CN117521606A - Intelligent report generation system and method for financial data - Google Patents
Intelligent report generation system and method for financial data Download PDFInfo
- Publication number
- CN117521606A CN117521606A CN202410010642.3A CN202410010642A CN117521606A CN 117521606 A CN117521606 A CN 117521606A CN 202410010642 A CN202410010642 A CN 202410010642A CN 117521606 A CN117521606 A CN 117521606A
- Authority
- CN
- China
- Prior art keywords
- financial data
- semantic
- financial
- report
- request
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 239000013598 vector Substances 0.000 claims abstract description 262
- 230000004927 fusion Effects 0.000 claims abstract description 116
- 230000004044 response Effects 0.000 claims abstract description 93
- 230000007246 mechanism Effects 0.000 claims abstract description 23
- 238000004140 cleaning Methods 0.000 claims abstract description 12
- 238000009826 distribution Methods 0.000 claims description 16
- 238000012937 correction Methods 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000004043 responsiveness Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 10
- 238000012545 processing Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000000638 solvent extraction Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/12—Use of codes for handling textual entities
- G06F40/126—Character encoding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/12—Accounting
- G06Q40/125—Finance or payroll
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses an intelligent report generation system and method for financial data. Firstly, data cleaning is carried out on financial data to obtain cleaned financial data, then semantic coding is carried out on the cleaned financial data to obtain financial data semantic feature vectors, then semantic coding is carried out on financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors, then responsiveness fusion based on a class attention mechanism is carried out on the sequence of report request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain report request-financial data semantic response fusion feature vectors serving as report request-financial data semantic response fusion features, and finally, a financial data report is generated based on the report request-financial data semantic response fusion features. Thus, the report generation requirements of different industries and scenes can be met.
Description
Technical Field
The present application relates to the field of intelligent report generation, and more particularly, to an intelligent report generation system and method for financial data.
Background
Financial data is an important basis for enterprise management and is also a focus of attention for investors and regulatory authorities. Financial statement is the basis for showing and analyzing the financial condition and performance of enterprises, and the accuracy of financial statement generation is critical to the management and management of enterprises. However, conventional financial data report generation systems generally require manual participation in report generation processes, including links such as data extraction, sorting, and report design, and such manual intervention is prone to errors, such as data entry errors, calculation errors, or format errors. And, the manual processing and arrangement of a large amount of financial data is a cumbersome task requiring a large amount of manpower and time. In addition, the conventional financial data report generation system is usually based on fixed templates and rules, can only generate reports with fixed formats and predefined formats, and cannot flexibly adapt to the requirements of different industries and scenes. Conventional systems often fail to meet if a user needs a customized report or is analyzed according to specific criteria.
Accordingly, an optimized intelligent report generating system for financial data is desired.
Disclosure of Invention
In view of the above, the present application provides an intelligent report generation system and method for financial data, which can improve the generation quality and efficiency of financial data report and reduce human intervention and errors, so as to meet the report generation requirements of different industries and scenes.
According to an aspect of the present application, there is provided an intelligent report generation method for financial data, including: acquiring financial data collected from a network data source; performing data cleaning on the financial data to obtain cleaned financial data; carrying out semantic coding on the cleaned financial data to obtain a financial data semantic feature vector; acquiring a financial statement request description submitted by a user; carrying out semantic coding on the financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors; performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain statement request-financial data semantic response fusion feature vectors as statement request-financial data semantic response fusion features; and generating a financial data report based on the report request-financial data semantic response fusion characteristics.
Further, performing semantic coding on the cleaned financial data to obtain financial data semantic feature vectors, including respectively passing the cleaned financial data through an embedding layer of a context encoder to respectively convert the cleaned financial data into embedding vectors to obtain a sequence of financial data embedding vectors; inputting the sequence of financial data embedding vectors into a converter of the context encoder to obtain a plurality of financial data semantic understanding feature vectors; and concatenating the plurality of financial data semantic understanding feature vectors to obtain the financial data semantic feature vector.
Further, performing semantic coding on the financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors, including: dividing the financial statement request description based on word granularity to obtain a sequence of financial statement request description words; and passing the sequence of financial report request descriptors through the context encoder to obtain the sequence of report request descriptor granularity semantic feature vectors.
Further, performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain a statement request-financial data semantic response fusion feature vector as a statement request-financial data semantic response fusion feature, including: performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors by using the following fusion formula to obtain the statement request-financial data semantic response fusion feature vectors; wherein, the fusion formula is:；/>；
wherein,representing the semantic feature vector of the financial data, +. >Representing 1 x->Matrix of->Equal to the dimension of the semantic feature vector of the financial data,/->Is 1 x->Matrix of->The number of the report request descriptor granularity semantic feature vectors in the sequence equal to the report request descriptor granularity semantic feature vectors is +.>Is a Sigmoid function->Is a weight coefficient>And->Convolution operation representing a 1 x 1 convolution kernel, < >>Representing each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +.>Representing each report request descriptor granularity in the sequence of the report request descriptor granularity semantic feature vectorsDimension of semantic feature vector->And representing the statement request-financial data semantic response fusion feature vector.
