Intelligent generation and dynamic updating method for business administration teaching case library

Abstract

The invention provides an intelligent generation and dynamic updating method for a business administration teaching case library. The method comprises the following steps: monitoring a real-time business data flow of a target industry; according to the characteristics of the business data flow, extracting a management decision event and generating an initial case framework; performing adversarial optimization on the initial case framework based on a generative adversarial network, and generating a complete case including text narration and matched data; the case timeliness is evaluated through a dynamic updating algorithm, when the case popularity is attenuated to a threshold value, a regeneration process is triggered, and the case is updated to the case library, many problems existing in a traditional business administration teaching case library are effectively solved, remarkable advantages are shown in the aspects of timeliness, intellectualization, data fusion, an updating mechanism and the like, and the practicability is high. And the quality and the teaching value of the case library are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing and updating, and particularly to an intelligent generation and dynamic updating method for a business administration teaching case library. Background Art

[0002] In the field of business administration teaching, the case teaching method, with its vivid and realistic scenario simulation, can effectively improve students' decision-making ability and practical cognitive level, and occupies an extremely important position in the teaching process. As the core support of case teaching, the quality and timeliness of the case library have a direct and crucial impact on the teaching effect. However, there are still many problems to be solved urgently in the construction and updating of the current business administration teaching case library:

[0003] I. Limited case acquisition channels and lagging updates: Traditional case libraries mostly rely on teachers to manually collect and organize, and the sources are mainly limited to academic journals, enterprise public reports, etc. These channels not only have a limited number of cases obtained, but also take a long time from the occurrence of an event to the case being sorted and stored in the library, making it difficult to reflect the rapidly changing business environment. For example, in emerging industries such as artificial intelligence and blockchain, the industry is developing rapidly, new business models and management challenges are emerging continuously, but traditional case libraries cannot incorporate these latest business practice cases in a timely manner, resulting in a disconnect between teaching content and real business.

[0004] II. Lack of intelligent means for case generation: In the past, case generation mostly relied on teachers' personal experience and subjective judgment to screen and extract case elements from complex business information, which was a cumbersome and inefficient process. At the same time, there are differences in the narrative structure, data integrity, etc. of cases generated manually, making it difficult to ensure the consistency of case quality. For example, when generating financial-related cases, the selection and analysis of financial data may vary among different teachers due to different personal understandings, resulting in the lack of generality and accuracy of case data.

[0005] III. Disconnection between cases and business data streams: Modern business activities generate a large amount of real-time data, such as enterprise operation data, market public opinion data, etc., but the existing case libraries have not fully utilized these data resources. The information in cases is often static and lagging, unable to reflect the impact of the dynamic changes of business data on management decisions. Take e-commerce enterprises as an example, daily sales data, user evaluation data, etc. can reflect the business conditions and market feedback of enterprises in real time, but traditional case libraries are difficult to incorporate these real-time data into cases, making it impossible for students to learn how to make scientific management decisions based on dynamic data through cases.

[0006] IV. Lack of an effective case dynamic update mechanism: As time goes by, the business environment changes, and the information and decision-making background in some cases may no longer be applicable. However, there is no corresponding mechanism in the case library to update or eliminate cases in a timely manner. This results in the continued use of outdated cases in teaching, and the knowledge and skills learned by students cannot meet actual business needs. For example, after a major change in the market competition pattern, if some cases on corporate market competition strategies are not updated in a timely manner, students may make wrong analyses and decisions based on the outdated case information;

[0007] Therefore, an intelligent generation and dynamic update method for a business administration teaching case library is proposed. Summary of the Invention

[0008] In view of this, embodiments of the present invention hope to provide an intelligent generation and dynamic update method for a business administration teaching case library to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial option.

[0009] To solve the above technical problems, a technical solution adopted by this application is: to provide an intelligent generation and dynamic update method for a business administration teaching case library, including the following steps: monitoring the real-time business data stream of the target industry; extracting management decision events according to the characteristics of the business data stream and generating an initial case framework; based on a generative adversarial network, performing adversarial optimization on the initial case framework to generate a complete case including text narration and supporting data; evaluating the timeliness of the case through a dynamic update algorithm, and triggering a regeneration process when the case popularity decays to a threshold, and updating it to the case library.

[0010] As a further preference of this technical solution: The characteristics of the business data stream include the semantic characteristics of enterprise annual report texts, the time-series change characteristics of financial data, and the sentiment tendency characteristics of market public opinion.

[0011] As a further preference of this technical solution: The generative adversarial network includes: a text generator: adopting a Transformer architecture, used to generate management scenario description texts, and the management scenario description texts output by the text generator include a narrative structure of a preset number of decision nodes to simulate the actual management decision-making scenario for learners to think about the decision-making process and influencing factors; a data generator: generating a matching financial indicator dataset through an LSTM network, meeting a preset error constraint; generating a matching financial indicator dataset through an LSTM network, and the data generator meets a preset error constraint when generating the financial indicator dataset to reflect the financial status and operating results of the enterprise; a cross-modal discriminator: used to calculate the consistency score between the management scenario description text and the financial indicator dataset.

