A service problem analysis method and device, electronic equipment, medium and product

Abstract

The application discloses a kind of service problem analysis method, device, electronic equipment, medium and product.The method comprises: obtaining vehicle-related service problem;The service problem is analyzed, and service intent is determined;Determine the target function module matched with the service problem;Wherein, the target function module is composed of at least one function algorithm;The function algorithm is used to construct the data processing logic of service problem analysis;According to the service intent, determine target function algorithm from the function algorithm of the target function module;According to the target function algorithm, generate the business report corresponding to the service problem.This technical solution can realize service problem intelligent matching function module, automatically optimize algorithm dimension and output business report, greatly reduce the use threshold of vehicle networking big data platform, improve the efficiency of whole vehicle research and development data analysis and platform function utilization rate.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a business problem analysis method, apparatus, electronic device, medium, and product. Background Technology

[0002] As intelligent connected vehicle technology matures, vehicle-to-everything (V2X) big data platforms aggregate massive amounts of vehicle operation data, including multi-dimensional monitoring information such as energy consumption, faults, operating conditions, and exhaust emissions. Big data platforms for vehicle R&D operations deploy dozens of data analysis modules, typically including energy consumption analysis, fault diagnosis, vehicle profiling, and user profiling. Each module is configured with multiple analysis algorithm dimensions and supports configuration of various input parameters such as vehicle filtering and statistical time periods.

[0003] Existing big data analytics systems still have significant shortcomings: First, the platform's functional modules are numerous and highly specialized, making it difficult for developers to quickly locate target functions and requiring manual consultation with maintenance personnel. The system does not support automatic matching of the optimal module based on natural language-based business requirements, resulting in a long learning curve for new users and low overall platform utilization. Second, the analytical algorithm dimensions must be manually selected by the user. For example, in energy consumption research, the system includes multiple analytical dimensions such as vehicle speed-fuel consumption correlation trends, road type mileage distribution, terrain mileage distribution, and total mass mileage distribution. Algorithm selection is highly dependent on the operator's professional experience, lacking sufficient intelligence. Third, the platform only outputs raw visualization results such as data tables and bar charts, requiring industry experts to combine their knowledge of vehicle engineering for manual analysis to extract research conclusions.

[0004] In summary, the industry urgently needs an intelligent agent analysis system that is deeply integrated with the vehicle networking big data platform, relies on domain expertise, and can achieve closed-loop processing across the entire chain. Summary of the Invention

[0005] This invention provides a business problem analysis method, device, electronic device, medium, and product, which can realize intelligent matching function modules for business problems, automatically select the best algorithm dimensions, and output business reports, significantly reducing the threshold for using the Internet of Vehicles big data platform and improving the efficiency of vehicle R&D data analysis and the utilization rate of platform functions.

[0006] According to one aspect of the present invention, a business problem analysis method is provided, the method comprising: Obtain vehicle-related business information; The business problem is analyzed to determine the business intent; Identify target functional modules that match the business problem; wherein, the target functional modules consist of at least one functional algorithm; the functional algorithms are used to construct the data processing logic for business problem analysis; Based on the stated business intent, the target functional algorithm is determined from the functional algorithms of the target functional module; Based on the target function algorithm, a business report corresponding to the business problem is generated.

[0007] According to another aspect of the present invention, a business problem analysis apparatus is provided, the apparatus comprising: The business issue acquisition unit is used to acquire vehicle-related business issues; The business intent determination unit is used to analyze the business problem and determine the business intent; A target functional module determination unit is used to determine target functional modules that match the business problem; wherein, the target functional module consists of at least one functional algorithm; the functional algorithm is used to construct the data processing logic for business problem analysis; The target function algorithm determination unit is used to determine the target function algorithm from the function algorithms of the target function module based on the business intent. The business report generation unit is used to generate a business report corresponding to the business problem based on the target function algorithm.

[0008] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a business problem analysis method according to any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement a business problem analysis method according to any embodiment of the present invention.

[0010] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements a business problem analysis method as described in any embodiment of the present invention.

