresponding to each training sample is trained through the training sample set corresponding to it, and according to the training results, the recognition training model is selectively adjusted, and the recognition training model is selectively adjusted again to ensure that the intention recognition model corresponding to each user matches his or her user, and the intelligence and accuracy of each intention recognition model are further optimized, thereby improving the accuracy of the intention recognition model and its user.

[0029] Specifically, the selecting any one user as a training sample and obtaining a training sample set corresponding to the training sample includes: Select any user as a training sample; Obtaining a preset user portrait database, wherein the user portrait database stores identity information of each user, historical language data of each user, and user portrait data of each user; Based on the training samples and a preset user portrait database, a training sample set corresponding to the training samples is generated.

[0030] Specifically, the recognition training model is trained based on the training sample set to obtain a training result including: Based on the historical language data corresponding to the training samples in the preset user portrait database, multiple groups of language data sets corresponding to the training samples are generated, wherein the language data sets include at least one data level and multiple dimensions of natural language at each data level; Based on multiple language data sets, the recognition training model is trained to obtain the work order data generated corresponding to each language data set; Based on the user portrait data of the training samples in the training sample set, the work order data generated corresponding to each group of language data sets are tested to obtain the accuracy rate corresponding to each group of language data sets; Based on the accuracy rate corresponding to each group of language data sets, the training results of the recognition training model are obtained, wherein the training results include the comprehensive training accuracy rate and the proportion of the number of matches.

[0031] Specifically, the comprehensive accuracy rate corresponding to each language data set is obtained by the following formula: The accuracy rate corresponding to the language data set=(accuracy rate corresponding to each data level in the language data set×sum of preset weight values ​​corresponding to the data levels) / number of multiple data levels in the language data set.

[0032] Specifically, based on the accuracy corresponding to each set of language data sets, the training results of the recognition training model are obtained, including: Determining the comprehensive training accuracy of the recognition training model based on the accuracy corresponding to each group of language data sets; Based on the accuracy rate corresponding to each set of language data sets, the proportion of the matching quantity is determined.

[0033] Specifically, the comprehensive accuracy of training is obtained by the following formula: The overall training accuracy = the sum of the accuracy corresponding to each language data set / the number of multiple language data sets.

[0034] Specifically, based on the accuracy rate corresponding to each language data set, the proportion of the number of matches is determined including: If the accuracy rate corresponding to the language data set exceeds a preset accuracy rate threshold, it is determined that the language data set meets the requirements; Otherwise, it is determined that the language data set does not meet the requirements; Based on the compliance of each language dataset, the proportion of compliance is determined.

[0035] Specifically, the proportion of the number of compliance is obtained by the following formula: The proportion of language datasets that meet the requirements = the number of language datasets that meet the requirements / the number of all language datasets.

[0036] Specifically, the preset accuracy threshold may be 90% or 92%. The settings of the preset accuracy thresholds here are only exemplary. In actual tests, those skilled in the art may set them according to actual needs, which will not be elaborated here.

[0037] Specifically, selectively adjusting the recognition training model based on the training results includes: If the comprehensive training accuracy reaches a preset comprehensive training accuracy threshold, and the matching quantity ratio reaches a preset matching quantity ratio threshold, the recognition training model is not adjusted; Otherwise, the recognition training model is adjusted to obtain an adjusted recognition training model.

[0038] Specifically, the preset comprehensive training accuracy threshold can be 90% or 85%, and the preset compliance ratio threshold can be 60% or 70%. The settings of the preset comprehensive training accuracy threshold and the preset compliance ratio threshold are only exemplary. In actual tests, those skilled in the art can set them according to actual needs, which will not be repeated here.

[0039] Specifically, after the recognition training model is adjusted to obtain the adjusted recognition training model, the method further includes: Based on the training sample set, the adjusted recognition training model is trained to obtain a training result of the adjusted recognition training model.

[0040] Specifically, based on the training result of the adjusted recognition training model or the training result of the recognition training model, selectively adjusting the level of the recognition training model or the adjusted recognition training model includes: If the comprehensive training accuracy in the training result of the adjusted recognition training model reaches the preset comprehensive training accuracy threshold, and the matching quantity ratio reaches the preset matching quantity ratio threshold, the adjusted recognition training model is not readjusted; Otherwise, the adjusted recognition training model is readjusted to obtain an adjusted recognition training model.