[0012] As a further preference of this technical solution: The real-time business data stream of the monitored target industry includes: determining multi-source data sources: The multi-source data sources include enterprise internal systems, industry databases, news media and information platforms, social media platforms, and government departments and regulatory agencies; data collection: Using web crawlers, API interfaces, and data collection software for data collection; using MQTT and AMQP real-time transmission protocols for data transmission and storage.

[0013] As a further preference of this technical solution: Extracting management decision events and generating an initial case framework according to the characteristics of the business data stream includes: cleaning and preprocessing the collected business data stream, and respectively extracting semantic features of enterprise annual report texts, temporal change features of financial data, and sentiment tendency features of market public opinions; extracting management decision events through rule matching and machine learning methods; constructing a framework structure including business administration case elements, filling the extracted relevant information into this framework structure, and finally evaluating and optimizing the initial case framework according to evaluation indicators.

[0014] As a further preference of this technical solution: The business administration case elements include case background, description of management decision events, decision-making objectives, decision-making basis, possible impacts, and result parts.

[0015] As a further preference of this technical solution: When extracting semantic features of enterprise annual report texts, using word embedding, deep learning models, and topic modeling techniques to convert the text into vector representations and identify main topics and key decision-making information; when analyzing temporal change features of financial data, using time series analysis methods to model and predict financial data, and identifying features such as trends and seasonality and abnormal fluctuations; when analyzing sentiment tendency features of market public opinions, using sentiment analysis techniques to classify the sentiment of public opinion data and count relevant indicators.

[0016] As a further preference of this technical solution: Calculating the consistency score between the management scenario description text and the financial indicator data set includes: respectively performing data preprocessing and feature extraction on the management scenario description text and the financial indicator data set, and mapping the features to the same feature space; calculating their consistency through semantic and data correlation analysis and pattern matching, and calculating a comprehensive score by assigning weights according to different analysis results, and finally comparing the score with a preset threshold to determine whether they match to form a reasonable and coherent case.

[0017] To solve the above technical problems, another technical solution adopted by this application is: A computer device, the computer device includes a processor and a memory coupled to the processor, and program instructions are stored in the memory. When the program instructions are executed by the processor, the processor executes the steps of an intelligent generation and dynamic update method of a business administration teaching case library as described above.

[0018] To solve the above technical problems, another technical solution adopted in this application is: a storage medium storing program instructions capable of implementing the method for intelligent generation and dynamic update of a business administration teaching case library as described above.

[0019] Due to the adoption of the above technical solutions in the embodiments of the present invention, it has the following advantages:

[0020] The present invention effectively solves many problems existing in the traditional business administration teaching case library, showing significant advantages in timeliness, intelligence, data fusion, and update mechanism, etc., and strongly improving the quality and teaching value of the case library.

[0021] Strong timeliness: The traditional case library relies on teachers to manually collect and organize, with limited acquisition channels and lagging updates, making it difficult to reflect the rapidly changing business environment. This solution can quickly capture the latest trends and changes in the industry by monitoring the real-time business data stream of the target industry. Such as the business models and management challenges of emerging industries, the changes in the market competition pattern, etc., can be timely incorporated into the case library to ensure that the teaching content is closely connected with the real business, and the knowledge and skills learned by students are more in line with actual needs.

[0022] High degree of intelligence: In the past, case generation mostly relied on teachers' personal experience and subjective judgment, with a cumbersome process and low efficiency, and it was difficult to ensure the consistency of case quality. This solution is based on a generative adversarial network, using a text generator, a data generator, and a cross-modal discriminator to achieve intelligent case generation. The text generator uses the Transformer architecture to generate text descriptions of management scenarios, the data generator generates a matching financial indicator dataset through the LSTM network, and the cross-modal discriminator ensures the consistency of the text and data, greatly improving the efficiency and quality of case generation, and ensuring the consistency of cases in terms of narrative structure, data integrity, etc.

[0023] High degree of data fusion: The traditional case library fails to make full use of the massive real-time data generated by modern business activities, such as enterprise operation data, market public opinion data, etc. This solution fully excavates the semantic features of enterprise annual report texts, the time-series change features of financial data, and the sentiment tendency features of market public opinion, and integrates these multi-source data into the case generation process, making the information in the cases more comprehensive and dynamic, capable of reflecting the impact of changes in business data on management decisions, and enabling students to learn to make scientific management decisions based on dynamic data.

[0024] Improvement of the dynamic update mechanism: The traditional case library lacks an effective dynamic case update mechanism. Some case information and decision-making backgrounds are no longer applicable over time, but they are still continuously used in teaching. This solution evaluates the timeliness of cases through a dynamic update algorithm. When the case popularity decays to a threshold, it triggers the regeneration process and updates the case to the case library. This ensures that the cases in the case library always maintain high timeliness and teaching value, enabling students to access the latest and most practical cases and improving the teaching effect.

[0025] The above summary is only for the purpose of the specification and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the present invention will be readily apparent by reference to the drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] To more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0027] Figure 1 is a flowchart of the method of the present invention;

[0028] Figure 2 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The following will describe the embodiments of the present disclosure in detail with reference to the drawings.

[0030] It should be clear that the following illustrates the embodiments of the present disclosure through specific specific examples. Those skilled in the art can easily understand other advantages and effects of the present disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present disclosure without creative efforts belong to the scope of protection of the present disclosure.