[0011] The technical solution of this invention involves: acquiring vehicle-related business issues; analyzing the business issues to determine the business intent; identifying target functional modules that match the business issues; determining a target functional algorithm from the functional algorithms of the target functional modules based on the business intent; and generating a business report corresponding to the business issues based on the target functional algorithm. This technical solution enables intelligent matching of functional modules to business issues, automatic optimization of algorithm dimensions, and output of business reports, significantly reducing the barrier to entry for using vehicle networking big data platforms and improving the efficiency of vehicle R&D data analysis and platform functionality utilization.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of a business problem analysis method provided in Embodiment 1 of the present invention; Figure 2 A flowchart of a business problem analysis process provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the structure of a business problem analysis device provided in Embodiment 3 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device that implements a business problem analysis method according to an embodiment of the present invention. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0016] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] Example 1 Figure 1 This is a flowchart of a business problem analysis method according to Embodiment 1 of the present invention. This embodiment is applicable to the analysis of vehicle business problems. The method can be executed by a business problem analysis device, which can be implemented in hardware and / or software and can be configured in a device. For example, the device can be a backend server or other device with communication and computing capabilities. Figure 1 As shown, the method includes: S110, Obtain vehicle-related business issues.

[0018] This solution utilizes a four-stage closed-loop architecture—requirement understanding, function matching, algorithm recommendation, and conclusion generation—to construct an intelligent analysis method suitable for R&D scenarios. To build a two-layer domain knowledge base system, a mechanism is established to complete function and algorithmic intelligent recommendations based on knowledge base retrieval. Simultaneously, analysis conclusions are automatically generated using Retrieval-Augmented Generation (RAG) technology. Specifically, this includes: A two-tiered system—a functional knowledge base and a domain knowledge base—is established for connected vehicle R&D scenarios. This system provides accurate and traceable knowledge support for function recommendation, algorithm recommendation, and conclusion generation. This system is the core foundation of this solution; subsequent function recommendation, algorithm recommendation, and conclusion generation all rely on this knowledge base for enhanced retrieval.

[0019] The functional knowledge base stores the functional semantic description documents corresponding to all functional modules of the platform. The functional semantic description documents are standardized functional description files oriented towards large language models. They clearly define the function name, business definition, applicable scenarios, applicable scenario descriptions for each algorithm dimension, and functional characteristic tags in natural language. They are the core knowledge carrier that ensures the matching of semantic retrieval and functional recommendation.

[0020] The domain knowledge base stores professional knowledge in the field of vehicle engineering and serves as the core support for generating analytical conclusions. Knowledge sources include internal enterprise technical standards, industry normative documents, accumulated expert experience, and historical analysis reports. Content covers categories such as energy consumption evaluation indicators, algorithm dimension interpretation rules, root causes of energy consumption anomalies, operating condition classification standards, and conclusion templates. Table 1 shows the dimensions of the domain knowledge base.

[0021] Table 1 The knowledge base construction steps are as follows: All functional modules within the vehicle data platform are traversed, and staff compile functional description documents for each module. The documents include a unique function ID, business text description, a list of applicable business scenarios, application conditions for each analysis algorithm, functional attribute tags, and related module information. After the documents are compiled, word embedding models are used to convert the text descriptions within the documents into high-dimensional numerical vectors, which are then uniformly stored in a vector database. Word embedding models can be Word2Vec, GloVe, BERT, etc. Word2Vec predicts words from context to obtain static word vectors; GloVe trains fixed word vectors based on global word co-occurrence frequencies; both are static embeddings, with each word corresponding to only one unique vector. The BERT word embedding layer relies on a bidirectional Transformer to capture contextual information and generate dynamic word vectors, which can distinguish the different semantics of the same word in different scenarios.

[0022] Knowledge base construction steps: First, collect professional materials in the vehicle technology field, including internal technical specifications, industry technical standards, experience summary documents, and historical analysis cases. Preprocess the raw documents using semantic segmentation to ensure the semantic integrity of each knowledge segment. Configure metadata tags for each knowledge segment, covering information such as knowledge category, applicable vehicle models, operating conditions, and validity period. Convert the knowledge segments into vector data using a word embedding model and store them in a vector database. Simultaneously, establish a keyword index to enable metadata-based filtering and retrieval.