Further, generating a financial data report based on the report request-financial data semantic response fusion feature, including: performing feature distribution optimization on the report request-financial data semantic response fusion feature vector to obtain an optimized report request-financial data semantic response fusion feature vector; and the optimized report request-financial data semantic response fusion feature vector passes through an AIGC-based report generator to obtain a generated financial data report.
Further, performing feature distribution optimization on the report request-financial data semantic response fusion feature vector to obtain an optimized report request-financial data semantic response fusion feature vector, including: performing feature correction on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain corrected feature vectors; and fusing the correction feature vector and the report request-financial data semantic response fusion feature vector to obtain the optimized report request-financial data semantic response fusion feature vector.
According to another aspect of the present application, there is provided an intelligent report generating system for financial data, comprising: a financial data acquisition module for acquiring financial data acquired from a network data source; the data cleaning module is used for cleaning the financial data to obtain cleaned financial data; the financial data semantic coding module is used for carrying out semantic coding on the cleaned financial data to obtain a financial data semantic feature vector; the financial statement acquisition module is used for acquiring a financial statement request description submitted by a user; the financial statement semantic coding module is used for carrying out semantic coding on the financial statement request description to obtain a sequence of statement request description word granularity semantic feature vectors; the fusion module is used for carrying out response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain statement request-financial data semantic response fusion feature vectors as statement request-financial data semantic response fusion features; and the financial data report generation module is used for generating a financial data report based on the report request-financial data semantic response fusion characteristic.
Further, the financial data semantic coding module comprises: the financial data embedding and encoding unit is used for respectively passing the cleaned financial data through an embedding layer of the context encoder so as to respectively convert the cleaned financial data into embedding vectors to obtain a sequence of financial data embedding vectors; a financial data conversion unit for inputting the sequence of financial data embedding vectors into a converter of the context encoder to obtain a plurality of financial data semantic understanding feature vectors; and a concatenation unit, configured to concatenate the plurality of financial data semantic understanding feature vectors to obtain the financial data semantic feature vector.
Further, the financial statement semantic coding module comprises: the word granularity dividing unit is used for dividing the financial statement request description based on word granularity to obtain a sequence of financial statement request description words; and the financial report context coding unit is used for enabling the sequence of the financial report request descriptors to pass through the context coder to obtain the sequence of the report request descriptor granularity semantic feature vectors.
Further, the fusion module is configured to: performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors by using the following fusion formula to obtain the statement request-financial data semantic response fusion feature vectors; wherein, the fusion formula is: ；/>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing the semantic feature vector of the financial data, +.>Representing 1 x->Matrix of->Equal to the dimension of the semantic feature vector of the financial data,/->Is 1 x->Matrix of->The number of the report request descriptor granularity semantic feature vectors in the sequence equal to the report request descriptor granularity semantic feature vectors is +.>Is a Sigmoid function->Is a weight coefficient>And->Convolution operation representing a 1 x 1 convolution kernel, < >>Representing each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +.>Each report request in the sequence representing the report request descriptor granularity semantic feature vectorSolving the dimension of the descriptor granularity semantic feature vector, < ->And representing the statement request-financial data semantic response fusion feature vector.
Firstly, data cleaning is carried out on financial data to obtain cleaned financial data, then, semantic coding is carried out on the cleaned financial data to obtain financial data semantic feature vectors, then, semantic coding is carried out on financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors, then, responsiveness fusion based on a class attention mechanism is carried out on the sequence of report request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain report request-financial data semantic response fusion feature vectors as report request-financial data semantic response fusion features, and finally, a financial data report is generated based on the report request-financial data semantic response fusion features. Thus, the report generation requirements of different industries and scenes can be met.
Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present application and together with the description, serve to explain the principles of the present application.
FIG. 1 illustrates a flow chart of a method of intelligent report generation for financial data in accordance with an embodiment of the present application.
FIG. 2 illustrates an architectural diagram of a smart report generation method for financial data according to an embodiment of the present application.
Fig. 3 shows a flowchart of sub-step S170 of the intelligent report generation method for financial data according to an embodiment of the present application.
Fig. 4 shows a flowchart of sub-step S171 of the intelligent report generation method for financial data according to an embodiment of the present application.
FIG. 5 illustrates a block diagram of an intelligent report generating system for financial data in accordance with an embodiment of the present application.
FIG. 6 illustrates an application scenario diagram of a smart report generation method for financial data according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, are also within the scope of the present application.
As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present application. It will be understood by those skilled in the art that the present application may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits have not been described in detail as not to unnecessarily obscure the present application.
Aiming at the technical problems, the technical conception of the method is that financial data are collected from a network data source, financial report request description submitted by a user is obtained at the same time, and data processing and semantic understanding algorithms are introduced into the rear end to conduct semantic analysis of the financial data and the financial report request description, so that financial data reports of different types, formats and styles are automatically generated according to the financial data and the user requirements, and in such a way, the generation quality and efficiency of the financial data report can be improved, manual intervention and errors are reduced, and report generation requirements of different industries and scenes are met.