[0031] Note that the following description pertains to various aspects of embodiments within the scope of the appended claims. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is for illustrative purposes only. Based on this disclosure, those skilled in the art should understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of the aspects set forth herein can be used to implement a device and / or practice a method. Additionally, this device and / or method can be implemented using other structures and / or functionality in addition to one or more of the aspects described herein.

[0032] It should also be noted that the diagrams provided in the following embodiments merely illustrate the basic concept of the present disclosure schematically. Only the components related to the present disclosure are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0033] In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0034] Figure 1 It is a schematic flowchart of a method for intelligent generation and dynamic update of a business administration teaching case library according to an embodiment of the present invention. Note that if there are substantially the same results, the method of the present application is not limited to Figure 1 the process sequence shown. As Figure 1 shown: A method for intelligent generation and dynamic update of a business administration teaching case library includes the following steps:

[0035] Step S100: Monitor the real-time business data stream of the target industry, where the characteristics of the business data stream include the semantic characteristics of enterprise annual report texts, the time-series change characteristics of financial data, and the sentiment tendency characteristics of market public opinion;

[0036] Specifically, monitoring the real-time business data stream of the target industry includes determining multi-source data sources: The multi-source data sources include enterprise internal systems, industry databases, news media and information platforms, social media platforms, and government departments and regulatory agencies; data collection: Use web crawlers, API interfaces, and data collection software for data collection; use MQTT and AMQP real-time transmission protocols for data transmission and storage.

[0037] Based on the above solution, the source set of the collected data can be set as S = (S1, S2,..., S n), where S1 can be an enterprise internal system, S2 can be an industry database, and so on. The amount of data obtained from the i-th source using web crawlers, API interfaces, data collection software, etc. is D i .

[0038] S200. Extract management decision events based on the characteristics of the business data stream and generate an initial case framework;

[0039] Specifically, step S200 may include cleaning and preprocessing the collected business data stream, respectively extracting semantic features of enterprise annual report texts, time series change features of financial data, and sentiment tendency features of market public opinions; extracting management decision events through rule matching and machine learning methods; constructing a framework structure including business administration case elements, and filling the extracted relevant information into the framework structure. Finally, quality evaluation and optimization are performed on the initial case framework according to evaluation indicators, where the business administration case elements include case background, description of management decision events, decision-making objectives, decision-making basis, possible impacts, and result parts.

[0040] Based on the above solution, when cleaning and preprocessing the collected business data stream, let the cleaning function be C(D), remove noise, duplicate, and incorrect data, and obtain the preprocessed data

[0041] Feature extraction:

[0042] Extraction of semantic features of enterprise annual report texts: Using word embedding, deep learning models, and topic modeling techniques, let the text feature extraction function be Convert the text into vector representation and identify the main topics and key decision-making information to obtain the text feature vector

[0043] Model and predict financial data using time series analysis methods. Let the time series analysis function be Identify features such as trends and seasonality, as well as abnormal fluctuations, to obtain the financial feature vector

[0044] Extraction of sentiment tendency features of market public opinions: With the help of sentiment analysis technology, let the sentiment analysis function be Classify the sentiment of public opinion data and count relevant indicators to obtain the public opinion feature vector

[0045] Extraction of management decision events: Extract management decision events through rule matching and machine learning methods. Rule matching is based on a pre-established rule set R = (r1, r2,..., r n ), if the data satisfies the i-th rule r i, it is recognized as the corresponding decision event E i . In terms of machine learning methods, a dataset L labeled with management decision event types is constructed, and supervised learning algorithms such as support vector machines, decision trees, random forests, etc. are used for training. Let the training function be T rain (L), and the feature vectors V text 、V finance 、V sentiment are used as inputs to train the model to predict management decision events and their types.

[0046] Initial case framework generation: Design a framework structure F including parts such as case background, management decision event description, decision goal, decision basis, possible impacts, and results. Fill the extracted relevant information I = {E i , V,...} into it, where V is a feature, and the initial case framework F init = Fill(F, I). Finally, the quality of the initial case framework is evaluated and optimized according to the evaluation index.

[0047] Step S300: Based on the generative adversarial network, perform adversarial optimization on the initial case framework to generate a complete case including text narrative and supporting data;

[0048] Specifically, in step S300, the generative adversarial network includes: a text generator: adopting a Transformer architecture, used to generate management scenario description text. The management scenario description text output by the text generator includes the narrative structure of a preset number of decision nodes to simulate the actual management decision scenario for learners to think about the decision-making process and influencing factors; a data generator: generating a matching financial indicator dataset through an LSTM network, meeting the preset error constraint; generating a matching financial indicator dataset through an LSTM network. When generating the financial indicator dataset, the data generator meets the preset error constraint to reflect the financial status and operating results of the enterprise; a cross-modal discriminator: used to calculate the consistency score between the management scenario description text and the financial indicator dataset.

[0049] When extracting the semantic features of enterprise annual report texts, word embedding, deep learning models, and topic modeling techniques are used to convert the text into vector representations and identify the main topics and key decision-making information; when analyzing the time-series change characteristics of financial data, time-series analysis methods are used to model and predict financial data to identify features such as trends, seasonality, and abnormal fluctuations; when analyzing the sentiment tendency characteristics of market public opinion, sentiment analysis techniques are used to classify the sentiment of public opinion data and count relevant indicators.