[0023] Specifically, the system responds to user input on the platform's interface to retrieve vehicle-related business questions. These inputs include text and voice input; the business questions are stored in natural language. For example, business questions could include analyzing the variation in vehicle fuel consumption at different speeds, compiling vehicle emissions data under mountainous road conditions, and identifying frequently failing parts in high-mileage vehicles.

[0024] S120. Analyze the business problem to determine the business intent.

[0025] Among them, business intent is a standardized analytical request extracted from the user's business question in natural language form. For example, if the business question is to collect data on vehicle fuel consumption under national highway driving conditions, the corresponding business intent is to conduct statistical analysis on vehicle energy consumption under national highway driving conditions.

[0026] In this embodiment, an artificial intelligence model is used to analyze the business problem and determine the business intent. For example, the artificial intelligence model can be a deep learning model, a large language model, etc.

[0027] Optionally, the business problem is parsed to determine the business intent, including: The business problem is analyzed using a pre-set large language model to determine the business intent.

[0028] Among them, large language models are giant neural networks based on deep learning, trained on massive amounts of text data, and their core capability is to understand and generate human natural language. These include the GPT series, Qwen, Llama, Claude, and others.

[0029] Specifically, the large language model employs a Transformer encoder-decoder structure, relying on a self-attention mechanism to capture the relationships between contextual words. The encoder consists of multiple stacked layers, each containing a multi-head self-attention module and a feedforward neural network. The encoder uses the self-attention mechanism to globally model the entire business question text sequence, calculating the association weight between each word and all other words in the sequence, simultaneously capturing the contextual dependencies within the text, and fully extracting the contextual semantic features of the entire sentence, forming a contextual feature vector containing global semantic information. The decoder also adopts a multi-layer stacked structure, with each layer containing masked multi-head self-attention, a cross-attention mechanism, and a feedforward neural network. Masked self-attention ensures that the model can only deduce based on the already generated text content and will not see the subsequent output content in advance; the cross-attention mechanism introduces the global semantic features output by the encoder into the decoder, enabling the decoder to associate the generated content with the complete semantics of the original business question, ensuring a strict correspondence between the output content and the input question.

[0030] Furthermore, the business problem text to be parsed is first fed into the encoder to complete global semantic encoding; the decoder combines the generated content with the semantic features from the encoder to perform word-by-word probability prediction, continuously rewriting the content. Relying on the collaborative work of the encoder and decoder, the model can deeply understand the contextual logic of natural language sentences, realizing the conversion of unstructured business problem text into standardized content.

[0031] In this solution, a large pre-trained language model is fine-tuned using a massive amount of historical business problem samples to achieve domain-specific adaptation training, resulting in a large language model tailored to specific business scenarios. Business problems are then input into the large language model, which predicts the business questions and outputs structured business intents.

[0032] By leveraging the Transformer encoding / decoding structure and multi-head self-attention, we can fully mine the contextual semantics, improve the accuracy of understanding business issues, and stably output standardized and consistent structured business intents.

[0033] S130. Determine the target functional module that matches the business problem; wherein the target functional module consists of at least one functional algorithm; the functional algorithm is used to construct the data processing logic for business problem analysis.

[0034] In this embodiment, functional algorithms are used to construct the data processing logic for business problem analysis. For example, functional algorithms may be vehicle speed time-series clustering algorithms, fuel consumption fitting algorithms, driving behavior recognition algorithms, fault code association analysis algorithms, location trajectory cleaning algorithms, etc.