FIG. 1 illustrates a flow chart of a method of intelligent report generation for financial data in accordance with an embodiment of the present application. FIG. 2 illustrates an architectural diagram of a smart report generation method for financial data according to an embodiment of the present application. As shown in fig. 1 and 2, the intelligent report generating method for financial data according to the embodiment of the present application includes the steps of: s110, acquiring financial data acquired from a network data source; s120, performing data cleaning on the financial data to obtain cleaned financial data; s130, carrying out semantic coding on the cleaned financial data to obtain a financial data semantic feature vector; s140, acquiring a financial statement request description submitted by a user; s150, carrying out semantic coding on the financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors; s160, performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain statement request-financial data semantic response fusion feature vectors as statement request-financial data semantic response fusion features; and S170, generating a financial data report based on the report request-financial data semantic response fusion characteristics.
It should be appreciated that the purpose of step S110 is to collect financial data from a network data source, which may include financial statements of the company, financial indicators, transaction data, etc., which are the basis for generating the financial statements. In step S120, the financial data collected from the network data source is cleaned and preprocessed, and the cleaned data should be freed of erroneous, missing or redundant data, ensuring the accuracy and integrity of the data for subsequent processing and analysis. In step S130, the cleaned financial data is semantically encoded to be converted into semantic feature vectors of the financial data, the purpose of the semantic encoding being to convert the financial data into a machine understandable and processed form for subsequent analysis and application. In step S140, a description of the financial statement request submitted by the user is obtained, and the user may describe the desired financial statement content and requirements by entering text or other forms. In step S150, the financial statement request description submitted by the user is semantically encoded and converted into a semantic feature vector sequence with the granularity of the statement request description word, so that the request of the user can be converted into a form which can be understood and processed by a machine. In step S160, the sequence of semantic feature vectors of the report request descriptor granularity and the semantic feature vectors of the financial data are fused based on the responsiveness of the class-based attention mechanism, and the fusion process can correlate and match the report request of the user with the financial data to obtain semantic response fusion features of the report request and the financial data, and the features can help to generate the financial data report meeting the requirements of the user. In step S170, based on the statement request-financial data semantic response fusion feature, a financial data statement meeting the user' S requirement is generated, and according to the statement request and the financial data semantic response fusion feature of the user, the system can perform corresponding calculation and processing, generate a statement containing the required financial data, and present the statement to the user. By executing the steps, the complete process from the collection, cleaning and encoding of the financial data to the processing of the report request of the user and the generation of the financial data report can be realized.
Specifically, in the technical solution of the present application, first, financial data collected from network data sources, such as databases, files, web pages, and the like, is obtained. It should be appreciated that the accuracy of the financial data is critical to the accuracy of the report generation of the financial data, however, the financial data comes from different systems or departments, there are differences in format and standards, and there may be erroneous, missing, and outliers in the financial data. Thus, data cleansing of the financial data is required to obtain cleansed financial data before semantic understanding of the financial data is performed.
Then, after the financial data is cleaned, in order to enable semantic understanding of the financial data to capture semantic association relations among data items in the financial data, semantic encoding needs to be performed on the cleaned financial data to obtain semantic feature vectors of the financial data. By semantically encoding the cleaned financial data, semantic relationships between data items in the financial data, such as similarity, correlation, hierarchical structure and the like between indexes, can be captured. The system is helpful for better understanding of financial data, and accurate calculation and reasoning are performed in subsequent report generation and analysis, so that a foundation is provided for subsequent report generation and data analysis.
Accordingly, in step S130, performing semantic encoding on the cleaned financial data to obtain a semantic feature vector of the financial data, including: respectively passing the cleaned financial data through an embedding layer of a context encoder to respectively convert the cleaned financial data into embedding vectors to obtain a sequence of financial data embedding vectors; inputting the sequence of financial data embedding vectors into a converter of the context encoder to obtain a plurality of financial data semantic understanding feature vectors; and concatenating the plurality of financial data semantic understanding feature vectors to obtain the financial data semantic feature vector.
It is worth mentioning that the Embedding Layer (Embedding Layer) is a kind of neural network Layer for converting input data (in this case cleaned financial data) into Embedding vectors. An embedded vector is a low-dimensional representation that can map high-dimensional discrete data (e.g., words, symbols, or data points) into a continuous vector space. The embedding layer maps each data point into a vector by learning the distribution and semantic relation of the data, thereby capturing the semantic features of the data. A transducer is a neural network model for sequential data processing. In this context, the converter is used as a context encoder for converting a sequence of financial data embedding vectors into a plurality of financial data semantic understanding feature vectors. The transducer model is composed of multiple attention mechanisms, by which semantic and dependency relationships are captured at different locations in the sequence. The converter model has strong modeling capability, and can perform effective semantic understanding and coding on the sequence data. Thus, in step 130, the embedding layer is configured to convert the cleaned financial data into a sequence of embedding vectors, and the converter is configured to convert the sequence of embedding vectors into a plurality of semantic understanding feature vectors of the financial data, and finally, the feature vectors are cascade-operated to obtain semantic feature vectors of the financial data, wherein the semantic feature vectors include semantic information of the cleaned financial data.