[0041] Specifically, the determining the training result of the user intention recognition model corresponding to the user corresponding to the training sample based on the level of the recognition training model or the adjusted recognition training model and the training result of the recognition training model or the adjusted recognition training model includes: If the level of the recognition training model is primary, and the training result of the recognition training model meets the preset additional training requirements, the recognition training model is used as the user intention recognition model corresponding to the user corresponding to the training sample, and it is determined that the user intention recognition model corresponding to the user is successfully trained; or, If the level of the recognition training model reaches the advanced level, and the training result of the recognition training model meets the preset training requirements, the recognition training model is used as the user intention recognition model corresponding to the user corresponding to the training sample, and it is determined that the user intention recognition model corresponding to the user is successfully trained; Otherwise, it is determined that the training of the user intention recognition model has failed, and the user intention recognition model is retrained.

[0042] Specifically, the preset training requirement may be that the comprehensive training accuracy reaches a preset comprehensive training accuracy threshold, and the compliance ratio reaches a preset compliance ratio threshold; the preset additional training requirement may be that the comprehensive training accuracy exceeds 5% of the preset comprehensive training accuracy threshold, and the compliance ratio reaches 10% of the preset compliance ratio threshold; the settings of the preset training requirements and the preset additional training requirements here are only exemplary descriptions. In actual tests, those skilled in the art can set them according to actual needs, and they will not be repeated here.

[0043] Step S103: Substituting the natural language information of the user into the user intention recognition model corresponding to the user, generating work order data corresponding to the natural language information, and recording the recognition record of the user intention recognition model; Step S104: Based on the work order data and a preset parsing database, an SQL structure corresponding to the work order data is generated and fed back to the user; Specifically, the work order data at least includes identification fields that match each dimension of each data level of the natural language information, the preset parsing database stores SQL statement structures corresponding to each identification field of each dimension of each data level and mapping logic corresponding to each SQL statement structure, and the generating of the SQL structure corresponding to the work order data based on the work order data and the preset parsing database includes: Based on the work order data and a preset parsing database, determine the SQL statement structure corresponding to each identification field corresponding to the work order data and the mapping logic corresponding to each SQL structure; Based on the SQL statement structure corresponding to each identification field corresponding to the work order data and the mapping logic corresponding to each SQL statement structure, an SQL structure corresponding to the work order data is generated.

[0044] Specifically, the method for generating the SQL structure corresponding to the work order data can adopt a trained neural network model or a trained machine learning model. The selection of the method for generating the SQL structure corresponding to the work order data here is only an exemplary description. In actual testing, technical personnel in this field can make a selection according to actual needs. As long as the SQL structure corresponding to the work order data can be generated according to the SQL statement structure corresponding to each identification field corresponding to the work order data and the mapping logic corresponding to each SQL statement structure, it will not be repeated here.

[0045] Step S105: Based on the feedback result of the user, selectively execute the SQL structure and display the execution result in a visual real-time manner.

[0046] Specifically, selectively executing the SQL structure based on the user's feedback result and visually displaying the execution result in real time includes: If there is no adjustment to the SQL structure in the user's feedback structure, the SQL structure is executed and the execution process and execution result of the SQL structure are obtained, and the execution process and execution result of the SQL structure are visually displayed in real time; or, If the feedback result of the user includes an adjustment to the SQL structure, and the adjustment range is lower than a preset adjustment range threshold, adjusting the SQL structure based on the feedback result of the user to obtain an adjusted SQL structure; The adjusted SQL structure is executed and the execution process and execution result of the adjusted SQL structure are obtained, and the execution process and execution result of the adjusted SQL structure are visualized and displayed in real time.

[0047] Specifically, the selectively executing the SQL structure based on the user's feedback result and visually displaying the execution result in real time also includes: If the feedback result of the user includes an adjustment to the SQL structure, and the adjustment range exceeds a preset adjustment range threshold, the SQL structure is not executed, and the feedback result of the user is fed back to the user intention recognition model corresponding to the user; and retraining the user intention recognition model corresponding to the user based on the feedback result of the user; The recognition record of the user intention recognition is marked to facilitate repeated recognition of the user intention, thereby improving the matching degree between the user intention recognition model corresponding to the user and the user.