[0035] Among them, the following algorithms are implemented: Vehicle speed time-series clustering algorithm: This algorithm constructs the data processing logic for classifying vehicle driving conditions, performing clustering processing on continuously collected vehicle speed data. Fuel consumption fitting algorithm: This algorithm constructs the data processing logic for calculating fuel consumption per 100 kilometers, combining mileage and fuel level data to complete index fitting calculations. Driving behavior recognition algorithm: This algorithm constructs the data processing logic for assessing the risks of rapid acceleration and braking, extracting behavioral features from vehicle dynamic data. Fault code association analysis algorithm: This algorithm constructs the data processing logic for vehicle fault diagnosis, enabling linked analysis of fault codes and vehicle operating parameters. Location trajectory cleaning algorithm: This algorithm constructs the data processing logic for vehicle path analysis, removing dirty data caused by satellite positioning drift.

[0036] In this solution, the target functional module is the vehicle data analysis functional module, which is stored in the form of a functional semantic description document. This functional semantic description document is a standardized functional description file oriented towards large language models. It clearly defines the functional name, business definition, applicable scenarios, applicable scenario descriptions for each algorithm dimension, and functional characteristic tags in natural language. It is the core knowledge carrier that ensures the matching of semantic retrieval and functional recommendation. For example, the target functional module could be energy consumption analysis, fault diagnosis, vehicle profiling, user profiling, etc.

[0037] In this embodiment, a search is conducted based on the pre-built relationship between business problems and functional modules to determine the target functional module that matches the business problem.

[0038] In this solution, the target functional module is selected from each functional module by calculating the semantic similarity between the business problem and the functional module.

[0039] Furthermore, keyword features of the business problem are extracted, and target functional modules are selected from each functional module based on these features. Keyword features are core entity words and business terms obtained by segmenting and filtering stop words from the business problem text. These features characterize the business category, data object, and analysis item corresponding to the business problem, serving as structured features for text matching. For example, keyword features could be "vehicle," "three months," "100 kilometers," or "fuel consumption."

[0040] S140. Based on the business intent, determine the target function algorithm from the function algorithms of the target function module.

[0041] In this embodiment, the intelligent agent recommends a target functional algorithm from the functional algorithms of the target functional module based on the business intent.

[0042] Specifically, users input their natural language requests through a web front-end. The intelligent agent first performs intent recognition and slot extraction, identifying the question type and key entities. If the intent is for feature recommendation or is explicit, the user's question is vectorized, and a hybrid search is performed in the feature knowledge base to retrieve the semantic description documents of relevant feature modules. After reordering, the best few recommendation results are selected and displayed to the user, including the feature name, applicable scenarios, and confidence level. After user confirmation, the algorithm recommendation stage begins.

[0043] In this solution, the intelligent agent adopts a layered microservice architecture deployed in a private cloud environment, which is divided into a user interaction layer, an intelligent decision-making layer, an analysis engine layer, a resource base layer, and an infrastructure layer from top to bottom.

[0044] The user interaction layer provides a unified interaction portal for end users, supporting text input, voice input, and file upload on web pages, and is responsible for preprocessing multimodal input, interactive feedback, and conversation history management. The intelligent decision-making layer is driven by a privately deployed large language model, enabling precise conversion of user needs and efficient task scheduling. Leveraging the semantic understanding, intent recognition, and logical reasoning capabilities of the large language model, it performs in-depth analysis and accurate judgment of user needs, supporting multi-turn dialogue context connection and automatic task flow orchestration. The analysis engine layer receives task instructions from the intelligent decision-making layer, linking with the knowledge base to perform semantic retrieval and recommendation matching. After the user manually triggers platform analysis and returns results, it receives chart data for structured analysis and conclusion generation. The resource foundation layer fully reuses existing resources of the big data platform, including the knowledge base, vector library, database, and cache, eliminating the need for redundant construction and reducing deployment costs. The infrastructure layer is provided by a private cloud, offering storage, computing, and network infrastructure services to ensure stable, secure, and scalable system operation. Core component selection is shown in Table 2.

[0045] Table 2 For example, after selecting energy consumption analysis as the target functional module, the agent automatically matches and pushes suitable target functional algorithms based on the parsed business intent. The energy consumption analysis module has multiple built-in functional algorithms, each corresponding to different business analysis objectives. The agent extracts keyword features from the business intent, retrieves the applicable scenario information corresponding to each functional algorithm in the functional knowledge base, and uses a semantic matching mechanism to filter out the target functional algorithm with the highest relevance.