Then, in order to understand the needs and requirements of the user in order to generate a financial statement that meets the user's expectations, it is necessary to first obtain a financial statement request description submitted by the user. In particular, in one specific example of the present application, users may submit their needs for financial statements through text descriptions, voice input, or other forms, such as requiring generation of profit sheets, cash flow sheets, or analysis reports of specific indicators for a certain period of time, or the like. And then, carrying out semantic coding on the financial report request description to extract context semantic association characteristic information based on word granularity in the financial report request description, thereby obtaining a sequence of semantic characteristic vectors of the report request description word granularity. Thus, the system can be helped to more accurately understand the report request description of the user, such as understand the report type, time range, index requirement and the like which are required to be generated by the user, thereby being beneficial to generating the report which meets the user's expectations.
Accordingly, in step S150, performing semantic encoding on the financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors, including: dividing the financial statement request description based on word granularity to obtain a sequence of the financial statement request description words; and passing the sequence of financial report request descriptors through the context encoder to obtain the sequence of report request descriptor granularity semantic feature vectors.
It should be appreciated that word granularity based partitioning is the partitioning of a financial statement request description into a sequence of words, which is done to break down the textual information of the description into individual words for subsequent semantic encoding and processing. The division based on word granularity has several uses: 1. semantic understanding: dividing the financial statement request description into words can help the system understand the meaning and semantics of each word. Each word may represent a particular concept or requirement, and the system may better understand the intent and needs of the user by semantically encoding each word. 2. Context modeling: sequences divided into words can help the system model context information. Each word can infer its meaning and semantics from the words that follow it. By dividing the financial statement request description into a sequence of words and inputting it into the context encoder, the system can capture the context and dependencies between words, thereby better understanding the semantics of the entire statement request. 3. Feature extraction: the division based on word granularity may translate the financial statement request description into a series of words, each of which may represent a particular feature or requirement. By semantically encoding each term, it can be converted into semantic feature vectors that can be used for subsequent feature extraction and matching to generate a financial data report that meets the needs of the user. In summary, the word granularity based partitioning can convert the financial statement request description into a sequence of processable words, thereby helping the system understand the needs of the user, model context information, and extract semantic features to generate a financial data statement.
Further, since the sequence of report request descriptor granularity semantic feature vectors provides report request descriptor granularity semantic feature information about user input, the financial data semantic feature vectors provide semantic association feature information about financial report data. In order to effectively fuse semantic features between a report request and financial data so as to help a system to better understand user requirements and map the user requirements to corresponding financial data to generate a report meeting user expectations, in the technical scheme of the application, the sequence of the report request descriptor granularity semantic feature vector and the financial data semantic feature vector are further fused based on responsiveness of a class attention mechanism so as to obtain a report request-financial data semantic response fusion feature vector. In particular, the manner of responsive fusion based on the class attention mechanism can weight semantic features of financial data according to word granularity semantic features described by report requests, so that the system can pay attention to the financial data related to the report requests and ignore data irrelevant to the requests, which is helpful to improving understanding and response capability of the system to user demands.
Accordingly, in step S160, performing a response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vector and the financial data semantic feature vector to obtain a statement request-financial data semantic response fusion feature vector as a statement request-financial data semantic response fusion feature, including: performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors by using the following fusion formula to obtain the statement request-financial data semantic response fusion feature vectors; wherein, the fusion formula is:；the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Representing the semantic feature vector of the financial data, +.>Representing 1 x->Matrix of->Equal to the dimension of the semantic feature vector of the financial data,/->Is 1 x->Matrix of->The number of the report request descriptor granularity semantic feature vectors in the sequence equal to the report request descriptor granularity semantic feature vectors is +.>Is a Sigmoid function->Is a weight coefficient>And->Convolution operation representing a 1 x 1 convolution kernel, < >>Representing each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +. >Representing the scale of each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +_>And representing the statement request-financial data semantic response fusion feature vector.
And then, the report request-financial data semantic response fusion feature vector passes through an AIGC-based report generator to obtain a generated financial data report. Thus, financial data reports of different types, formats and styles can be automatically generated according to the financial data and the user requirements, so that the generation quality and efficiency of the financial data report are improved, and manual intervention and errors are reduced.
Accordingly, in step S170, as shown in fig. 3, based on the report request-financial data semantic response fusion feature, a financial data report is generated, including: s171, carrying out feature distribution optimization on the statement request-financial data semantic response fusion feature vector to obtain an optimized statement request-financial data semantic response fusion feature vector; and S172, enabling the optimized report request-financial data semantic response fusion feature vector to pass through an AIGC-based report generator to obtain a generated financial data report.
It should be understood that, in step S171, the purpose of optimizing the feature distribution of the report request-financial data semantic response fusion feature vector is to better meet the requirements for generating the financial data report by adjusting the weight and distribution of the feature vector. By optimizing the distribution of feature vectors, the degree of matching between the report request and the financial data can be improved, thereby generating a more accurate and useful financial data report. In step S172, the optimized report request-financial data semantic response fusion feature vector is input into an AIGC-based (Artificial Intelligence for Generating Content, artificial intelligence generated content) report generator to generate a final financial data report. AIGC is an artificial intelligence based content generation technology, which utilizes techniques such as machine learning, natural language processing and the like to automatically generate a financial data report meeting the requirements according to input feature vectors and rules. And through the report generator of the AIGC, a financial data report can be quickly generated according to the optimized feature vector, and the generation efficiency and accuracy are improved. In summary, step S171 adjusts the weight and distribution of the feature vector fused by the report request-financial data semantic response through feature distribution optimization to improve the matching degree, step S172 generates a final financial data report from the optimized feature vector by using an AIGC-based report generator, and the two steps are used together to generate an accurate and useful financial data report.