[0048] Specifically, the preset adjustment range threshold may be 5% or 8%. The setting of the preset adjustment range threshold here is only an example. In actual testing, those skilled in the art may set it according to actual needs, which will not be elaborated here.

[0049] Based on the above steps S101 to S105, the user's identity information is obtained and the user intention recognition model is retrieved to implement a user intention recognition model for each user, thereby improving the accuracy of the recognition of the user's natural language information. The user's natural language information is then substituted into the corresponding user intention recognition model to generate work order data corresponding to the natural language information, thereby implementing the automatic generation of work order data, and recording the recognition record of the user intention recognition model to record each recognition result for the convenience of subsequent model optimization and comparison of recognition results. Then, based on the work order data and the preset parsing database, a corresponding SQL structure is generated and fed back to the user. Then, combined with the user's feedback results, the SQL structure is selectively executed and the execution results are visualized and displayed in real time, thereby implementing intelligent interaction between the user and the generated SQL structure to ensure that the executed SQL structure meets the user's needs, thereby improving the intelligent management and analysis of the work order data, improving work efficiency, reducing labor costs, and avoiding the work order management method in the prior art that requires professionals to have in-depth technical knowledge and complex operation processes, making it difficult for ordinary users to directly use it, thereby affecting the efficiency and accuracy of work order management and increasing the technical problem of labor costs.

[0050] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effects of the present invention, different steps do not have to be performed in such an order, and they can be performed simultaneously (in parallel) or in other orders. These changes are within the scope of protection of the present invention.

[0051] Furthermore, the present invention also provides a report generation device based on big data intelligent interaction.

[0052] See attached Figure 2 , Figure 2 1 is a main structural block diagram of a report generation device based on big data intelligent interaction according to an embodiment of the present invention. Figure 2As shown, the report generation device based on big data intelligent interaction in the embodiment of the present invention interacts with multiple user terminals, the user intention recognition model 204 corresponding to each user of the device, the communication device 202, the natural language input device 205, the preset parsing database 203 and the control unit 201, the communication device 202 is connected to the multiple user terminals for communication, and is used to realize the information interaction between the device and the multiple user terminals, the natural language input device 205 is used to collect the natural language information of the user, the control unit 201 includes a processor 2011 and a memory 2012, the memory 2012 can be configured to store the program code 2013 for executing the report generation method based on big data intelligent interaction of the above method embodiment, the processor 2011 can be configured to execute the program code 2013 in the memory 2012, and the program code 2013 includes but is not limited to the program code 2013 for executing the report generation method based on big data intelligent interaction of the above method embodiment. For the convenience of explanation, only the part related to the embodiment of the present invention is shown, and the specific technical details are not disclosed, please refer to the method part of the embodiment of the present invention. The control unit 201 can be a control unit 201 device formed by various electronic devices.

[0053] Specifically, the device also includes a preset user portrait database and a preset user basic recognition model, wherein the user basic recognition model includes at least a preset basic language comparison library and a preset basic recognition module; The user portrait database stores the identity information of each user, the historical language data of each user, and the user portrait data of each user; The preset parsing database stores the SQL statement structure corresponding to each identification field of each dimension of each data level and the mapping logic corresponding to each SQL statement structure.

[0054] Specifically, the construction method of the preset user basic identification model 204 can adopt a trained neural network model or a trained machine learning model. The selection of the construction method of the preset user basic identification model here is only an exemplary description. In actual testing, technical personnel in this field can make a selection according to actual needs, which will not be repeated here.

[0055] Specifically, the natural language input device 205 can be a keyboard input or a voice input. The selection of the natural language input device 205 here is only an exemplary explanation. In actual testing, technical personnel in this field can make a selection according to actual needs, as long as it is possible to collect the user's natural language information through the natural language input device 205. No further details will be given here.

[0056] Specifically, the specific communication connection method between the communication device 202 and multiple user terminals is not restrictive. Technicians can set it according to actual usage requirements as long as the communication connection can be achieved. For example, it can be connected via WiFi or Bluetooth, etc. Technicians can set it according to actual usage requirements.