[0050] More specifically, for the extraction of semantic features of enterprise annual report texts: During the word embedding process, it can be assumed that the dimension of the word vector space is d. For the word w i in the text, its word vector representation is The processing of text T by a deep learning model (such as BERT) can be expressed as (where l is the number of words in the text, and w i is the weight coefficient). Through topic modeling (such as the LDA model), assuming the number of topics is k and the topic distribution is θ, θ = LDA(T), the main topics and key decision-making information are identified accordingly.

[0051] Extraction of the characteristics of the time series change of financial data: In time series analysis, for example, the ARIMA model can be used. Assuming the time series is y t , the model is expressed as:

[0052]

[0053] where p is the autoregressive order, q is the moving average order, and are the model parameters, ∈ t is white noise, which is used to identify features such as trends and seasonality and abnormal fluctuations.

[0054] Extraction of the sentiment tendency characteristics of market public opinion: In sentiment analysis technology, a dictionary-based method is adopted. Assuming the sentiment dictionary is D sentiment , for the word w in the text T sentiment , if w ∈ D sentiment , its sentiment tendency score s(w) is marked according to the dictionary. The sentiment tendency score of the text is:

[0055]

[0056] where n is the number of words in the text, and sentiment classification and relevant indicators are counted accordingly.

[0057] Calculate the consistency score between the management scenario description text and the financial indicator dataset, including: performing data preprocessing and feature extraction on the management scenario description text and the financial indicator dataset respectively, and mapping the features to the same feature space; calculating their consistency through semantic and data association analysis and pattern matching, and calculating the comprehensive score by assigning weights according to different analysis results. Finally, compare the score with the preset threshold to determine whether they match to form a reasonable and coherent case. The specific steps include: performing data preprocessing and feature extraction on the management scenario description text T manage and the financial indicator dataset D finance respectively. Assuming the text feature extraction function is F text (T manage ), the text feature vector V text is obtained. The data feature extraction function is F data (D finance ), and the data feature vector V data is obtained.

[0058] Map the feature vectors V text and V data to the same feature space. Let the mapping function be Map(V text , V data ), and obtain the mapped feature vector V mapped .

[0059] Through semantic-data correlation analysis, let the correlation analysis function be Ana(V mapped ), and calculate the consistency score G1 of the two; through pattern matching, let the matching function be Match(V mapped ), and calculate the matching score G2. Assign weights ω3 and ω4 (ω3 + ω4 = 1) according to different analysis results, calculate the comprehensive score G = ω3G1 + ω4G2, and finally compare the score G with the preset threshold g to determine whether the two match to form a reasonable and coherent case. If G ≥ g, it matches; otherwise, it does not match. Repeat the above process until it matches.

[0060] S400. Evaluate the timeliness of the case through the dynamic update algorithm. When the case popularity decays to the threshold, trigger the regeneration process and update it to the case library.

[0061] More specifically, step S400 may include:

[0062] Define the case popularity index: Construct the case popularity function where c represents the case, s is the number of factors affecting popularity, p i is the weight coefficient of each factor, and f i is the influence function of the i-th factor on the popularity of case c. For example, the factor f1(c) can represent the number of times the case is used in teaching within a certain time, and f2(c) is the number of news reports related to the case, etc.

[0063] Real-time monitor the case popularity: According to the set time interval Δt, for each case c j (j = 1, 2,..., N), N is the total number of cases in the case library, calculate its popularity value H(c j ) t in real time, where t represents the current time.

[0064] Case popularity decay model: Assume that the case popularity decays exponentially with time, and the decay model is H(c j ) t+Δt = H(c j ) t × e -λΔt , where λ is the decay coefficient, reflecting the decay speed of the case popularity with time.

[0065] Threshold comparison and judgment: Set the case popularity threshold T heat, the real-time heat value H(c j ) t With threshold T heat For comparison. If H(c j ) t ≤T heat , it is determined that the case heat has decayed to the threshold, triggering the regeneration process.

[0066] Regeneration process: For case c where the heat decays to the threshold j , re-run the process of monitoring the real-time business data flow of the target industry, extracting management decision events and generating an initial case framework, and generating a complete case through adversarial optimization based on a generative adversarial network in steps S100-S300 to obtain an updated case

[0067] Case Library Update: Updated cases Replace the original case c in the case library j , complete the dynamic update of the case library, and ensure that the cases in the case library always maintain high timeliness and teaching value.

[0068] The present invention also provides a simulation embodiment for practical application of the method according to the present invention:

[0069] 1. Monitor real-time business data flows in target industries.

[0070] Data source and collection:

[0071] Internal enterprise system: The sales data of the past month was obtained from the sales management system of the well-known smartphone brand "a certain mobile phone". It was found that its weekly sales in first-tier cities were 5,000, 5,200, 5,300, and 5,100 units respectively; the current inventory quantity of each model of mobile phone was learned from the inventory management system, such as the inventory of a certain X5 model was 10,000 units.

[0072] Industry database: We learned from the database of a professional market research organization that the market growth rate of the entire smartphone industry this quarter is 3%, and the market share of a certain mobile phone is 10%, ranking fourth in the industry.

[0073] News media and information platforms: According to news websites, competitor Ruifeng Mobile Phone recently launched a new model with innovative camera function, and it is expected that sales will reach 80,000 units in the first month after its launch.