In step S171, as shown in fig. 4, performing feature distribution optimization on the report request-financial data semantic response fusion feature vector to obtain an optimized report request-financial data semantic response fusion feature vector, including: s1711, performing feature correction on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain corrected feature vectors; and S1712, fusing the correction feature vector and the report request-financial data semantic response fusion feature vector to obtain the optimized report request-financial data semantic response fusion feature vector.
In particular, in the above technical solution, the sequence of the report request descriptor granularity semantic feature vector and the financial data semantic feature vector express the word granularity text semantic feature of the financial report request description and the encoding semantic feature of the financial data respectively, so that when the response fusion based on the class attention mechanism is performed on the sequence of the report request descriptor granularity semantic feature vector and the financial data semantic feature vector, the semantic feature fusion effect based on different data modalities of the sequence of the report request descriptor granularity semantic feature vector and the financial data semantic feature vector needs to be improved.
That is, considering the financial statement request description and the source data modality of the financial data, it is desirable to suppress Gao Weiji which changes corresponding to different feature distribution directions due to source data modality differences when it is fused to the high-dimensional geometric transformation-based features in the high-dimensional feature space, thereby improving the fusion effect.
Based on the above, the applicant of the present application further performs fusion feature geometric variation correction based on rotation control on the sequence of the report request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain corrected feature vectors.
Accordingly, in one example, in step S1711, performing feature correction on the sequence of report request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain corrected feature vectors includes: performing feature correction on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors by using the following correction formula to obtain corrected feature vectors; wherein, the correction formula is:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Is a cascading feature vector after cascading the sequence of the semantic feature vector with the granularity of the report request descriptor,/- >Is the financial data semantic feature vector, +.>、And->The feature values of the cascade feature vector, the financial data semantic feature vector and the correction feature vector, respectively,/->And->Respectively 1-norm and 2-norm of the feature vector,/->Is the length of the feature vector, and +.>Is a weight superparameter,/->An exponential operation representing a numerical value, the exponential operation representing the calculation of a natural exponential function value that is a power of the numerical value.
Specifically, in order to promote the perception and cognition capability of the feature to the spatial transformation corresponding to different feature distributions when the response fusion based on the attention-like mechanism is carried out, the rotation control of the feature distribution of the feature vector in different directions is carried out from the vector dimension based on the sequence of the statement request descriptor granularity semantic feature vector and the distance structure parameter of the financial data semantic feature vector, so that the rotation invariance of the fused feature is reserved through relative rotation unwrapping, and the high-dimensional geometric change of the feature distribution caused by the geometric transformation in the high-dimensional feature space when the feature is fused is avoided. Therefore, the correction feature vector is fused with the statement request-financial data semantic response fusion feature vector, so that the fusion expression effect of the statement request-financial data semantic response fusion feature vector can be improved, and the data quality of the generated financial data statement obtained by the statement generator based on AIGC is improved. Thus, financial data reports of different types, formats and styles can be automatically generated according to the financial data and the user requirements, so that the generation quality and efficiency of the financial data report are improved, manual intervention and errors are reduced, and the report generation requirements of different industries and scenes are met.
In summary, according to the intelligent report generation method for financial data, which is based on the embodiment of the application, the generation quality and efficiency of the financial data report can be improved, and manual intervention and errors are reduced, so that the report generation requirements of different industries and scenes are met.
FIG. 5 illustrates a block diagram of an intelligent report generating system 100 for financial data according to an embodiment of the present application. As shown in fig. 5, an intelligent report generating system 100 for financial data according to an embodiment of the present application includes: a financial data acquisition module 110 for acquiring financial data collected from a network data source; the data cleaning module 120 is configured to perform data cleaning on the financial data to obtain cleaned financial data; a financial data semantic coding module 130, configured to perform semantic coding on the cleaned financial data to obtain a financial data semantic feature vector; the financial statement acquisition module 140 is used for acquiring a financial statement request description submitted by a user; the financial statement semantic coding module 150 is configured to perform semantic coding on the financial statement request description to obtain a sequence of statement request descriptor granularity semantic feature vectors; the fusion module 160 is configured to perform responsiveness fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain a statement request-financial data semantic response fusion feature vector as a statement request-financial data semantic response fusion feature; and a financial data report generation module 170 for generating a financial data report based on the report request-financial data semantic response fusion characteristics.
In one possible implementation, the financial data semantic coding module 130 includes: the financial data embedding and encoding unit is used for respectively passing the cleaned financial data through an embedding layer of the context encoder so as to respectively convert the cleaned financial data into embedding vectors to obtain a sequence of financial data embedding vectors; a financial data conversion unit for inputting the sequence of financial data embedding vectors into a converter of the context encoder to obtain a plurality of financial data semantic understanding feature vectors; and a concatenation unit, configured to concatenate the plurality of financial data semantic understanding feature vectors to obtain the financial data semantic feature vector.