[0046] The recommendation logic is built upon the semantic connection between business intent and the applicable scenarios of the algorithm: if a user needs to assess overall energy consumption levels or investigate abnormal fuel consumption issues, the speed-to-fuel-consumption trend analysis algorithm is prioritized; if a user studies the energy consumption impact of road conditions, two types of algorithms are matched: road type mileage distribution and terrain mileage distribution; if a user examines the effect of vehicle load on energy consumption, the total mass mileage distribution algorithm is pushed. The final recommendation results are output in list format, along with the recommendation criteria and matching confidence levels. Users can select a solution with a single click or manually modify the algorithm combinations.

[0047] Optionally, based on the business intent, the target functional algorithm is determined from the functional algorithms of the target functional module, including: Extract the second keyword feature from the business intent; Based on the second keyword feature, the target function algorithm is determined from the function algorithms of the target function module.

[0048] The second keyword feature is the core feature word that represents the business intent. For example, the second keyword feature could be load capacity, road conditions, fuel consumption, etc.

[0049] In this solution, text feature extraction techniques are used to extract the second keyword features from the business intent. For example, text feature extraction techniques can be TF-IDF keyword extraction algorithm, NER extraction method, etc.

[0050] Furthermore, the semantic similarity between the second keyword feature and the description of the applicable scenario of the functional algorithm is calculated, and the similarity values ​​are sorted from high to low. The target functional algorithm is then selected from all the functional algorithms included in the target functional module.

[0051] By extracting core keywords of business intent and performing semantic similarity matching with algorithm scenarios, we can accurately identify suitable functional algorithms and reduce misrecommendations from irrelevant algorithms.

[0052] S150. Generate a business report corresponding to the business problem based on the target function algorithm.

[0053] In this solution, vehicle operation data corresponding to the target function algorithm is retrieved from the data platform. By analyzing the vehicle operation data, a business report corresponding to the business problem is generated.

[0054] Specifically, after the user clicks "Start Analysis" on the front end, the big data platform performs calculations and returns chart data and statistical indicators. The intelligent agent receives the results data returned by the platform and performs subsequent processing.

[0055] Furthermore, after obtaining the chart data returned by the platform, the bar charts and tables are first converted into structured semantic descriptions, extracting key information such as data dimensions, statistical values, distribution characteristics, and outlier intervals. Then, the key chart features are used as query conditions to retrieve relevant evaluation criteria, root cause knowledge of anomalies, and conclusion templates from the domain knowledge base. Metadata filtering is used during the retrieval process to ensure the applicability of the knowledge. Finally, the structured chart descriptions and the retrieved domain knowledge are used as context to input into a large language model to generate structured conclusions that include an analysis overview, key findings, root cause inferences, and optimization suggestions. All key arguments are marked with knowledge source identifiers.

[0056] In this embodiment, users provide satisfaction ratings and textual feedback on the recommendation function, recommendation algorithm, and generated conclusions. The system periodically supplements the knowledge base with new functional semantic description documents or domain knowledge based on frequently asked user questions and feedback. New content is then processed through segmentation, vectorization, and storage to achieve continuous evolution of the knowledge base.

[0057] Optionally, based on the target function algorithm, a business report corresponding to the business problem is generated, including: Determine the screening time; Based on the filtering time, determine the vehicle operation data corresponding to the target function algorithm; The vehicle operation data is analyzed to generate a business report.

[0058] The screening time refers to the screening time for vehicle operation data.

[0059] In this solution, the filtering time is determined in response to user input. For example, the input can be text input, voice input, etc.

[0060] Specifically, after determining the screening time, vehicle operation data that falls within the screening period are selected from the vehicle operation dataset corresponding to the target function algorithm.

[0061] Furthermore, the vehicle operation data, such as bar charts and tables, returned by the data platform are processed and transformed into structured semantic descriptions, from which key information such as data dimensions, statistical values, distribution characteristics, and anomaly intervals are extracted. In the domain knowledge retrieval stage, the extracted key features of the charts are used as query conditions to retrieve corresponding evaluation criteria, root cause knowledge of anomalies, and conclusion templates from the domain knowledge base. When generating conclusions, the structured descriptions of the charts and the retrieved domain knowledge together form the context and are input into the large language model to generate a business report covering an analysis overview, key findings, root cause inferences, and optimization suggestions. Simultaneously, the corresponding knowledge source identifiers are labeled for each key conclusion. Table 3 shows an example of the business report.

[0062] Table 3 By flexibly setting data filtering time periods according to business needs, accurately extracting vehicle operation data corresponding to the target function algorithm, and conducting data analysis only within the valid dataset, the volume of data processing is greatly reduced, invalid calculations are reduced, and business reports are generated quickly. At the same time, the content of the report analysis is strictly aligned with the time range corresponding to the business problem, thereby improving the relevance and accuracy of the data analysis results.

[0063] This solution constructs a two-layer knowledge system comprising a functional knowledge base and a domain knowledge base, completing layered construction, semantic segmentation, metadata annotation, and hybrid retrieval and result rearrangement. Based on the knowledge base, it enables intelligent recommendation of functional modules, integrating vector semantic retrieval and keyword retrieval, and through fine-tuning, accurately mapping natural language business requirements to platform functions. It also allows for multi-round interactive confirmation when business intent is unclear. Furthermore, it performs semantic matching between business intent and algorithm application scenarios within the knowledge base, automatically selecting the best analysis algorithm and simultaneously displaying the reasons for recommendation and its credibility. Simultaneously, leveraging retrieval enhancement generation technology, it performs structured parsing of chart data, retrieving evaluation indicators, anomaly causes, and document templates from the knowledge base, outputting professional analysis content with source tags and traceability. The entire system adopts an intelligent agent task orchestration architecture, relying on atomic tasks for dynamic scheduling. After completing data queries on the platform, it automatically reads chart information to generate analysis conclusions, creating a complete business loop encompassing requirement identification, functional matching, algorithm optimization, and automatic conclusion output.

[0064] The technical solution of this invention involves: acquiring vehicle-related business issues; analyzing the business issues to determine the business intent; identifying target functional modules that match the business issues; determining a target functional algorithm from the functional algorithms of the target functional modules based on the business intent; and generating a business report corresponding to the business issues based on the target functional algorithm. By implementing this technical solution, it is possible to intelligently match business issues with functional modules, automatically optimize algorithm dimensions, and output business reports, significantly reducing the barrier to entry for using vehicle networking big data platforms and improving the efficiency of vehicle R&D data analysis and the utilization rate of platform functions.

[0065] Example 2 Figure 2 This is a flowchart of a business problem analysis process provided in Embodiment 2 of the present invention. The relationship between this embodiment and the above embodiments is a detailed description of the business problem analysis process. Figure 2 As shown, the method includes: S210, Obtain vehicle-related business issues.

[0066] S220. Analyze the business problem to determine the business intent.

[0067] S230. The business problem is vectorized to generate a business vector; based on the business vector, a first module is selected from the predetermined functional modules; wherein, the functional module is the vehicle task execution module.

[0068] In this solution, the business problem text is first preprocessed, including word segmentation, removal of stop words and invalid modifiers, and retention of valid text content such as business objects, analysis metrics, and constraints. A pre-trained word embedding model is then used to map each preprocessed word into a low-dimensional numerical vector. These vectors are then integrated into a unified sentence vector using mean pooling or concatenation aggregation. Finally, a business vector representing the semantic information of the entire sentence is obtained, completing the vectorization of the business problem.

[0069] Pre-trained word embedding models are trained on massive amounts of text corpora beforehand, enabling them to uncover contextual semantic relationships between words. The model maps words in natural language to a high-dimensional real-valued vector space, transforming semantically similar words into numerical vectors with similar distances, while words with significant semantic differences have more distant vectors, thus converting textual semantics into computer-computable numerical representations. Examples of word embedding models include Word2Vec, GloVe, and BERT. Word2Vec predicts words from context to obtain static word vectors; GloVe trains fixed word vectors based on global word co-occurrence frequencies; both are static embeddings, with each word corresponding to only one unique vector. BERT's word embedding layer relies on a bidirectional Transformer to capture contextual information and generate dynamic word vectors, capable of distinguishing the different semantics of the same word in different contexts.

[0070] Specifically, the semantic similarity between the business vector and each functional module is calculated, and the first module is selected from each functional module based on the magnitude of the semantic similarity. The semantic similarity includes cosine similarity, Euclidean distance similarity, and Manhattan distance similarity, with cosine similarity being the preferred method.

[0071] Optionally, based on the business vector, a first module is selected from a pre-determined set of functional modules, including: Calculate the semantic similarity between the business vector and each of the functional modules; Based on the semantic similarity, the first module is selected from the functional modules.

[0072] In this scheme, the semantic similarity between the business vector and each functional module is calculated, and all semantic similarity results are sorted from high to low. The first module with the best matching result is then selected.

[0073] By using vector similarity for semantic matching and selecting the best match, the accuracy of matching business problems with functional modules can be effectively improved.

[0074] S240. Extract the first keyword feature from the business problem; based on the first keyword feature, select the second module from the predetermined functional modules.

[0075] In this embodiment, the first keyword feature is a core feature word that characterizes the business problem. For example, the first keyword feature could be a quarter, load capacity, road conditions, fuel consumption, etc.

[0076] In this solution, text feature extraction techniques are used to extract the first keyword feature from the business problem. For example, text feature extraction techniques could be the TF-IDF keyword extraction algorithm, NER extraction method, etc.

[0077] Furthermore, the semantic similarity between the first keyword feature and the description of the applicable scenario of the functional algorithm is calculated. The second module is selected from the predetermined functional modules according to the similarity values ​​from high to low.

[0078] S250. The first module and the second module are identified as target functional modules.

[0079] Specifically, the first module and the second module are jointly identified as the target functional module. For example, if the first module is the energy consumption analysis module and the second module is the driving condition statistics module, then the energy consumption analysis module and the driving condition statistics module are jointly identified as the target functional module.

[0080] S260. Based on the business intent, determine the target function algorithm from the function algorithms of the target function module.

[0081] S270. Generate a business report corresponding to the business problem based on the target function algorithm.

[0082] The technical solution of this invention involves: acquiring vehicle-related business questions; analyzing the business questions to determine business intent; vectorizing the business questions to generate business vectors; selecting a first module from pre-determined functional modules based on the business vectors; extracting a first keyword feature from the business questions; selecting a second module from pre-determined functional modules based on the first keyword feature; determining the first and second modules as target functional modules; determining a target functional algorithm from the functional algorithms of the target functional modules based on the business intent; and generating a business report corresponding to the business questions based on the target functional algorithm. By executing this technical solution, two paths—vector semantic retrieval and keyword feature selection—are integrated to obtain the first and second modules respectively. Combining these two methods to determine the target functional module not only relies on vector matching to capture deep semantic needs but also utilizes keywords to lock in the core business theme, avoiding omissions and mismatches caused by a single matching method. Subsequently, the target functional algorithm is accurately selected based on the business intent, and a business report is automatically generated. The entire process achieves full automation of business question analysis, intelligent module matching, algorithm optimization, and report output, significantly reducing manual configuration costs and effectively improving the accuracy and efficiency of data analysis.

[0083] Example 3 Figure 3 This is a schematic diagram of a business problem analysis device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: Business issue acquisition unit 310 is used to acquire vehicle-related business issues; The business intent determination unit 320 is used to analyze the business problem and determine the business intent; The target functional module determination unit 330 is used to determine the target functional module that matches the business problem; wherein, the target functional module consists of at least one functional algorithm; the functional algorithm is used to construct the data processing logic for business problem analysis; The target function algorithm determination unit 340 is used to determine the target function algorithm from the function algorithms of the target function module based on the business intent; The business report generation unit 350 is used to generate a business report corresponding to the business problem according to the target function algorithm.

[0084] Optionally, the target functional module determination unit 330 includes: The first module determines the subunit, which is used to vectorize the business problem and generate a business vector; based on the business vector, the first module is selected from the pre-determined functional modules; wherein, the functional module is the vehicle task execution module; The second module determines the sub-unit, which is used to extract the first keyword feature in the business problem; based on the first keyword feature, the second module is selected from the pre-determined functional modules; The target functional module determination subunit is used to determine the first module and the second module as target functional modules.

[0085] Optionally, the first module determines the sub-units, specifically used for: Calculate the semantic similarity between the business vector and each of the functional modules; Based on the semantic similarity, the first module is selected from the functional modules.

[0086] Optionally, the business intent determination unit 320 is specifically used for: The business problem is analyzed using a pre-set large language model to determine the business intent.

[0087] Optionally, the target function algorithm determination unit 340 is specifically used for: Extract the second keyword feature from the business intent; Based on the second keyword feature, the target function algorithm is determined from the function algorithms of the target function module.

[0088] Optionally, the business report generation unit 350 is specifically used for: Determine the screening time; Based on the filtering time, determine the vehicle operation data corresponding to the target function algorithm; The vehicle operation data is analyzed to generate a business report.

[0089] The business problem analysis device provided in this embodiment of the invention can execute a business problem analysis method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0090] Example 4 Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0091] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0092] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0093] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a business problem analysis method.

[0094] In some embodiments, a business problem analysis method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the business problem analysis method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform a business problem analysis method by any other suitable means (e.g., by means of firmware).

[0095] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0096] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0097] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0098] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0099] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0100] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0101] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from ROM 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0102] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0103] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A business problem analysis method, characterized in that, include: Obtain business-related information about vehicles; The business problem is analyzed to determine the business intent; Identify target functional modules that match the business problem; wherein, the target functional modules consist of at least one functional algorithm; the functional algorithms are used to construct the data processing logic for business problem analysis; Based on the stated business intent, the target functional algorithm is determined from the functional algorithms of the target functional module; Based on the target function algorithm, a business report corresponding to the business problem is generated.

2. The method according to claim 1, characterized in that, Identify the target functional modules that match the business problem, including: The business problem is vectorized to generate a business vector; based on the business vector, a first module is selected from a pre-determined set of functional modules; wherein, the functional module is the vehicle task execution module; Extract the first keyword feature from the business problem; based on the first keyword feature, select the second module from the predetermined functional modules; The first module and the second module are identified as the target functional modules.

3. The method according to claim 2, characterized in that, Based on the business vector, a first module is selected from the predetermined functional modules, including: Calculate the semantic similarity between the business vector and each of the functional modules; Based on the semantic similarity, the first module is selected from the functional modules.

4. The method according to claim 1, characterized in that, The business problem is analyzed to determine the business intent, including: The business problem is analyzed using a pre-set large language model to determine the business intent.

5. The method according to claim 1, characterized in that, Based on the stated business intent, the target functional algorithm is determined from the functional algorithms of the target functional module, including: Extract the second keyword feature from the business intent; Based on the second keyword feature, the target function algorithm is determined from the function algorithms of the target function module.

6. The method according to claim 1, characterized in that, Based on the target function algorithm, a business report corresponding to the business problem is generated, including: Determine the screening time; Based on the filtering time, determine the vehicle operation data corresponding to the target function algorithm; The vehicle operation data is analyzed to generate a business report.

7. A business problem analysis device, characterized in that, include: The business issue acquisition unit is used to acquire vehicle-related business issues; The business intent determination unit is used to analyze the business problem and determine the business intent; A target functional module determination unit is used to determine target functional modules that match the business problem; wherein, the target functional module consists of at least one functional algorithm; the functional algorithm is used to construct the data processing logic for business problem analysis; The target function algorithm determination unit is used to determine the target function algorithm from the function algorithms of the target function module based on the business intent. The business report generation unit is used to generate a business report corresponding to the business problem based on the target function algorithm.

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

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

10. A computer program product comprising a computer program that, when executed by a processor, implements a business problem analysis method according to any one of claims 1-6.

Images