[0074] Social media platform: Using social media monitoring tools, we found that the negative comments about a certain mobile phone on Weibo mainly focused on battery life, with 3,000 discussions on battery life in the past week, while positive comments mostly revolved around appearance design, with 2,000 discussions.

[0075] Government departments and regulatory agencies: It is learned from the policy documents issued by the government that the import tariff on smartphones may be increased by 5% within the next six months.

[0076] 2. Extract management decision events and generate an initial case framework according to the characteristics of business data streams.

[0077] Feature extraction and analysis:

[0078] Semantic features of the enterprise annual report text: By analyzing the annual report of a certain mobile phone in the previous year, it is found that it emphasizes technological innovation and market expansion, but mentions less about the user experience. Combining with the negative evaluations on current social media, it is speculated that there may be deficiencies in the product optimization of the enterprise.

[0079] Temporal change characteristics of financial data: By analyzing the financial data of a certain mobile phone in the past three years, it is found that the proportion of R & D investment has decreased year by year, from 15% three years ago to 10% this year; while the proportion of marketing expenses has increased year by year, from 12% to 18%. At the same time, the net profit growth rates in the past two years were 10% and 5% respectively, but only 2% this year.

[0080] Emotional tendency characteristics of market public opinion: Based on the evaluations on social media, the overall emotional tendency of a certain mobile phone in the market public opinion is negative, and the proportion of negative evaluations reaches 60%, mainly concentrated in aspects such as battery life and system smoothness.

[0081] Extraction of management decision events:

[0082] According to the above feature analysis, the following management decision events are extracted:

[0083] In view of the fact that competitors have launched innovative products and the competition for market share is fierce, a certain mobile phone needs to decide whether to increase R & D investment to enhance product competitiveness.

[0084] Considering that the import tariff may increase, it is necessary to decide whether to adjust the product pricing strategy.

[0085] Regarding the negative evaluations on social media, it is necessary to decide how to improve the product user experience.

[0086] Generation of the initial case framework:

[0087] Case background: A certain mobile phone is a brand with a certain share in the smartphone market, but faces huge pressure from competitors, and at the same time, the market environment and policies are constantly changing.

[0088] Description of management decision events:

[0089] Event 1: Whether to increase the proportion of R & D investment to 15% in the next year.

[0090] Event 2: Whether to increase the product price by 3% after the tariff increase.

[0091] Event 3: Whether to establish a dedicated user experience improvement team to solve the problems of battery life and system smoothness within three months.

[0092] Decision-making objectives:

[0093] Event 1: Improve product competitiveness and strive to increase the market share to 12% within one year.

[0094] Event 2: Maintain the profit level on the premise of ensuring that the sales volume is not greatly affected.

[0095] Event 3: Reduce the proportion of negative reviews to 30% within three months.

[0096] Decision-making basis:

[0097] Event 1: The threat of innovative products from competitors and the lack of product competitiveness caused by the decline in the proportion of R & D investment.

[0098] Event 2: Policy information on the possible increase in import tariffs and data on the impact of past price adjustments on sales volume.

[0099] Event 3: Negative review data on social media and pain points problems feedback by users.

[0100] Possible impacts and results:

[0101] Event 1: Increasing R & D investment may lead to a decline in profits in the short term, but is expected to increase the market share in the long term; if not increased, it may be further squeezed by competitors.

[0102] Event 2: Increasing the price may lead to a decline in sales volume, but can relieve cost pressure; maintaining the original price may affect profits.

[0103] Event 3: Establishing an improvement team may increase costs, but can effectively improve the user experience and enhance the brand image; not taking action may lead to a continuous increase in negative reviews.

[0104] 3. Optimize the initial case framework through a generative adversarial network to generate a complete case.

[0105] Text Generator: Adopts the Transformer architecture to generate detailed management scenario description texts. For example, for the R & D investment decision event, the following text is generated: "Under the background of increasingly fierce market competition, a certain mobile phone is facing the dilemma of insufficient product innovation. The new model launched by the competitor, Ruifeng Mobile Phone, has attracted a large number of consumers with its innovative camera function and is expected to impact the market share of a certain mobile phone. In this case, the management of a certain mobile phone needs to decide whether to increase the proportion of R & D investment from the current 10% to 15% in the next year to enhance the technical content and competitiveness of the product."

[0106] Data Generator: Generates a matching financial indicator dataset through the LSTM network. Assume the generated data shows that if the proportion of R & D investment is increased to 15%, the R & D expenses in the next year will increase from the current 500 million yuan to 750 million yuan, and after the successful launch of the product, the sales revenue in the next year is expected to increase from 5 billion yuan to 6 billion yuan, and the net profit will increase from 500 million yuan to 600 million yuan.

[0107] Cross-modal Discriminator: Conducts a consistency assessment on the generated management scenario description text and the financial indicator dataset. After evaluation, the two match logically. For example, the increase in R & D investment mentioned in the text to enhance product competitiveness is reflected in the growth of sales revenue and net profit in the dataset. Therefore, a complete case containing text narration and supporting data is generated.

[0108] 4. Evaluate the timeliness of the case through a dynamic update algorithm and update the case library:

[0109] Case Popularity Evaluation:

[0110] The evaluation indicators for case popularity include the usage frequency of the case in teaching, the number of relevant news reports, the popularity of discussions on social media, etc. In the next two months, the case of a certain mobile phone was used 10 times in teaching, there were 8 relevant news reports, and the number of discussions on this case on social media reached 500.

[0111] As time goes by, the market situation has changed. The sales volume of the new model of the competitor, Ruifeng Mobile Phone, far exceeds expectations, reaching 100,000 units, while the market share of a certain mobile phone has dropped to 9%. At the same time, the government officially announced a 5% increase in the import tariff on smart phones.

[0112] Trigger the regeneration process:

[0113] Due to the significant changes in the market situation, the popularity of the case of a certain mobile phone has gradually declined, and the relevant data and decision-making situations no longer conform to the current reality. When the case popularity evaluation score is lower than the preset threshold (preset to 60 points, and the current score is 40 points), trigger the regeneration process.

[0114] Regenerate and update the case library:

[0115] Re - monitor the real - time business data stream of the target industry to obtain the latest data. For example, it is learned that a certain mobile phone plans to launch a new model with fast - charging function to cope with competition, with an estimated R & D cost of 300 million yuan and an expected sales volume of 80,000 units after listing.

[0116] According to the new data and situations, re - extract management decision - making events, such as whether to accelerate the R & D progress of the new model and how to adjust the pricing strategy of the new model, and generate a new initial case framework.

[0117] Optimize again through the generative adversarial network to generate an updated complete case and update it to the case library to ensure the timeliness and practicality of the case library.

[0118] The present invention also provides an actual embodiment that continues the case of "a certain mobile phone" in the previous smartphone industry, presenting the consistency score between the management scenario description text and the financial indicator dataset:

[0119] 1. Generate the management scenario description text and the financial indicator dataset

[0120] Management scenario description text:

[0121] In order to cope with the impact of the new model of the competitor Ruifeng Mobile Phone, a certain mobile phone decides to increase R & D investment and plans to increase the proportion of R & D investment from 10% to 15% in the next year. It is expected to launch more competitive products through technological innovation, thereby increasing the market share. If the R & D is successful and the market acceptance of the product is good, it is expected that the sales volume will increase by 20% within half a year after the new model is launched, and drive the sales revenue in the following year to increase by 30%.

[0122] Financial indicator dataset:

[0123] The generated financial indicator dataset shows that the current annual sales revenue of a certain mobile phone is 5 billion yuan, and the R & D cost is 500 million yuan (accounting for 10%). If the proportion of R & D investment is increased to 15%, the R & D cost will increase to 750 million yuan. It is expected that the sales volume will increase by 22% within half a year after the new model is launched, and the sales revenue in the following year will increase by 32% to reach 6.5 billion yuan.

[0124] 2. Calculate the consistency score:

[0125] Data pre - processing and feature extraction:

[0126] Text feature extraction: After tokenizing the management scenario description text and removing stop words, key information such as "increase R & D investment", "increase market share", "sales volume increase by 20%", "sales revenue increase by 30%", etc. are extracted, and these information are transformed into feature vectors. For example, "increase R & D investment" is encoded as a dimension, and its value is set to 1 to indicate the existence of this information; for "sales volume increase by 20%", the growth rate can be used as the value of this dimension.

[0127] Data feature extraction: Key data such as current sales revenue, R & D expenses, expected proportion of R & D investment, expected sales volume growth rate, and sales revenue growth rate are extracted from the financial indicator dataset, and are also transformed into feature vectors.

[0128] Feature mapping:

[0129] Map the text feature vector and the data feature vector to the same feature space. For example, both are mapped to a 5-dimensional vector space, corresponding to R & D investment, market share, sales volume growth, sales revenue growth, and other relevant factors respectively.

[0130] Consistency calculation:

[0131] Semantic-data correlation analysis:

[0132] In terms of R & D investment: The text mentions that the proportion of R & D investment is increased from 10% to 15%, and the R & D expenses in the dataset increase from 500 million yuan to 750 million yuan. After calculation, the proportion is exactly 15% (750÷5000 = 15%). The correlation degree of this part is relatively high, and it can be scored 9 points (out of 10).

[0133] In terms of sales volume growth: The text expects a 20% increase in sales volume, and the dataset expects a 22% increase. The two are relatively close, and the correlation degree is good, and it can be scored 8 points.

[0134] In terms of sales revenue growth: The text expects a 30% increase, and the dataset is 32%. They are also relatively matching, and the correlation degree can be scored 8 points.

[0135] Pattern matching:

[0136] Overall, the causal relationship described in the text (increase R & D investment - increase market share - increase in sales volume and sales revenue) is reasonably reflected in the dataset. The change trend of the data is consistent with the logic described in the text, and the pattern matching degree is relatively high, and it can be scored 8 points.

[0137] Comprehensive score calculation:

[0138] Assume that the weight of semantic-data correlation analysis is 0.7 and the weight of pattern matching is 0.3.

[0139] The average score of semantic-data correlation analysis = (9 + 8 + 8)÷3 = 8.33 points.

[0140] Comprehensive score = 8.33×0.7 + 8×0.3 = 8.23 points.

[0141] 3. Result judgment:

[0142] Set the threshold of the consistency score to 7 points. Since the calculated comprehensive score of 8.23 points is higher than the threshold, it indicates that the management scenario description text and the financial indicator dataset match each other and jointly constitute a reasonable and coherent business administration case, which can be used in the teaching case library. If the score is lower than the threshold, the text or data needs to be adjusted and the consistency assessment is carried out again until the score meets the requirements.

[0143] The present invention effectively solves many problems existing in the traditional business administration teaching case library, shows significant advantages in timeliness, intelligence, data fusion and update mechanism, etc., and effectively improves the quality and teaching value of the case library.

[0144] Strong timeliness: The traditional case library relies on teachers to manually collect and organize, with limited acquisition channels and lagging updates, making it difficult to reflect the rapidly changing business environment. This solution can quickly capture the latest trends and changes in the industry by monitoring the real-time business data stream of the target industry. Such as the business models and management challenges of emerging industries, the changes in the market competition pattern, etc., can be timely incorporated into the case library to ensure that the teaching content is closely connected with the real business, and the knowledge and skills learned by students are more in line with the actual needs.

[0145] High degree of intelligence: In the past, case generation mostly relied on teachers' personal experience and subjective judgment, with a cumbersome process and low efficiency, and it was difficult to guarantee the consistency of case quality. This solution is based on the generative adversarial network and uses a text generator, a data generator and a cross-modal discriminator to realize the intelligent generation of cases. The text generator uses the Transformer architecture to generate management scenario description text, the data generator generates a matching financial indicator dataset through the LSTM network, and the cross-modal discriminator ensures the consistency of the text and data, greatly improving the efficiency and quality of case generation, and ensuring the consistency of cases in terms of narrative structure, data integrity, etc.

[0146] High degree of data fusion: The traditional case library fails to make full use of the massive real-time data generated by modern business activities, such as enterprise operation data, market public opinion data, etc. This solution fully excavates the semantic features of enterprise annual report texts, the time-series change features of financial data and the sentiment tendency features of market public opinion, and integrates these multi-source data into the case generation process, making the information in the case more comprehensive and dynamic, and can reflect the impact of business data changes on management decisions, enabling students to learn to make scientific management decisions based on dynamic data.

[0147] Improvement of the dynamic update mechanism: The traditional case library lacks an effective case dynamic update mechanism. Some case information and decision-making backgrounds are no longer applicable over time, but they are still continuously used in teaching. This solution evaluates the timeliness of cases through a dynamic update algorithm. When the case popularity decays to a threshold, it triggers the regeneration process and updates it to the case library. This ensures that the cases in the case library always maintain a high timeliness and teaching value, enabling students to access the latest and most practical cases and improving the teaching effect.

[0148] It should be noted that the various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same and similar parts among the various embodiments, reference can be made to each other. For system-type embodiments, since they are basically similar to method embodiments, they are described relatively simply. For the relevant parts, reference can be made to the partial description of the method embodiments.

[0149] The electronic device according to an embodiment of the present disclosure includes a memory and a processor. The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0150] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In an embodiment of the present disclosure, the processor is used to run the computer-readable instructions stored in the memory, so that the electronic device executes all or part of the steps of the method for intelligent generation and dynamic update of a business administration teaching case library according to the foregoing embodiments of the present disclosure.

[0151] Those skilled in the art should be able to understand that, in order to solve the technical problem of how to obtain good user experience effects, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included in the protection scope of the present disclosure.

[0152] As Figure 2 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. It shows a schematic structural diagram of an electronic device suitable for implementing the electronic device in the embodiments of the present disclosure. Figure 2 The shown electronic device is only an example and should not bring any limitations to the functions and usage scopes of the embodiments of the present disclosure.

[0153] AsFigure 2 As shown, the electronic device may include a processor (such as a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). In the RAM, various programs and data required for the operation of the electronic device are also stored. The processor, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0154] Generally, the following devices may be connected to the I / O interface: an input device including, for example, a sensor or a visual information acquisition device; an output device including, for example, a display screen; a storage device including, for example, a magnetic tape, a hard disk, etc.; and a communication device. The communication device may allow the electronic device to communicate wirelessly or wireline with other devices (such as edge computing devices) to exchange data. Although Figure 2 an electronic device with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. Instead, more or fewer devices may be implemented or had.

[0155] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program codes for performing the method shown in the flowchart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by the processor, all or part of the steps of an intelligent generation and dynamic update method of a business administration teaching case library according to an embodiment of the present disclosure are executed.

[0156] For a detailed description of this embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, and details will not be repeated here.

[0157] A computer-readable storage medium according to an embodiment of the present disclosure stores non-temporary computer-readable instructions thereon. When the non-temporary computer-readable instructions are run by a processor, all or part of the steps of an intelligent generation and dynamic update method of a business administration teaching case library according to the foregoing embodiments of the present disclosure are executed.

[0158] The above-mentioned computer-readable storage medium includes but is not limited to: optical storage media (such as CD-ROMs and DVDs), magneto-optical storage media (such as MOs), magnetic storage media (such as magnetic tapes or removable hard disks), media with built-in rewritable non-volatile memories (such as memory cards), and media with built-in ROMs (such as ROM cartridges).

[0159] For a detailed description of this embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0160] The basic principles of the present disclosure have been described above in connection with specific embodiments. However, it should be noted that the advantages, benefits, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, benefits, effects, etc. are essential for each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purposes of illustration and facilitating understanding, rather than limitations, and the above details do not limit the present disclosure to necessarily adopt the above specific details for implementation.

[0161] In the present disclosure, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The block diagrams of the devices, apparatuses, equipment, and systems involved in the present disclosure are only illustrative examples and do not intend to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open-ended terms meaning "including but not limited to", and can be used interchangeably with each other. The word "or" and "and" used herein refer to the word "and / or", and can be used interchangeably with each other, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to", and can be used interchangeably with each other.

[0162] In addition, as used herein, the "or" used in the listing of items starting with "at least one" indicates a disjunctive listing, so that for example, the listing of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the term "exemplary" does not mean that the described examples are preferred or better than other examples.

[0163] It should also be noted that in the systems and methods of the present disclosure, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present disclosure.

[0164] Various changes, substitutions, and alterations to the techniques described herein may be made without departing from the teachings defined by the appended claims. In addition, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and acts described above. Current or later-developed processes, machines, manufactures, compositions of events, means, methods, or acts that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein may be utilized. Accordingly, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or acts within their scope.

[0165] The foregoing description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0166] The foregoing description has been presented for purposes of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although several example aspects and embodiments have been discussed above, those skilled in the art will recognize some of their variations, modifications, alterations, additions, and subcombinations.

Claims

1. An intelligent generation and dynamic update method for a business administration teaching case library, characterized in that, Including the following steps: Monitoring the real-time business data stream of the target industry; Extracting management decision events and generating an initial case framework according to the characteristics of the business data stream; Performing adversarial optimization on the initial case framework based on a generative adversarial network to generate a complete case including text narration and supporting data; Evaluating the timeliness of the case through a dynamic update algorithm, triggering a regeneration process when the case popularity decays to a threshold, and updating it to the case library.

2. The intelligent generation and dynamic update method of a business administration teaching case library according to claim 1, characterized in that: The characteristics of the business data stream include the semantic characteristics of enterprise annual report texts, the time-series change characteristics of financial data, and the sentiment tendency characteristics of market public opinions.

3. The intelligent generation and dynamic update method of a business administration teaching case library according to claim 1, characterized in that: The generative adversarial network includes: Text generator: Adopting a Transformer architecture, used to generate management scenario description texts. The management scenario description texts output by the text generator include the narrative structure of a preset number of decision nodes to simulate the actual management decision scenario for learners to consider the decision-making process and influencing factors; Data generator: Generating a matching financial indicator data set through an LSTM network, meeting the preset error constraint; generating a matching financial indicator data set through an LSTM network. When generating the financial indicator data set, the data generator meets the preset error constraint to reflect the financial status and operating results of the enterprise; Cross-modal discriminator: Used to calculate the consistency score between the management scenario description text and the financial indicator data set.

4. The intelligent generation and dynamic update method of a business administration teaching case library according to claim 1, characterized in that: The monitoring of the real-time business data stream of the target industry includes: Determining multi-source data sources: The multi-source data sources include enterprise internal systems, industry databases, news media and information platforms, social media platforms, and government departments and regulatory agencies; Data collection: Using web crawlers, API interfaces, and data collection software for data collection; Adopting MQTT and AMQP real-time transmission protocols for data transmission and storage.

5. The intelligent generation and dynamic update method of a business administration teaching case library according to claim 1, characterized in that: The extracting of management decision events and generating of the initial case framework according to the characteristics of the business data stream includes: Cleaning and preprocessing the collected business data stream, and respectively extracting the semantic characteristics of enterprise annual report texts, the time-series change characteristics of financial data, and the sentiment tendency characteristics of market public opinions; Extracting management decision events through rule matching and machine learning methods; Constructing a framework structure including business administration case elements, filling the extracted relevant information into the framework structure, and finally evaluating and optimizing the initial case framework according to evaluation indicators.

6. The intelligent generation and dynamic update method of a business administration teaching case library according to claim 5, characterized in that: The business administration case elements include case background, management decision event description, decision-making goal, decision-making basis, possible impacts, and result parts.

7. An intelligent generation and dynamic update method for a business administration teaching case library according to claim 1, characterized in that: When extracting the semantic characteristics of enterprise annual report texts, using word embedding, deep learning models, and topic modeling techniques to convert the text into vector representations and identify the main topics and key decision-making information; When analyzing the time-series change characteristics of financial data, using time-series analysis methods to model and predict financial data, and identifying characteristics such as trends, seasonality, and abnormal fluctuations; When analyzing the sentiment tendency characteristics of market public opinions, using sentiment analysis techniques to classify the sentiment of public opinion data and count relevant indicators.

8. An intelligent generation and dynamic update method for a business administration teaching case library according to claim 1, characterized in that: The calculating of the consistency score between the management scenario description text and the financial indicator data set includes: Perform data preprocessing and feature extraction on the management scenario description text and the financial indicator dataset respectively, and map the features to the same feature space; calculate the consistency between the two through semantic and data association analysis and pattern matching, and assign weights according to different analysis results to calculate the comprehensive score. Finally, compare the score with the preset threshold to determine whether the two match to form a reasonable and coherent case.

9. 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 instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for intelligent generation and dynamic update of a business administration teaching case library according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to execute the method for intelligent generation and dynamic update of a business administration teaching case library according to any one of claims 1-8.

Images