In one possible implementation, the financial statement semantic coding module 150 includes: the word granularity dividing unit is used for dividing the financial statement request description based on word granularity to obtain a sequence of the financial statement request description words; and the financial report context coding unit is used for enabling the sequence of the financial report request descriptors to pass through the context coder to obtain the sequence of the report request descriptor granularity semantic feature vectors.
In one possible implementation, the fusing module 160 is configured to: performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors by using the following fusion formula to obtain the statement request-financial data semantic response fusion feature vectors; wherein, the fusion formula is:；/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein,representing the semantic feature vector of the financial data, +.>Representing 1 x->Matrix of->Equal to the dimension of the semantic feature vector of the financial data,/->Is 1 x->Matrix of->The number of the report request descriptor granularity semantic feature vectors in the sequence equal to the report request descriptor granularity semantic feature vectors is +.>Is a Sigmoid function->As the weight coefficient of the light-emitting diode,and->Convolution operation representing a 1 x 1 convolution kernel, < >>Representing each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +.>Representing granularity semantic features of the report request descriptorsThe scale of semantic feature vectors of each report request descriptor granularity in the sequence of vectors, +.>And representing the statement request-financial data semantic response fusion feature vector.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described intelligent report generating system for financial data 100 have been described in detail in the above description of the intelligent report generating method for financial data with reference to fig. 1 to 4, and thus, repetitive descriptions thereof will be omitted.
As described above, the intelligent report generating system 100 for financial data according to the embodiment of the present application may be implemented in various wireless terminals, for example, a server or the like having an intelligent report generating algorithm for financial data. In one possible implementation, the intelligent report generating system 100 for financial data according to embodiments of the present application may be integrated into a wireless terminal as one software module and/or hardware module. For example, the intelligent report generating system 100 for financial data may be a software module in the operating system of the wireless terminal or may be an application developed for the wireless terminal; of course, the intelligent report generating system 100 for financial data may also be one of the many hardware modules of the wireless terminal.
Alternatively, in another example, the intelligent report generating system for financial data 100 and the wireless terminal may be separate devices, and the intelligent report generating system for financial data 100 may be connected to the wireless terminal through a wired and/or wireless network and transmit interactive information in a contracted data format.
FIG. 6 illustrates an application scenario diagram of a smart report generation method for financial data according to an embodiment of the present application. As shown in fig. 6, in this application scenario, first, financial data collected from a network data source (e.g., D1 illustrated in fig. 6) and a financial report request description submitted by a user (e.g., D2 illustrated in fig. 6) are acquired, and then the financial data and the financial report request description are input into a server (e.g., S illustrated in fig. 6) where an intelligent report generation algorithm for financial data is deployed, wherein the server can process the financial data and the financial report request description using the intelligent report generation algorithm for financial data to generate a financial data report.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The embodiments of the present application have been described above, the foregoing description is exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. Intelligent report generator for financial dataA method, comprising: acquiring financial data collected from a network data source; performing data cleaning on the financial data to obtain cleaned financial data; carrying out semantic coding on the cleaned financial data to obtain a financial data semantic feature vector; acquiring a financial statement request description submitted by a user; carrying out semantic coding on the financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors; performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain statement request-financial data semantic response fusion feature vectors as statement request-financial data semantic response fusion features; generating a financial data report based on the report request-financial data semantic response fusion characteristics; the method for performing response fusion on the sequence of the statement request descriptor granularity semantic feature vector and the financial data semantic feature vector based on a class attention mechanism to obtain a statement request-financial data semantic response fusion feature vector as a statement request-financial data semantic response fusion feature comprises the following steps: performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors by using the following fusion formula to obtain the statement request-financial data semantic response fusion feature vectors; wherein, the fusion formula is: Wherein (1)>Representing the semantic feature vector of the financial data, +.>Representing 1 x->Matrix of->Equal to the dimension of the semantic feature vector of the financial data,/->Is 1 x->Matrix of->The number of the report request descriptor granularity semantic feature vectors in the sequence equal to the report request descriptor granularity semantic feature vectors is +.>Is a Sigmoid function->Is a weight coefficient>And->Convolution operation representing a 1 x 1 convolution kernel, < >>Representing each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +.>Representing the scale of each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +_>And representing the statement request-financial data semantic response fusion feature vector.
2. The intelligent report generating method for financial data according to claim 1, wherein semantically encoding the cleaned financial data to obtain a financial data semantic feature vector, comprising: respectively passing the cleaned financial data through an embedding layer of a context encoder to respectively convert the cleaned financial data into embedding vectors to obtain a sequence of financial data embedding vectors; inputting the sequence of financial data embedding vectors into a converter of the context encoder to obtain a plurality of financial data semantic understanding feature vectors; and concatenating the plurality of financial data semantic understanding feature vectors to obtain the financial data semantic feature vector.
3. The intelligent report generating method for financial data according to claim 2, wherein semantically encoding the financial report request description to obtain a sequence of report request descriptor granularity semantic feature vectors comprises: dividing the financial statement request description based on word granularity to obtain a sequence of financial statement request description words; and passing the sequence of financial report request descriptors through the context encoder to obtain the sequence of report request descriptor granularity semantic feature vectors.