[0057] Specifically, the communication device 202 can be a wireless communication device in the prior art, or it can be a wired communication device. The selection of the communication device 202 here is only an exemplary description. In actual testing, technical personnel in this field can make a selection according to actual needs. As long as information interaction between the system and multiple user terminals can be achieved through the communication device 202, it will not be repeated here.

[0058] In one implementation, the description of the specific implementation functions may refer to steps S101 to S105.

[0059] The report generation device 200 based on big data intelligent interaction is used to execute Figure 1 The embodiment of the report generation method based on big data intelligent interaction shown in the figure has similar technical principles, technical problems solved and technical effects produced. Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working process and relevant instructions of the report generation device 200 based on big data intelligent interaction can refer to the contents described in the embodiment of the report generation method based on big data intelligent interaction, which will not be repeated here.

[0060] It is understood by those skilled in the art that the present invention implements all or part of the processes in the method of the above embodiment, and can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor 2011, the steps of each of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium that can carry the computer program code. It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electric carrier signal and telecommunication signal.

[0061] Furthermore, the report generation device 200 based on big data intelligent interaction of the present invention also includes a computer-readable storage medium. In a computer-readable storage medium embodiment according to the present invention, the computer-readable storage medium can be configured to store a program code 2013 for executing the report generation method based on big data intelligent interaction of the above method embodiment, and the program code 2013 can be loaded and run by the processor 2011 to implement the above report generation method based on big data intelligent interaction. For ease of explanation, only the parts related to the embodiment of the present invention are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present invention. The computer-readable storage medium can be a memory 2012 device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present invention is a non-temporary computer-readable storage medium.

[0062] Further, it should be understood that since the setting of each module is only for illustrating the functional units of the device of the present invention, the physical devices corresponding to these modules may be the processor 2011 itself, or a part of the software, a part of the hardware, or a part of the combination of software and hardware in the processor 2011. Therefore, the number of each module in the figure is only schematic.

[0063] Those skilled in the art will appreciate that the modules in the device can be adaptively split or merged. Such splitting or merging of specific modules will not cause the technical solution to deviate from the principle of the present invention, and therefore, the technical solutions after splitting or merging will fall within the protection scope of the present invention.

[0064] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A report generation method based on big data intelligent interaction, characterized in that: The method is applied to a report generation device based on big data intelligent interaction, wherein the device interacts with multiple user terminals, and the method comprises the following steps: Obtaining natural language information of a user and identity information of the user; Based on the identity information of the user, retrieve a user intention recognition model corresponding to the user, wherein the level of the user intention recognition model is one of primary, advanced and ultimate; Substituting the natural language information of the user into the user intention recognition model corresponding to the user, generating work order data corresponding to the natural language information, and recording the recognition record of the user intention recognition model; Based on the work order data and a preset parsing database, an SQL structure corresponding to the work order data is generated and fed back to the user; Based on the user's feedback result, the SQL structure is selectively executed and the execution result is visualized and displayed in real time.

2. The report generation method based on big data intelligent interaction according to claim 1 is characterized in that: Before calling the user intention recognition model corresponding to the user based on the identity information of the user, the method includes: Generate a user intention recognition model corresponding to each user; The user intention recognition model corresponding to each user is trained to obtain a trained user intention recognition model corresponding to the user.

3. The report generation method based on big data intelligent interaction according to claim 2 is characterized in that: Generating a user intention recognition model corresponding to each user includes: Obtaining a preset user basic recognition model, wherein the user basic recognition model at least includes a preset basic language comparison library and a preset basic recognition module; The identity information of each user and the preset user basic recognition model are obtained to generate a user intention recognition model corresponding to each user, wherein the level of the recognition training model is primary.

4. The report generation method based on big data intelligent interaction according to claim 3 is characterized in that: The step of training the user intention recognition model corresponding to each user to obtain the trained user intention recognition model corresponding to the user includes: Select any user as a training sample, and obtain a training sample set corresponding to the training sample, wherein the training sample set includes the identity information of the training sample and the user portrait data of the training sample; Based on the training sample set, the recognition training model is trained to obtain a training result; Based on the training results, selectively adjusting the recognition training model; and selectively adjusting the level of the recognition training model or the adjusted recognition training model based on the training result of the adjusted recognition training model or the training result of the recognition training model; Based on the level of the recognition training model or the adjusted recognition training model, and the training result of the recognition training model or the adjusted recognition training model, the training result of the user intention recognition model corresponding to the user corresponding to the training sample is determined.

5. The report generation method based on big data intelligent interaction according to claim 4 is characterized in that: The selecting any one user as a training sample and obtaining a training sample set corresponding to the training sample comprises: Select any user as a training sample; Obtaining a preset user portrait database, wherein the user portrait database stores identity information of each user, historical language data of each user, and user portrait data of each user; Based on the training samples and a preset user portrait database, a training sample set corresponding to the training samples is generated.

6. The report generation method based on big data intelligent interaction according to claim 4 is characterized in that: The training of the recognition training model based on the training sample set to obtain the training result includes: Based on the historical language data corresponding to the training samples in the preset user portrait database, multiple groups of language data sets corresponding to the training samples are generated, wherein the language data sets include at least one data level and multiple dimensions of natural language at each data level; Based on multiple language data sets, the recognition training model is trained to obtain the work order data generated corresponding to each language data set; Based on the user portrait data of the training samples in the training sample set, the work order data generated corresponding to each group of language data sets are tested to obtain the accuracy rate corresponding to each group of language data sets; Based on the accuracy rate corresponding to each group of language data sets, the training results of the recognition training model are obtained, wherein the training results include the comprehensive training accuracy rate and the proportion of the number of matches.

7. The report generation method based on big data intelligent interaction according to claim 6 is characterized in that: The determining, based on the level of the recognition training model or the adjusted recognition training model and the training result of the recognition training model or the adjusted recognition training model, of the training result of the user intention recognition model corresponding to the user corresponding to the training sample includes: If the level of the recognition training model is primary, and the training result of the recognition training model meets the preset additional training requirements, the recognition training model is used as the user intention recognition model corresponding to the user corresponding to the training sample, and it is determined that the user intention recognition model corresponding to the user is successfully trained; or, If the level of the recognition training model reaches the advanced level, and the training result of the recognition training model meets the preset training requirements, the recognition training model is used as the user intention recognition model corresponding to the user corresponding to the training sample, and it is determined that the user intention recognition model corresponding to the user is successfully trained; Otherwise, it is determined that the training of the user intention recognition model has failed, and the user intention recognition model is retrained.

8. The report generation method based on big data intelligent interaction according to claim 7 is characterized in that: The work order data at least includes identification fields that match each dimension of each data level of the natural language information, the preset parsing database stores SQL statement structures corresponding to each identification field of each dimension of each data level and mapping logic corresponding to each SQL statement structure, and the generating of the SQL structure corresponding to the work order data based on the work order data and the preset parsing database includes: Based on the work order data and a preset parsing database, determine the SQL statement structure corresponding to each identification field corresponding to the work order data and the mapping logic corresponding to each SQL structure; Based on the SQL statement structure corresponding to each identification field corresponding to the work order data and the mapping logic corresponding to each SQL statement structure, an SQL structure corresponding to the work order data is generated.

9. The report generation method based on big data intelligent interaction according to claim 8 is characterized in that: The selectively executing the SQL structure based on the user's feedback result and visually displaying the execution result in real time includes: If there is no adjustment to the SQL structure in the user's feedback structure, the SQL structure is executed and the execution process and execution result of the SQL structure are obtained, and the execution process and execution result of the SQL structure are visually displayed in real time; or, If the feedback result of the user includes an adjustment to the SQL structure, and the adjustment range is lower than a preset adjustment range threshold, adjusting the SQL structure based on the feedback result of the user to obtain an adjusted SQL structure; The adjusted SQL structure is executed and the execution process and execution result of the adjusted SQL structure are obtained, and the execution process and execution result of the adjusted SQL structure are visualized and displayed in real time.

10. A report generation device based on big data intelligent interaction, characterized in that: The device interacts with multiple user terminals, and includes a user intention recognition model corresponding to each user, a communication device, a natural language input device, a preset parsing database and a control unit. The communication device is communicatively connected to the multiple user terminals to realize information interaction between the system and the multiple user terminals. The natural language input device is used to collect natural language information of the user. The control unit includes a processor and a memory, and the memory is suitable for storing multiple program codes. The program code is suitable for being loaded and run by the processor to execute the report generation method based on big data intelligent interaction as described in any one of claims 1 to 9.

Images