4. A method of generating a statement of financial data as claimed in claim 3, wherein generating a statement of financial data based on the statement request-financial data semantic response fusion feature comprises: performing feature distribution optimization on the report request-financial data semantic response fusion feature vector to obtain an optimized report request-financial data semantic response fusion feature vector; and the optimized report request-financial data semantic response fusion feature vector passes through an AIGC-based report generator to obtain a generated financial data report.
5. The intelligent report generating method for financial data according to claim 4, wherein performing feature distribution optimization on the report request-financial data semantic response fusion feature vector to obtain an optimized report request-financial data semantic response fusion feature vector comprises: performing feature correction on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain corrected feature vectors; and fusing the correction feature vector and the report request-financial data semantic response fusion feature vector to obtain the optimized report request-financial data semantic response fusion feature vector.
6. An intelligent report generating system for financial data, comprising: a financial data acquisition module for acquiring financial data acquired from a network data source; the data cleaning module is used for cleaning the financial data to obtain cleaned financial data; the financial data semantic coding module is used for carrying out semantic coding on the cleaned financial data to obtain a financial data semantic feature vector; the financial statement acquisition module is used for acquiring a financial statement request description submitted by a user; the financial statement semantic coding module is used for carrying out semantic coding on the financial statement request description to obtain a sequence of statement request description word granularity semantic feature vectors; the fusion module is used for carrying out response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors to obtain statement request-financial data semantic response fusion feature vectors as statement request-financial data semantic response fusion features; the financial data report generation module is used for generating a financial data report based on the report request-financial data semantic response fusion characteristics; wherein, the fusion module is used for: performing response fusion based on a class attention mechanism on the sequence of the statement request descriptor granularity semantic feature vectors and the financial data semantic feature vectors by using the following fusion formula to obtain the statement request-financial data semantic response fusion feature vectors; wherein, the fusion formula is: Wherein (1)>Representing the semantic feature vector of the financial data, +.>Representing 1 x->Matrix of->Equal to the dimension of the semantic feature vector of the financial data,is 1 x->Matrix of->The number of the report request descriptor granularity semantic feature vectors in the sequence equal to the report request descriptor granularity semantic feature vectors is +.>Is a Sigmoid function->Is a weight coefficient>And->Convolution operation representing a 1 x 1 convolution kernel, < >>Representing each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +.>Representing the scale of each report request descriptor granularity semantic feature vector in the sequence of report request descriptor granularity semantic feature vectors, +_>And representing the statement request-financial data semantic response fusion feature vector.
7. An intelligent report generating system for financial data according to claim 6, wherein said financial data semantic coding module comprises: the financial data embedding and encoding unit is used for respectively passing the cleaned financial data through an embedding layer of the context encoder so as to respectively convert the cleaned financial data into embedding vectors to obtain a sequence of financial data embedding vectors; a financial data conversion unit for inputting the sequence of financial data embedding vectors into a converter of the context encoder to obtain a plurality of financial data semantic understanding feature vectors; and a concatenation unit, configured to concatenate the plurality of financial data semantic understanding feature vectors to obtain the financial data semantic feature vector.
8. The intelligent report generating system for financial data of claim 7, wherein the financial report semantic coding module comprises: the word granularity dividing unit is used for dividing the financial statement request description based on word granularity to obtain a sequence of financial statement request description words; and the financial report context coding unit is used for enabling the sequence of the financial report request descriptors to pass through the context coder to obtain the sequence of the report request descriptor granularity semantic feature vectors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410010642.3A CN117521606B (en) | 2024-01-04 | 2024-01-04 | Intelligent report generation system and method for financial data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410010642.3A CN117521606B (en) | 2024-01-04 | 2024-01-04 | Intelligent report generation system and method for financial data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117521606A true CN117521606A (en) | 2024-02-06 |
| CN117521606B CN117521606B (en) | 2024-03-19 |
Family
ID=89753423
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410010642.3A Active CN117521606B (en) | 2024-01-04 | 2024-01-04 | Intelligent report generation system and method for financial data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117521606B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2303886A1 (en) * | 1997-09-19 | 1999-03-25 | William J. Clancey | Creating and editing electronic documents |
| US20050086135A1 (en) * | 2003-10-20 | 2005-04-21 | Li-Chin Lu | Automatic bookkeeping system |
| EP1705606A1 (en) * | 2005-03-21 | 2006-09-27 | Li-Chih Lu | Automatic bookkeeping system |
| US7610233B1 (en) * | 1999-12-22 | 2009-10-27 | Accenture, Llp | System, method and article of manufacture for initiation of bidding in a virtual trade financial environment |
| JP2011076557A (en) * | 2009-10-02 | 2011-04-14 | Pronexus Inc | Database and providing system of business financial information |
| WO2012090222A1 (en) * | 2010-12-29 | 2012-07-05 | Esssar Investments Limited | System and method for converting & presenting financial information |
| US20210201359A1 (en) * | 2019-12-30 | 2021-07-01 | Genesys Telecommunications Laboratories, Inc. | Systems and methods relating to automation for personalizing the customer experience |
| US20210216862A1 (en) * | 2020-01-15 | 2021-07-15 | Beijing Jingdong Shangke Information Technology Co., Ltd. | System and method for semantic analysis of multimedia data using attention-based fusion network |
| KR20220101817A (en) * | 2021-01-12 | 2022-07-19 | 최희준 | Method and system of an artificial intelligence for predicting financial information |
| CN116150361A (en) * | 2022-12-27 | 2023-05-23 | 暨南大学 | Event extraction method, system and storage medium for financial statement notes |
| CN116610803A (en) * | 2023-07-19 | 2023-08-18 | 北京每日信动科技有限公司 | Industrial chain excellent enterprise information management method and system based on big data |
| CN116842964A (en) * | 2023-07-18 | 2023-10-03 | 杭州鑫策科技有限公司 | Business process generation method and system based on semantic analysis |
-
2024
- 2024-01-04 CN CN202410010642.3A patent/CN117521606B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2303886A1 (en) * | 1997-09-19 | 1999-03-25 | William J. Clancey | Creating and editing electronic documents |
| US7610233B1 (en) * | 1999-12-22 | 2009-10-27 | Accenture, Llp | System, method and article of manufacture for initiation of bidding in a virtual trade financial environment |
| US20050086135A1 (en) * | 2003-10-20 | 2005-04-21 | Li-Chin Lu | Automatic bookkeeping system |
| EP1705606A1 (en) * | 2005-03-21 | 2006-09-27 | Li-Chih Lu | Automatic bookkeeping system |
| JP2011076557A (en) * | 2009-10-02 | 2011-04-14 | Pronexus Inc | Database and providing system of business financial information |
| WO2012090222A1 (en) * | 2010-12-29 | 2012-07-05 | Esssar Investments Limited | System and method for converting & presenting financial information |
| US20210201359A1 (en) * | 2019-12-30 | 2021-07-01 | Genesys Telecommunications Laboratories, Inc. | Systems and methods relating to automation for personalizing the customer experience |
| US20210216862A1 (en) * | 2020-01-15 | 2021-07-15 | Beijing Jingdong Shangke Information Technology Co., Ltd. | System and method for semantic analysis of multimedia data using attention-based fusion network |
| KR20220101817A (en) * | 2021-01-12 | 2022-07-19 | 최희준 | Method and system of an artificial intelligence for predicting financial information |
| CN116150361A (en) * | 2022-12-27 | 2023-05-23 | 暨南大学 | Event extraction method, system and storage medium for financial statement notes |
| CN116842964A (en) * | 2023-07-18 | 2023-10-03 | 杭州鑫策科技有限公司 | Business process generation method and system based on semantic analysis |
| CN116610803A (en) * | 2023-07-19 | 2023-08-18 | 北京每日信动科技有限公司 | Industrial chain excellent enterprise information management method and system based on big data |
Non-Patent Citations (1)
| Title |
|---|
| 李娜;赵歆波;: "一种整合语义对象特征的视觉注意力模型", 哈尔滨工业大学学报, no. 05, 10 May 2020 (2020-05-10) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117521606B (en) | 2024-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109033068B (en) | Method and device for reading and understanding based on attention mechanism and electronic equipment | |
| CN115203380A (en) | Text processing system and method based on multi-mode data fusion | |
| CN115146488B (en) | Variable business process intelligent modeling system and method based on big data | |
| CN115783923B (en) | Elevator fault mode identification system based on big data | |
| CN115796173A (en) | Data processing method and system for supervision submission requirements | |
| CN116629275B (en) | Intelligent decision support system and method based on big data | |
| CN112052684A (en) | Named entity identification method, device, equipment and storage medium for power metering | |
| CN116245513B (en) | Automatic operation and maintenance system and method based on rule base | |
| CN116911929B (en) | Advertisement service terminal and method based on big data | |
| CN115145551A (en) | Intelligent auxiliary system for machine learning application low-code development | |
| CN115878003B (en) | Method and system for automating RPA webpage operation based on Transformer | |
| CN114332519A (en) | Image description generation method based on external triple and abstract relation | |
| CN116151845A (en) | Product full life cycle management system and method based on industrial Internet of things technology | |
| CN115409018A (en) | Company public opinion monitoring system and method based on big data | |
| CN112784580A (en) | Financial data analysis method and device based on event extraction | |
| CN115269882A (en) | Intellectual property retrieval system and method based on semantic understanding | |
| CN117316462A (en) | Medical data management method | |
| CN117521606B (en) | Intelligent report generation system and method for financial data | |
| CN117744855A (en) | Load prediction system and method based on machine learning | |
| CN115309895A (en) | Progress management system based on research and development project text description and management method thereof | |
| CN114726870A (en) | Hybrid cloud resource arrangement method and system based on visual dragging and electronic equipment | |
| CN116681428B (en) | Intelligent recycling management system and method for electronic equipment | |
| CN117171413B (en) | Data processing system and method for digital collection management | |
| CN118095294B (en) | Project document generation method and device based on artificial intelligence and electronic equipment | |
| CN116977021B (en) | Automatic pushing method for system butt joint based on big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |