Enterprise internal decision-making method, system and platform architecture based on multi-source data information aggregation

Abstract

The invention provides an enterprise internal decision-making method, system and platform architecture based on multi-source data information aggregation, relates to the technical field of enterprise data management, and aims to obtain multi-source data information including enterprise internal data information and enterprise external data information, avoid the problem of single and isolated data sources and improve the enterprise internal decision-making efficiency. Deep learning, block chains, natural language processing technologies and big data analysis are integrated, multi-source data information is aggregated and analyzed, prediction topic creation, topic recommendation, prediction topic and reward distribution are realized, enterprise internal employee participation and external multi-source data are integrated, enterprise internal decision making is assisted, large-scale and high-frequency decision making requirements are supported, and the enterprise internal decision making efficiency is improved. And moreover, the wide market participation degree is improved, the accuracy and efficiency of enterprise internal decision making are improved, the participation enthusiasm of employees is aroused, and the comprehensiveness of enterprise internal decision making is improved. In the system and platform architecture provided by the invention, flexible replacement and expansion of components can be realized in combination with enterprise internal and external strategies and modular system design, in the subsequent system development and maintenance process, new functions or modules are easy to add, and in cooperation with a layered platform architecture, enterprise development or market environment change is flexibly adapted; and the fast-paced business requirement is met.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise data management, and more specifically, to an enterprise internal decision-making method, system and platform architecture based on multi-source data information aggregation. Background Art

[0002] Internal decision-making is crucial for a company. It not only impacts its daily operations but also directly influences its strategic positioning, market competitiveness, and long-term development. Internal decisions determine key factors such as product types, production methods, and output, ensuring the efficient allocation and utilization of corporate resources.

[0003] Currently, internal corporate decision-making primarily relies on traditional methods such as managers' experience, intuition, and limited data collection, processing, and analysis. However, in the face of an increasingly complex and volatile market environment, the limitations of this traditional approach are becoming increasingly apparent. First, traditional methods rely solely on a single and isolated data source, relying solely on data from internal employees involved in strategic planning. They lack effective mechanisms for integrating multiple external data sources (such as industry dynamics, market trends, and social media information). This leads to limited information coverage and data dimensionality, making it difficult to fully reflect the complex factors influencing internal corporate decision-making. Second, traditional methods, which rely on managers' experience and intuition, are highly subjective and lack information aggregation, failing to fully utilize existing data, limiting the effectiveness and efficiency of decision-making. Furthermore, traditional methods lag behind in the technical means of collecting, processing, and analyzing limited data, and their application of emerging technologies (such as blockchain and artificial intelligence) is limited and immature. These methods struggle to process large-scale, multi-dimensional, or real-time data. Consequently, managers lack comprehensive and in-depth data support for complex issues, making it difficult to support large-scale, high-frequency decision-making. This leads to insufficient basis for decision-making, hindering the scientific accuracy and timeliness of their decisions. Finally, the decision support system that traditional methods rely on has a fixed structure. The structure of the system and the relationship between its components are usually determined during the design process. During corporate strategic transformation, business expansion, or market fluctuations, it cannot flexibly adapt to changes in corporate development or market environment, and it is difficult to meet fast-paced business needs. It is also difficult to integrate and share internal and external information, which limits the comprehensiveness and accuracy of corporate decision-making.

[0004] In addition, when traditional methods are used to mobilize employees to participate in corporate decision-making, they often use stereotyped and one-sided methods such as questionnaires, which are difficult to mobilize the enthusiasm of corporate employees to participate. In order to mobilize the enthusiasm of employees to participate, the existing technology discloses a gamification management platform based on the enterprise. The platform includes: user management module, task management module, point management module, ranking and reward module, activity management module, data analysis and monitoring module, employee account registration, login, role assignment and authority management, task establishment, assignment and submission, and also through the reward and punishment system unit, to manage each employee's points, establish a ranking form according to the employee's points, and establish a reward and punishment mechanism to reward or punish employees. Through gamification, the enthusiasm of employees to participate in operational decision-making has been mobilized to a certain extent, which has promoted the improvement of the company's management level. However, on the one hand, this method mainly relies on the data of employees in modules such as tasks, points, and activity, but this information It may not be able to fully reflect the complex situation of corporate operations and market changes, resulting in incomplete information when making decisions. Although the platform provides functions such as task management and point rewards and punishments, it lacks in-depth data analysis and decision-making support tools, making it difficult to assist managers in making scientific and accurate decisions. On the other hand, it fails to fully consider the different needs and behavioral motivations of employees. Incentive methods based solely on material rewards are difficult to stimulate employee enthusiasm and participation in the long term, and may cause employees to engage in short-term speculative behavior to obtain rewards, leading to abuse or cheating, and undermining market fairness and stability. In addition, employees have a poor experience with the decision-making support system based on traditional methods, which is inconvenient to interact with and has a high participation threshold, limiting the possibility of broad participation and affecting the comprehensiveness and accuracy of internal corporate decision-making.

[0005] To sum up, the traditional methods of internal decision-making in enterprises have significant defects in data analysis, multi-information aggregation, decision support systems and employee participation enthusiasm, resulting in insufficient accuracy, scientific efficiency and comprehensiveness of internal decision-making. Scientific, effective and comprehensive internal decision-making in enterprises is inseparable from multiple aspects such as multi-source information acquisition, aggregation processing, analysis and the construction of decision support systems. Summary of the Invention

[0006] In order to solve the problems of low accuracy, low efficiency, unscientific and incomplete internal decision-making methods in traditional enterprises, the present invention proposes an internal decision-making method, system and platform for enterprises based on multi-source data information aggregation, integrating deep learning, blockchain, natural language processing technology and big data analysis, aggregating multi-source information, improving the accuracy and efficiency of internal decision-making in enterprises, mobilizing the enthusiasm of employees to participate, and improving the comprehensiveness of internal decision-making in enterprises.

[0007] In order to achieve the above technical effects, the technical solutions of the present invention are as follows:

[0008] In the first aspect, the present application proposes an internal enterprise decision-making method based on multi-source data information aggregation, comprising the following steps:

[0009] Acquire multi-source data information and pre-process the multi-source data information; the multi-source data information includes internal enterprise data information and external enterprise data information;

[0010] Creating topics, which are forecasted events related to the company's internal and external strategies, and are implemented using decentralized smart contracts based on blockchain technology;

[0011] Analyze multi-source data information based on deep learning models and natural language processing technology, and generate matching recommended topics for corporate employees based on the analysis results;

[0012] Presenting recommended topics to company employees for them to participate in topic prediction. The prediction process includes employees placing bets or trades on topics to form bet or trade prices. The prediction process is implemented based on decentralized smart contracts based on blockchain technology.

[0013] Determine whether the prediction results of the topics predicted by the company's employees meet the results set internally by the company. If so, the decentralized smart contract based on blockchain technology settles and distributes rewards to the employees participating in the prediction, and assists the company's internal decision-making based on the prediction results of the topics predicted by the company's employees. Otherwise, conduct internal research and judgment.

[0014] Preferably, the pre-processing of multi-source data information includes: multi-source data cleaning, multi-source data integration and multi-source data transformation; the decentralized smart contracts based on blockchain technology include: event creation contracts, betting contracts, transaction contracts, settlement contracts and governance contracts, and the decentralized smart contracts based on blockchain technology are all constructed using smart contract development languages;

[0015] The internal data information of the enterprise includes basic data information of enterprise employees, behavioral data information of enterprise employees and enterprise business decision-making information. The basic data information of enterprise employees includes: education level, age, work experience, interests, expertise and work ability; the external data information of the enterprise includes historical market transaction data information, historical external environment impact data information and current external environment impact data information.

[0016] Preferably, the topic is created by a creator with specific permissions. When the topic is created, a decentralized event creation contract based on blockchain technology is implemented. The event creation contract defines a data structure, which includes an event structure and a verification function. The verification function is used to authenticate the creator's identity, verify whether the creator's permissions are legal, and perform format and logic checks on the created topic.

[0017] The topic creation process is as follows:

[0018] S21: The event creation contract is triggered by the creation request submitted by the creator;

[0019] S22: The event creation contract calls the verification function to verify whether the creator's permissions are legal. If so, S23 is executed; otherwise, the event creation contract terminates the topic creation process and returns an error prompt;

[0020] S23: The event creation contract performs a format and logic check on the created topic. If the check fails, the event creation contract terminates the topic creation process and returns an error message, requiring the creator to correct it and submit it. If the check passes, S24 is executed.

[0021] S24: The event creation contract packages the created topics into topics for enterprise employees to participate in prediction and broadcasts them to the blockchain network. The nodes in the blockchain perform consensus verification. When the consensus verification passes, the topic is recorded in the blockchain ledger, and the topic creation is successful. All created topics constitute the topic library.

[0022] Preferably, when analyzing multi-source data information based on a deep learning model and natural language processing technology, the deep learning model is first trained and optimized using historical market transaction data information, historical external environmental impact data information, and historical enterprise internal decision-making data, and the model parameters of the deep learning model are dynamically adjusted;

[0023] Deep learning models and natural language processing technology are used to analyze the basic data information of corporate employees, corporate employee behavior data information, current market transaction data information, and current external environment impact data information, match topics for corporate employees from the topic library, and then generate final recommended topics. The process of generating the final recommended topics is: first filter out irrelevant topics, then perform similarity calculations and refine topics; after the topics are generated, recommend topics through the topic recommendation algorithm, then sort the topics, set topic weights, and obtain the final recommendations; corporate employees provide feedback on the final recommended topics.

[0024] Preferably, the process of enterprise employees participating in topic prediction is implemented based on a decentralized betting contract or transaction contract of blockchain technology. The betting contract includes an enterprise employee betting structure, a price calculation function, a balance check function, and a betting record function. The enterprise employee betting structure includes an enterprise employee identity record, the enterprise employee selected topic, topic prediction betting result options, topic prediction betting quantity, and the time when the enterprise employee participated in topic prediction betting; the transaction contract includes: an enterprise employee transaction structure, a price calculation function, a balance check function, and a transaction record function. The enterprise employee transaction structure includes an enterprise employee identity record, the enterprise employee selected topic, topic prediction transaction result options, topic prediction transaction quantity, and the time when the enterprise employee participated in topic prediction transaction, indicating that the enterprise employees buy and sell contracts through a bidding model. The bidding model refers to a trading method in which the contract price is determined by the quotes of the buyer and seller during the contract transaction process;

[0025] When enterprise employees choose to participate in topic prediction by betting, the process of enterprise employees participating in topic prediction is as follows:

[0026] S31: The betting contract is triggered after the company employee selects the topic prediction and topic prediction betting result options and determines the topic prediction betting amount;

[0027] S32: The betting contract calls the balance check function to check whether the token or points balance in the enterprise employee account is sufficient to pay the bet amount. If so, execute step S33; otherwise, the balance is insufficient, the betting contract is terminated, the event prediction process ends, and the enterprise employee is prompted to recharge;

[0028] S33: The betting contract calls the price calculation function to calculate the latest betting price of the betting result option based on the current betting situation of the predicted event;

[0029] S34: After the latest betting price is determined, the contract deducts the corresponding tokens or points from the enterprise employee's account and calls the betting record function to record the enterprise employee's betting information in the blockchain ledger; at the same time, the betting contract updates the betting ratio and betting amount information of the predicted event in the market data;

[0030] When enterprise employees choose to participate in topic forecasting by means of transactions, the process of enterprise employees participating in topic forecasting is as follows:

[0031] S301: The transaction contract is triggered after the enterprise employee selects the topic prediction and topic prediction transaction result options and determines the topic prediction transaction quantity;

[0032] S302: The transaction contract calls the balance check function to check whether the token or points balance in the enterprise employee's account is sufficient to pay the transaction amount. If so, step S33 is executed; otherwise, the balance is insufficient, the transaction contract is terminated, the event prediction process ends, and the enterprise employee is prompted to recharge;

[0033] S303: The trading contract calls the price calculation function to calculate the final transaction price of the transaction result option based on the current predicted event transaction status. The transaction price is determined by matching the quotes of the buyer and seller.

[0034] S304: After the final transaction price is determined, the transaction contract deducts the corresponding tokens or points from the enterprise employee's account and calls the transaction record function to record the enterprise employee's transaction information in the blockchain ledger. At the same time, the transaction contract updates the transaction ratio and transaction amount information of the predicted event in the market data.

[0035] When settling accounts and distributing rewards to employees of companies that participated in the prediction, a decentralized settlement contract based on blockchain technology is implemented. The settlement contract includes: a result confirmation function, a valid bet or transaction share calculation function, a reward distribution function, and an account balance update function;

[0036] The process of settling accounts and distributing rewards to employees who participated in the forecast is as follows:

[0037] S41: The result of enterprise employees participating in event prediction triggers the settlement contract;

[0038] S42: The settlement contract calls the valid bet or transaction share calculation function, traverses the enterprise employee bet or transaction information records in the blockchain ledger, filters out valid bets or transactions according to preset rules, and calculates the valid bet or transaction share for each result option;

[0039] S43: Based on the calculated valid bet or transaction share and the total amount of tokens or points in the reward pool, the settlement contract calls the reward allocation function and calculates the reward amount that each participating enterprise employee should receive according to the set settlement algorithm;

[0040] S44: The settlement contract calls the account balance update function to add the calculated reward amount to the account balance of the enterprise employee who placed the correct bet or transaction, and records the settlement process and results in detail in the blockchain ledger, including the event results, user bets or transactions, and reward distribution details.

[0041] Preferably, the method further includes a governance process. When the balance of an enterprise employee's account meets the set conditions, the enterprise employee can participate in the formulation of internal market rules and the review of forecast topics. The governance process is based on a decentralized governance contract based on blockchain technology. The governance contract includes a governance structure, a voting mechanism function, a proposal management function, and an authority control function. The process of enterprise employees participating in governance is as follows:

[0042] When an enterprise employee submits a governance proposal, the governance contract calls the proposal management function, records the proposal content into the governance structure, and displays it for other enterprise employees to review and discuss;

[0043] The governance contract calls the voting mechanism function and starts the voting process according to the set voting method. Employees vote on the proposal according to their own wishes, and the voting records are stored in the blockchain ledger.

[0044] After the voting is completed, the governance decision is executed according to the rules determined by the proposal management function and the permission control function.

[0045] Preferably, the rewards also adopt a multi-level incentive mechanism, which includes material rewards, career development and spiritual recognition. The material rewards include "bonuses and coupons", the career development includes "training opportunities and job promotions", and the spiritual recognition includes "senior level recognition and exclusive titles".

[0046] The method also includes risk management and control, and the risk judgment preset standards of the risk management and control are: abnormal betting or transaction amount, abnormal betting or transaction frequency, abnormal betting or transaction behavior pattern, abnormal source and flow of funds; for risks with different preset standards, preset risk response strategies include: betting or transaction restriction strategy, investigation and warning strategy, and market intervention strategy.

[0047] According to the above technical solution, the multi-level incentive mechanism can take into account the principles of behavioral psychology and economics, integrate material rewards and spiritual incentives, stimulate employees' intrinsic enthusiasm for participation, isolate corporate employees to actively participate in internal corporate decision-making forecasts in the long term, and improve the comprehensiveness and accuracy of internal corporate decision-making.

[0048] Preferably, decentralized smart contracts based on blockchain technology use multiple dimensions to select blockchain platforms, including transaction processing speed, throughput indicators, and:

[0049] The prediction event processing capability indicator per unit time is used to measure the processing efficiency of the blockchain platform during operations such as the creation, betting, trading, and settlement of highly concurrent prediction events;

[0050] Dynamic node expansion adaptability indicators are used to measure the stability of system performance, data consistency, and transaction processing capabilities during the dynamic increase or decrease of nodes on the blockchain platform;

[0051] Smart contract vulnerability detection coverage indicator, used to evaluate the blockchain platform's ability to detect and prevent smart contract vulnerabilities;

[0052] Data desensitization and authorized access granularity indicators are used to measure the blockchain platform's ability to desensitize sensitive information during data storage and transmission, as well as its flexibility in implementing fine-grained authorized access control;

[0053] The smart contract development language and enterprise technology stack compatibility index evaluation indicator is used to measure the degree of integration between the smart contract development language supported by the blockchain platform and the enterprise's existing technology system;

[0054] The blockchain platform introduces a node trust weighted consensus mechanism, assigns different voting weights according to the credibility of internal nodes of the enterprise, and adopts an optimized index structure for storage to improve data query efficiency. At the same time, an event tracing data storage model is designed to store the complete life cycle data of each predicted event in a chain structure.

[0055] In a second aspect, the present application proposes an enterprise internal decision-making system based on multi-source data information aggregation, wherein the system is deployed in a distributed network environment and includes:

[0056] A multi-source data information acquisition and processing module is used to acquire multi-source data information and pre-process the multi-source data information; the multi-source data information includes internal enterprise data information and external enterprise data information;

[0057] The topic creation module is used to create topics, which are predicted events related to the internal and external strategies of the enterprise. The topic creation is realized by decentralized smart contracts based on blockchain technology;

[0058] The topic recommendation module analyzes multi-source data information based on deep learning models and natural language processing technology, and generates matching recommended topics for enterprise employees based on the analysis results;

[0059] The prediction participation module displays recommended topics to corporate employees so that they can participate in topic prediction. The prediction participation process is implemented based on the decentralized smart contract of blockchain technology.

[0060] The prediction reward distribution module is used to determine whether the prediction results of the topics predicted by enterprise employees meet the results set within the enterprise. If so, the decentralized smart contract based on blockchain technology settles and distributes rewards to the employees participating in the prediction, and assists the internal decision-making of the enterprise based on the prediction results of the topics predicted by the employees. Otherwise, internal research and judgment will be conducted.

[0061] According to the above technical solution, a modular design is adopted, which can realize flexible replacement and expansion of components. In the subsequent system development and maintenance process, it is easy to add new functions or modules, which has high flexibility and adaptability.

[0062] Thirdly, this application proposes an enterprise internal decision-making platform architecture based on multi-source data information aggregation. The platform architecture adopts a layered design, including: a data acquisition and preprocessing layer, an intelligent engine layer, a business logic layer for predicting betting transaction rules, an enterprise employee interface layer, and a data storage layer;

[0063] The data acquisition and preprocessing layer is used to deploy a multi-source data information acquisition and processing module, the intelligent engine layer is used to deploy a topic creation module and a topic recommendation module, the predicted betting and transaction rules business logic layer is used to deploy a prediction participation module and a prediction reward distribution module; the enterprise employee interface layer is used to provide a friendly interactive interface for enterprise employees; the data storage layer is used to save predicted betting or transaction records, enterprise employee information and other data.

[0064] The data output by the multi-source data information acquisition and processing module flows through the data acquisition and processing layer and enters the intelligent engine layer. The intelligent engine layer includes deep learning models, natural language processing components and intelligent event generators, and is responsible for topic generation and topic recommendation;

[0065] The business logic layer of the predicted betting transaction rules is equipped with a centralized smart contract platform, which is responsible for predicting transaction management, rule formulation and market monitoring; the transaction cycle on the decentralized smart contract platform includes: initiating contract creation, broadcasting, verification, storage, execution and settlement.

[0066] According to the above technical solution, the enterprise employee interface layer provides enterprise employees with a friendly interactive interface, improving the enterprise employee experience. Combined with convenient and smooth operating procedures and topic recommendation functions, it lowers the threshold for employee participation, improves the usability and attractiveness of the platform architecture, promotes the active participation of all employees, and enhances the enterprise's market vitality and data quality.

[0067] Compared with the prior art, the beneficial effects of the technical solution of the present invention are:

[0068] The present invention proposes an enterprise internal decision-making method, system and platform architecture based on the aggregation of multi-source data information. The multi-source data information obtained includes enterprise internal data information and enterprise external data information, avoiding the problem of single and isolated data sources, integrating deep learning, blockchain, natural language processing technology and big data analysis, aggregating and analyzing multi-source data information, realizing the creation of prediction topics, recommended topics, forecast topics and reward distribution, integrating internal employee participation and external multi-source data, assisting internal enterprise decision-making, supporting large-scale and high-frequency decision-making needs, and improving broad market participation, improving the accuracy and efficiency of internal enterprise decision-making, mobilizing employee participation enthusiasm, and improving the comprehensiveness of internal enterprise decision-making. In the system and platform architecture proposed by the present invention, combined with the internal and external strategies of the enterprise and modular system design, flexible replacement and expansion of components can be achieved. In the subsequent system development and maintenance process, new functions or modules can be easily added. With the layered platform architecture, it can flexibly adapt to enterprise development or market environment changes and meet fast-paced business needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A schematic diagram showing a flow chart of an enterprise internal decision-making method based on multi-source data information aggregation proposed in an embodiment of the present invention;

[0070] Figure 2 A schematic diagram showing the process of creating a topic proposed in an embodiment of the present invention;

[0071] Figure 3 A schematic diagram showing a process of generating topics for matching and recommending enterprise employees, as proposed in an embodiment of the present invention;

[0072] Figure 4 A schematic diagram showing a process for settling accounts and distributing rewards to enterprise employees who participate in forecasting, as proposed in an embodiment of the present invention;

[0073] Figure 5 A schematic diagram showing a multi-level incentive mechanism proposed in an embodiment of the present invention;

[0074] Figure 6 A structural diagram showing an enterprise internal decision-making system based on multi-source data information aggregation proposed in an embodiment of the present invention;

[0075] Figure 7 A diagram showing the architecture of an enterprise internal decision-making platform based on multi-source data information aggregation proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0076] The accompanying drawings are for illustrative purposes only and are not to be construed as limiting this patent;

[0077] In order to better illustrate this embodiment, some parts of the drawings may be omitted, enlarged, or reduced, and do not represent the actual size;

[0078] It is understandable to those skilled in the art that descriptions of certain well-known contents may be omitted in the drawings.

[0079] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0080] The positional relationships described in the drawings are for illustrative purposes only and should not be construed as limiting this patent;

[0081] Example 1

[0082] This embodiment proposes an enterprise internal decision-making method based on multi-source data information aggregation. The flowchart of this method is shown in Figure 1 , including the following steps:

[0083] S1: Acquire multi-source data information and pre-process the multi-source data information; the multi-source data information includes internal enterprise data information and external enterprise data information;

[0084] S2: Create an agenda item, which is a forecasted event related to the company's internal and external strategies. The agenda item is created using a decentralized smart contract based on blockchain technology.

[0085] S3: Analyzes multi-source data information based on deep learning models and natural language processing technology, and generates matching recommended topics for enterprise employees based on the analysis results;

[0086] S4: Presenting recommended topics to employees so they can participate in topic prediction. The prediction process is implemented based on decentralized smart contracts using blockchain technology.

[0087] S5: Determine whether the prediction results of the topics predicted by the company's employees meet the results set within the company. If so, the decentralized smart contract based on blockchain technology settles and distributes rewards to the company's employees who participated in the prediction, and assists the company's internal decision-making based on the prediction results of the topics predicted by the company's employees. Otherwise, conduct internal research and judgment.

[0088] In this embodiment, based on the prediction results of the topics predicted by the company's employees, the process of assisting the company's internal decision-making depends on the topics created during the specific implementation. Some created topics do not have clear and fixed expected results within the company, so the prediction results of the topics predicted by the company's employees are directly used for internal decision-making; for some topics, the company has preset expected results in advance. In this case, if the prediction results of the topics predicted by the company's employees do not meet the results set internally, the company can combine information from other channels for further research and judgment.

[0089] The enterprise internal decision-making method based on multi-source data information aggregation proposed in this embodiment obtains multi-source data information including enterprise internal data information and enterprise external data information, avoiding the problem of single and isolated data sources under traditional methods, integrating deep learning, blockchain, natural language processing technology and big data analysis, aggregating and analyzing multi-source data information, and realizing the creation of prediction topics, recommended topics, forecasted topics and reward distribution, assisting enterprise internal decision-making, supporting large-scale and high-frequency decision-making needs, and promoting broad market participation, improving the accuracy and efficiency of enterprise internal decision-making, mobilizing the enthusiasm of employees to participate, and improving the comprehensiveness of enterprise internal decision-making.

[0090] Example 2

[0091] In this embodiment, preprocessing of multi-source data includes multi-source data cleansing, multi-source data integration, and multi-source data transformation. Blockchain-based decentralized smart contracts include event creation contracts, betting contracts, transaction contracts, settlement contracts, and governance contracts. These blockchain-based decentralized smart contracts are all built using smart contract development languages such as Solidity and Vyper.

[0092] Internal enterprise data information includes basic employee data, employee behavior data, and information related to business decision-making. Basic employee data includes: education, age, work experience, interests, expertise, and work ability; external enterprise data information includes historical market transaction data, historical external environment impact data (such as economy, politics, trade, and market, etc.), and current external environment impact data.

[0093] In this embodiment, topics are created by creators with specific permissions. When a topic is created, a decentralized event creation contract based on blockchain technology is implemented. The event creation contract defines a data structure, which includes an event structure and a verification function. The verification function is used to authenticate the creator's identity and verify whether the creator's permissions are legal. At the same time, it performs format and logic checks on the created topic.

[0094] In this embodiment, if Figure 2 As shown in the figure, the process of creating a topic is as follows:

[0095] S21: The event creation contract is triggered by the creation request submitted by the creator;

[0096] S22: The event creation contract calls the verification function to verify whether the creator's permissions are legal. If so, S23 is executed; otherwise, the event creation contract terminates the topic creation process and returns an error prompt;

[0097] S23: The event creation contract performs a format and logic check on the created topic. If the check fails, the event creation contract terminates the topic creation process and returns an error message, requiring the creator to correct it and submit it. If the check passes, S24 is executed.

[0098] S24: The event creation contract packages the created topics into topics for enterprise employees to participate in prediction and broadcasts them to the blockchain network. The nodes in the blockchain perform consensus verification. When the consensus verification passes, the topic is recorded in the blockchain ledger, and the topic creation is successful. All created topics constitute the topic library.

[0099] In this embodiment, when analyzing multi-source data information based on deep learning models and natural language processing technology, the deep learning model is first trained and optimized using historical market transaction data information, historical external environmental impact data information, and historical enterprise internal decision-making data, and the model parameters of the deep learning model are dynamically adjusted.

[0100] Use deep learning models and natural language processing technology to analyze basic data information of corporate employees, corporate employee behavior data information, current market transaction data information, and current external environment impact data information.

[0101] In this embodiment, the analysis of data information aims to deeply explore the value of market data and provide comprehensive and accurate data support for internal enterprise decision-making. Therefore, this embodiment analyzes basic employee data, employee behavior data, current market transaction data, and current external environmental impact data. It integrates multiple data processing technologies and algorithms and has powerful data collection, storage, analysis, and visualization capabilities. Compared with conventional data analysis tools currently on the market, it is more focused on the data analysis needs of internal enterprise decision-making scenarios.

[0102] The multi-source data integration obtained in this embodiment covers all aspects and related data sources, including but not limited to the following:

[0103] Employee behavior data: This collects all employee activity data on the platform, including browsing event records, betting or transaction records (including betting or transaction events, outcome options, betting or transaction amounts, betting or transaction times, etc.), account operation records (deposits, withdrawals, account settings, etc.), and records of participation in governance activities. Through front-end tracking technology and back-end logging systems, employee behavior data is collected in real time and standardized for subsequent analysis. The betting or trading mentioned here depends on the topic in which the employee participates. Employees can "bet" by splitting the same bet into different lottery tickets or "trade" through bidding and matching similar to stock trading. The method used, betting or trading, depends on the predicted topic and the number of participants. "Trading" refers to the actions of employees in making bids and buying and selling contracts during the process of participating in topic predictions, which is parallel to "betting."

[0104] Market transaction data: This records the details of all bets or trades on all prediction issues in the market, including changes in the betting or trading share of each outcome option, betting or trading price fluctuations, transaction time, and information about both parties to the transaction. This data is directly obtained from the blockchain ledger and backend database to ensure data accuracy and completeness.

[0105] External Data Sources: To enrich data dimensions and improve forecast accuracy, we integrate external data sources such as industry news, market research reports, social media sentiment data, and macroeconomic indicators. We regularly acquire external data through data interfaces or web crawlers and integrate and analyze it with internal data. For example, we collaborate with professional financial news websites to obtain real-time economic data and industry news to inform market trend forecasts.

[0106] A variety of data extraction techniques and algorithms were used in the data analysis process, which mainly includes the following steps:

[0107] Data cleaning and preprocessing: Collected raw data is cleaned to remove invalid, redundant, and erroneous data. Examples include addressing missing values, correcting data format errors, and filtering outliers. Furthermore, data is standardized and normalized to make data from different sources and types comparable. For example, user bet or transaction amounts are converted to a standard currency unit and timestamps are formatted.

[0108] Data Mining and Feature Engineering: Data mining algorithms (such as association rule mining, cluster analysis, and classification algorithms) are used to discover underlying patterns and regularities in the data. Feature engineering techniques are used to extract valuable features, such as user behavior characteristics (e.g., betting or trading frequency, stability of betting or trading preferences), market trend characteristics (e.g., trends in betting or trading ratios, price fluctuations), and event correlation characteristics (e.g., correlations between different event outcomes). These features serve as input for subsequent prediction models and analytical reports.

[0109] Predictive model construction and training: Build predictive models based on machine learning and deep learning algorithms, such as regression models (for predicting market trends, betting or trading price movements, etc.) and classification models (for predicting event outcomes, user behavior classification, etc.). Use historical data to train and optimize predictive models, and continuously adjust model parameters to improve prediction accuracy. For example, a neural network algorithm is used to build an event outcome prediction model. Training on a large amount of historical forecast event data enables the model to learn the patterns in how different factors influence event outcomes.

[0110] Data analysis and visualization: Comprehensively evaluate and interpret the analysis results to generate a detailed data analysis report. The report content includes market overview analysis (such as the number of participants, distribution of betting or transaction amounts, ranking of popular prediction events, etc.), user behavior analysis (such as betting or transaction preferences of different user groups, activity analysis, etc.), prediction model evaluation (such as evaluation of indicators such as model accuracy, recall rate, F1 value), and market risk assessment (such as risk indicator calculation, risk trend analysis, etc.). At the same time, the analysis results are presented to corporate decision makers and relevant users in an intuitive and easy-to-understand manner through data visualization technology (such as charts, dashboards, etc.). For example, a line graph showing the change in betting or transaction ratio over time can be drawn to show changes in market trends; a user activity heat map can be created to intuitively present the participation of users in different regions or departments.

[0111] Match topics from the topic library for corporate employees and then generate the final recommended topics. For the specific process, see Figure 3, including: based on the analysis results of corporate employee behavioral data information, such as collecting corporate employee behavioral data on the network platform (such as clicks, browsing, comments, etc.) and storing it in the topic library. This process can use natural language processing technology and distributed storage technology to ensure the real-time and scalability of the data. Match the interests of corporate employees with the topics in the topic library. This process can use machine learning models (such as content recommendations) for matching to improve the accuracy of matching. Based on the matching results, generate personalized topics related to the interests of corporate employees. This process combines the basic information of corporate employees to obtain corporate employee portraits and contextual information, and uses natural language processing technology to generate personalized topics. Then, use the recommendation algorithm to generate a list of topics related to the interests of corporate employees. This process can use deep learning models (such as neural networks, convolutional neural networks, etc.) for topic generation to improve the diversity and relevance of topics.

[0112] The process of generating the final recommended topics includes: first, filtering irrelevant topics from the topic list to exclude topics that are not relevant to employee needs to ensure the accuracy of the recommended results. This can be done using rule-based or machine learning filters. Next, similarity calculation is performed to refine the topics. This process calculates the similarity between employee interests and topics to determine their relevance. Similarity calculations can be performed using algorithms such as cosine similarity and Jaccard similarity to improve the accuracy of recommendations. After topic generation, a topic recommendation algorithm is used to recommend topics. Topics are ranked based on their relevance and employee interests. Ranking algorithms such as LambdaRank and Learning to Rank can be used to prioritize the recommended topics. Next, topic weights are assigned to obtain the final recommendations. Employees provide feedback on the final recommended topics, and the weights are updated and adjusted based on the positive and negative feedback. This process can use online learning algorithms (such as reinforcement learning and online gradient descent) to update the weights, ensuring the real-time and accurate recommendation results.

[0113] In this embodiment, the process of enterprise employees participating in topic prediction is implemented based on a decentralized betting or trading contract based on blockchain technology. The betting contract includes an enterprise employee betting structure, a price calculation function, a balance check function, and a betting record function. The enterprise employee betting structure includes an enterprise employee identity record, the topic selected by the enterprise employee, the topic prediction betting result option, the topic prediction betting quantity, and the time when the enterprise employee participated in the topic prediction betting; the trading contract includes: an enterprise employee transaction structure, a price calculation function, a balance check function, and a transaction record function. The enterprise employee transaction structure includes an enterprise employee identity record, the topic selected by the enterprise employee, the topic prediction transaction result option, the topic prediction transaction quantity, and the time when the enterprise employee participated in the topic prediction transaction, indicating that the enterprise employees buy and sell contracts through a bidding model. The bidding model refers to a trading method in which the contract price is determined by the quotations of the buyer and seller during the contract transaction process.

[0114] When enterprise employees choose to participate in topic prediction by betting, the process of enterprise employees participating in topic prediction is as follows:

[0115] S31: The betting contract is triggered after the company employee selects the topic prediction and topic prediction betting result options and determines the topic prediction betting amount;

[0116] S32: The betting contract calls the balance check function to check whether the token or points balance in the enterprise employee account is sufficient to pay the bet amount. If so, execute step S33; otherwise, the balance is insufficient, the betting contract is terminated, the event prediction process ends, and the enterprise employee is prompted to recharge;

[0117] S33: The betting contract calls the price calculation function to calculate the latest betting price of the betting result option based on the current betting situation of the predicted event; in this process, the existing number of bets on the betting result option, the betting distribution of other betting result options, and the preset price adjustment algorithm (such as adjusting the price according to the change in the betting ratio) are comprehensively considered to ensure that the betting price can reflect market expectations in real time.

[0118] S34: After the latest betting price is determined, the contract deducts the corresponding tokens or points from the company employee's account and calls the betting record function to record the company employee's betting information in the blockchain ledger. Simultaneously, the betting contract updates the betting ratio and betting amount information for the predicted event in the market data. In this embodiment, the medium for company employees to participate in topic prediction and betting is tokens or points.

[0119] When enterprise employees choose to participate in topic forecasting by means of transactions, the process of enterprise employees participating in topic forecasting is as follows:

[0120] S301: The transaction contract is triggered after the enterprise employee selects the topic prediction and topic prediction transaction result options and determines the topic prediction transaction quantity;

[0121] S302: The transaction contract calls the balance check function to check whether the token or points balance in the enterprise employee's account is sufficient to pay the transaction amount. If so, step S33 is executed; otherwise, the balance is insufficient, the transaction contract is terminated, the event prediction process ends, and the enterprise employee is prompted to recharge;

[0122] S303: The trading contract calls the price calculation function to calculate the final transaction price of the transaction result option based on the current predicted event transaction status. The transaction price is determined by matching the quotes of the buyer and seller.

[0123] S304: After the final transaction price is determined, the transaction contract deducts the corresponding tokens or points from the enterprise employee's account and calls the transaction record function to record the enterprise employee's transaction information in the blockchain ledger. At the same time, the transaction contract updates the transaction ratio and transaction amount information of the predicted event in the market data.

[0124] When settling accounts and distributing rewards to employees of participating companies, a decentralized settlement contract based on blockchain technology is implemented. The settlement contract includes: result confirmation function, valid bet or transaction share calculation function, reward distribution function, and account balance update function;

[0125] like Figure 4 As shown in the figure, the process of settling accounts and distributing rewards to employees of enterprises participating in the forecast is as follows:

[0126] S41: The result of enterprise employees participating in event prediction triggers the settlement contract;

[0127] S42: The settlement contract calls the valid betting or transaction share calculation function, traverses the enterprise employee betting or transaction information records in the blockchain ledger, filters out valid bets or transactions according to preset rules, and calculates the valid betting or transaction share of each result option; during the calculation process, some abnormal betting or transaction behaviors may be excluded, such as violations of betting or transaction rules, large abnormal bets or transactions, etc., to ensure the fairness of settlement.

[0128] S43: Based on the calculated valid betting or transaction share and the total amount of tokens or points in the reward pool, the settlement contract calls the reward allocation function to calculate the reward amount that each betting or trading enterprise employee should receive according to the set settlement algorithm; in this embodiment, the source of the reward pool can be a certain proportion of the total betting or transaction amount or a reward fund pre-set by the market operator.

[0129] S44: The settlement contract calls the account balance update function to add the calculated reward amount to the account balance of the enterprise employee who placed the correct bet or transaction, and records the settlement process and results in detail in the blockchain ledger, including the event results, user bets or transactions, and reward distribution details.

[0130] In this embodiment, if Figure 5 As shown in the figure, the reward also adopts a multi-level incentive mechanism, which includes material rewards, career development and spiritual recognition. The material rewards include "bonuses and coupons", the career development includes "training opportunities and job promotion", and the spiritual recognition includes "high-level recognition and exclusive titles", which fully mobilize the enthusiasm of employees, improve market activity, and ensure the long-term stable development of the prediction market. The specific design is as follows: (1) Accuracy reward. Multi-dimensional evaluation indicators are used: 1) Number of consecutive accurate predictions: The number of times an employee accurately predicts the results of an event is recorded and counted. For example, different levels such as 3, 5, and 10 consecutive accurate predictions are set. Each time a new level is reached, the reward intensity is increased accordingly. For example, if an employee makes 3 consecutive accurate predictions, he or she can get a basic reward bonus, such as an additional 10% token or point reward; if an employee makes 5 consecutive accurate predictions, the reward bonus is increased to 20%; if an employee makes 10 or more consecutive accurate predictions, in addition to a higher proportion of tokens or points rewards (such as 50%), he or she may also receive a special honor medal or a prominent display position in the market rankings, thereby improving the employee's visibility and reputation in the company's internal prediction market. 2) Forecast Accuracy: Employees are evaluated based on their overall forecast accuracy over a specific time period (e.g., one month, one quarter, or one year). Accuracy is divided into different intervals, such as 80%-85%, 85%-90%, 90%-95%, and 95% and above, with each interval corresponding to a different reward level. The higher the accuracy, the more generous the rewards, which may include not only tokens or points, but also priority participation in high-value forecast events, access to professional training courses or industry seminars, etc., incentivizing employees to continuously improve their forecast accuracy.

[0131] (2) Activity Rewards. Activity of betting or trading behavior: Count the total number of bets or transactions made by employees in a certain period. Different levels are divided according to the number of bets or transactions, such as 10-20 bets or transactions per month, 20-50 times, 50 times and above, etc. The more bets or transactions, the higher the activity reward. The reward can be in the form of tokens or points, or a chance to win a lottery (prizes include physical gifts, internal corporate benefits, value-added services, etc.), which increases the fun and attractiveness of the reward.

[0132] (3) Special Contribution Rewards

[0133] Information provision and analytical contributions. Employees who provide valuable external information (such as industry dynamics and market trends) that contributes to forecasted events, or who conduct in-depth data analysis and share insights, will be awarded special contribution rewards after evaluation and approval by market operators. These rewards can take the form of tokens or points, public recognition (such as letters of commendation published on internal company communication platforms), invitations to participate in internal expert seminars, and more. This encourages employees to actively share valuable information, enrich market information sources, and improve overall forecasting capabilities.

[0134] Through this diversified and clearly structured reward and incentive mechanism, we comprehensively assess employee performance in areas such as forecast accuracy, activity, special contributions, and teamwork, stimulating their intrinsic motivation, fostering a positive and healthy market competition environment, and promoting the continued prosperity and development of the company's internal forecasting market. Furthermore, during implementation, we will continuously optimize and adjust the reward mechanism based on market feedback and corporate development needs to ensure its effectiveness and adaptability.

[0135] This embodiment proposes an internal enterprise decision-making method based on multi-source data information aggregation, which also includes risk management. The preset risk judgment standards for risk management are: abnormal betting or transaction amounts, abnormal betting or transaction frequencies, abnormal betting or transaction behavior patterns, and abnormal sources and flows of funds; for risks with different preset standards, preset risk response strategies are provided, including: betting or transaction restriction strategies, investigation and warning strategies, and market intervention strategies.

[0136] Abnormal betting or trading amounts: Set a cap on the amount a single employee can bet or trade within a certain period of time. If an employee's betting or trading amount exceeds this cap, it will be considered a potential risk. For example, if a single employee's betting or trading amount within an hour is capped at 1,000 tokens, a risk warning will be triggered if the employee's betting or trading amount reaches or exceeds this value within that period.

[0137] Abnormal betting or trading frequency: Monitor the betting or trading frequency of company employees. If users place bets or trades frequently within a short period of time (e.g., the number of bets or trades per minute exceeds a certain threshold), there may be a risk of market manipulation or irrational betting or trading. Reasonable thresholds are set based on historical betting or trading data and the average market betting or trading frequency. For example, bets or trades exceeding 5 times per minute are considered abnormal.

[0138] Abnormal betting or trading behavior patterns: Data analysis is used to identify patterns in employee betting or trading behavior, such as consecutive bets or trades on the same outcome, and unusual betting or trading preferences within specific topics or time periods. Machine learning algorithms (such as cluster analysis and anomaly detection algorithms) are used to model and analyze employee betting or trading behavior, identifying deviations from normal patterns. For example, if an employee places bets or trades on the same outcome more than 10 times in a row, and this behavior is clearly inconsistent with overall market trends, this behavior will be marked as abnormal.

[0139] Betting or Trading Restrictions: Employees with unusual betting or trading amounts or frequencies will be automatically restricted from further betting or trading. Measures such as temporarily freezing account betting or trading permissions, lowering betting or trading limits, or extending betting or trading intervals can be implemented to prevent further escalation of risk. For example, if an employee's betting or trading amount exceeds the upper limit, the system will immediately freeze their account betting or trading permissions for one hour and reduce their daily betting or trading limit to 50%.

[0140] Investigation and Alert Strategy: When unusual betting or trading patterns, or unusual sources and flows of funds, are detected, an investigation is initiated. First, an alert is sent to company employees, requesting an explanation of their betting or trading behavior. Second, the anomaly is reported to the company's internal risk management team or relevant departments for further investigation and verification. For example, a system message or email alert may be sent to the user, informing them of the unusual betting or trading behavior and requesting a reasonable explanation within 24 hours.

[0141] Market intervention strategy: In extreme cases, such as when large-scale abnormal betting or trading leads to market disorder or forecast results deviate significantly from normal ranges, market intervention measures will be implemented. This may include suspending or canceling some betting or trading transactions, adjusting betting or trading rules (such as extending betting or trading hours, modifying settlement methods, etc.), issuing market announcements to stabilize market sentiment, etc. For example, if a forecast topic receives a large number of abnormal bets or trades in a short period of time, resulting in a serious imbalance in the betting or trading ratio, betting or trading on that topic may be suspended, and the market situation will be reassessed before deciding whether to continue or adjust the betting or trading rules.

[0142] In step S5, it is determined whether the prediction results of the topics in which the enterprise employees participate in the prediction meet the results set internally by the enterprise. The results set internally by the enterprise include: the result determination time scheduled within the enterprise, official statistical release data and internal decisions made by the enterprise. The source of the results is authoritative and reliable. For example, for market data prediction related to the enterprise's business, the results may come from reports released by authoritative market research institutions; for internal project progress predictions, the results are based on the evaluation and decision-making of relevant departments within the enterprise.

[0143] Assisted internal corporate decision-making can include: market and sales forecasting. Based on employee betting or transaction analysis of market trends, consumer behavior, etc., predict future changes in market demand for products or services, evaluate competitors' market strategies, and predict their potential impact on the company's market share. Leverage historical sales data, seasonal factors, market trends, etc. to predict future sales and sales growth rates. Analyze the effectiveness of sales channels and predict the potential contribution of different channels to sales. Decision-making effect prediction: Based on the topics and forecast results of employee participation in decision-making, predict the degree to which the decision will improve corporate performance, evaluate the possibility and risks of decision implementation, and the role of employee participation in decision-making in improving decision execution.

[0144] In this embodiment, the decentralized smart contract based on blockchain technology adopts multiple dimensions to select the blockchain platform. The multiple dimensions include transaction processing speed, throughput indicators, and also include:

[0145] The prediction event processing capability indicator per unit time is used to measure the processing efficiency of the blockchain platform during operations such as the creation, betting, trading, and settlement of highly concurrent prediction events;

[0146] Dynamic node expansion adaptability indicators are used to measure the stability of system performance, data consistency, and transaction processing capabilities during the dynamic increase or decrease of nodes on the blockchain platform;

[0147] Smart contract vulnerability detection coverage indicator, used to evaluate the blockchain platform's ability to detect and prevent smart contract vulnerabilities (such as reentrancy attacks, overflow attacks, etc.);

[0148] Data desensitization and authorized access granularity indicators are used to measure the blockchain platform's ability to desensitize sensitive information (such as employee betting or transaction records, corporate strategy-related forecast data, etc.) during data storage and transmission, as well as the flexibility of implementing fine-grained authorized access control;

[0149] The Smart Contract Development Language and Enterprise Technology Stack Compatibility Index is an evaluation indicator used to measure the degree of integration between the smart contract development languages supported by the blockchain platform (such as Solidity, Vyper, etc.) and the enterprise's existing technology system (including development frameworks, tool chains, and programming languages);

[0150] The blockchain platform introduces a node trust weighted consensus mechanism, assigns different voting weights according to the credibility of internal nodes of the enterprise (such as evaluation based on factors such as the node's historical performance and the importance of the department to which it belongs), and adopts an optimized index structure (such as B+ tree index, hash index, etc.) for storage to improve data query efficiency. At the same time, an event tracing data storage model is designed to store the complete life cycle data of each prediction event (including detailed records of each stage such as creation, betting or trading, settlement, etc.) in a chain structure.

[0151] Furthermore, innovative network configuration and security policy adaptation enable targeted blockchain network configuration based on the enterprise's internal network topology and security policies. For example, multiple regional subnetworks are established within the enterprise, with blockchain nodes divided according to different business functions and security levels. Network access control lists (ACLs) and virtual private network (VPN) technologies are used to achieve secure communication and data isolation between nodes. The innovative use of "dynamic network traffic monitoring and intelligent routing" technology monitors network traffic in real time. Based on forecasts of market activity and node load, network routing strategies are automatically adjusted to optimize data transmission paths, improve network transmission efficiency, and ensure system stability and reliability under high concurrency conditions.

[0152] Through these innovative approaches to blockchain platform selection and adaptation, our internal prediction market system is able to fully leverage the advantages of blockchain technology, providing the enterprise with a secure, efficient, flexible, and easily scalable prediction market operating environment, effectively supporting its decision-making optimization and business development needs. During implementation, we will continue to monitor blockchain technology developments and continuously optimize and refine our platform selection and adaptation strategies to ensure the system maintains leading technical standards and superior performance.

[0153] Example 3

[0154] like Figure 6 As shown, this application proposes an enterprise internal decision-making system based on multi-source data information aggregation, which is deployed in a distributed network environment and includes:

[0155] A multi-source data information acquisition and processing module is used to acquire multi-source data information and pre-process the multi-source data information; the multi-source data information includes internal enterprise data information and external enterprise data information;

[0156] The topic creation module is used to create topics, which are predicted events related to the internal and external strategies of the enterprise. The topic creation is realized by decentralized smart contracts based on blockchain technology;

[0157] The topic recommendation module analyzes multi-source data information based on deep learning models and natural language processing technology, and generates matching recommended topics for enterprise employees based on the analysis results;

[0158] The prediction participation module displays recommended topics to corporate employees for them to participate in topic prediction. The prediction participation process is implemented based on the decentralized smart contract of blockchain technology. In the process of participating in topic prediction, corporate employees place bets or trades on the betting or trading results displayed on the topic. The prediction participation module displays betting or trading result options, betting or trading prices, and balances.

[0159] The prediction reward distribution module is used to determine whether the prediction results of the topics predicted by enterprise employees meet the results set within the enterprise. If so, the decentralized smart contract based on blockchain technology settles and distributes rewards to the employees participating in the prediction, and assists the internal decision-making of the enterprise based on the prediction results of the topics predicted by the employees. Otherwise, internal research and judgment will be conducted.

[0160] A distributed network environment consists of interconnected computer systems distributed across different locations, with no central node, and each point connected to at least two lines. In this network environment, if any line fails, communication can be routed through other links, greatly improving the reliability and stability of the system. For decision-making systems, even if certain nodes or links fail, the system can still maintain normal operation, ensuring the continuity and accuracy of decision-making. In this embodiment, the distributed network environment adopts the following design:

[0161] (1) Network topology design.

[0162] Hybrid topology applications: A combination of star and mesh topologies is used. Key nodes are connected using a mesh topology to ensure redundant links and high reliability, such as multi-link configurations between a corporate headquarters data center and key servers in branch offices. A star topology is used within branches or for non-critical nodes and core nodes, facilitating network management and reducing complexity.

[0163] Regionalized network segmentation: Divide the network into regional zones based on geographic distribution, business functions, or security requirements. Each zone is relatively independent and has strict access controls, connected by high-performance routers or gateways. For example, segmentation by department ensures security, optimizes traffic flow, and ensures system stability.

[0164] (2) Network Node Deployment and Management

[0165] Distributed node load balancing: Using intelligent algorithms, similar to the principles of CDN (Content Delivery Network), business requests such as data queries, betting or trading operations are dynamically allocated based on the request content, location and node load, ensuring load balancing across nodes, improving response speed and processing efficiency.

[0166] Dynamic node expansion and management: Leveraging cloud computing and virtualization technologies, new nodes are automatically created and integrated based on business needs, enabling flexible adjustment of network resources, ensuring system performance and scalability, and adapting to enterprise development.

[0167] (3) Network security and protection mechanisms

[0168] Distributed firewall and IDS (Intrusion Detection System) integration: Firewalls and IDS are deployed in each region or key node, working together and sharing information to form a distributed security protection system. If an intrusion is detected in one location, local defenses can be implemented and the entire network notified, improving overall security.

[0169] Data encryption and enhanced transmission security: High-strength encryption algorithms (such as SSL / TLS, important protocols in the field of network security) are used in enterprise operations and data transmission and storage. Sensitive data is encrypted during storage, and digital certificate authentication technology is used to ensure data confidentiality, integrity, and the authenticity of the identities of both parties in the transmission.

[0170] The distributed network environment built through these measures provides high availability, scalability, performance optimization and data security for the company's internal decision-making system, strongly supporting the healthy development of corporate decision-making-related work and the forecast market.

[0171] From a system perspective, enterprise employees' betting or trading operations and input information are equivalent to inputting a series of key information into the system, including the user's identity (used to uniquely identify the betting or trading user), the unique identifier of the selected prediction topic, the betting or trading outcome option, the number of tokens or points bet or traded, and the betting or trading timestamp. This information is encapsulated into a betting or trading request data structure, which serves as the input for the system to process betting or trading operations.

[0172] When an enterprise employee issues a betting or transaction request, the system will process and manage the betting or transaction request. The basic unit of processing and management is the order. When an enterprise employee submits a betting or transaction request, the order matching engine converts it into an order. Each order contains all the information of the above betting or transaction operation and is assigned a unique order number for tracking and management in the system.

[0173] The order matching engine monitors newly generated orders in real time and matches them with existing unmatched orders. The matching process is based on the following principles and algorithms:

[0174] Price Priority: For different betting or trading outcomes for the same prediction topic, we prioritize the betting or trading options based on price, either from high to low (for buyer orders, meaning orders seeking to buy bets or trade shares at a higher price) or from low to high (for seller orders, meaning orders seeking to sell bets or trade shares at a lower price). We prioritize matching orders with the most favorable prices to ensure fair and efficient trading.

[0175] Time priority principle: If the price is the same, orders are matched based on the time they were generated. Orders submitted first are given priority to prevent unfair trading order caused by factors such as network delays.

[0176] Matching Algorithm Implementation: Efficient algorithms (such as those based on binary search trees or hash tables to store and retrieve orders) are used to quickly locate matching orders. When a new order enters the system, the engine searches the existing order list for a matching order. If a matching order is found (for example, a buyer's order and a seller's order agree on the prediction topic, outcome options, and price), the transaction is matched, the status of both orders is updated to matched, and the corresponding token or point transfer operation is performed (the bet or transaction amount is deducted from the buyer's account and added to the seller's account). If no fully matching order is found, the new order is added to the list of unmatched orders, awaiting a possible subsequent matching opportunity.

[0177] Example 4

[0178] like Figure 7 As shown, this application proposes an enterprise internal decision-making platform architecture based on multi-source data information aggregation. The platform architecture adopts a layered design, including: a data acquisition and pre-processing layer, an intelligent engine layer, a business logic layer for predicting betting or trading rules, an enterprise employee interface layer, and a data storage layer;

[0179] The data acquisition and preprocessing layer is used to deploy multi-source data information acquisition and processing modules, the intelligent engine layer is used to deploy topic creation modules and topic recommendation modules, and the predicted betting or transaction rules business logic layer is used to deploy the predicted participation module and the predicted reward distribution module; the enterprise employee interface layer is used to provide a friendly interactive interface for enterprise employees. In this embodiment, the interface provides enterprise employee registration and login methods, which are integrated with enterprise identity authentication. After login, hot topics related to the enterprise's strategic development are displayed, and a search function is provided; the data storage layer is used to save predicted betting or transaction records, enterprise employee information and other data.

[0180] The data output by the multi-source data information acquisition and processing module flows through the data acquisition and processing layer and enters the intelligent engine layer. The intelligent engine layer includes deep learning models, natural language processing components and intelligent event generators, and is responsible for topic generation and topic recommendation;

[0181] The business logic layer of the predicted betting or trading rules is equipped with a centralized smart contract platform, which is responsible for predicting trading management, rule formulation and market monitoring; the transaction cycle on the decentralized smart contract platform includes: initiating contract creation, broadcasting, verification, storage, execution and settlement.

[0182] In the platform architecture proposed in this embodiment, in the selection of the front-end technology stack, on the basis of ensuring that basic interaction requirements are met, a number of innovative elements are incorporated to provide an excellent user experience and efficient operation process.

[0183] (1) Technology selection innovation driven by user experience

[0184] The cross-platform and progressive web application (PWA) fusion innovation innovatively uses a cross-platform framework (such as ReactNative or Flutter) combined with progressive web application (PWA) technology to build front-end applications. This fusion approach allows the system to provide a smooth and efficient user experience on mobile devices in the form of native applications, and to achieve convenient access on the desktop through a web browser without the need for users to install additional software. Through PWA technology, the system has functions such as offline access and push notifications, which greatly improves user convenience and participation. For example, even if employees are in a weak network or no network environment on their mobile devices, they can still browse previously loaded forecast event information, view personal betting or transaction history, etc. After the network is restored, the data will be automatically synchronized and updated to ensure that users can stay connected to the forecast market anytime, anywhere.

[0185] Optimization and innovation of interactive design and animation effects. Based on the analysis of corporate employee behavior and the principles of human-computer interaction, the interactive design and animation effects of the front-end interface have been deeply optimized and innovated. For example, on the forecast topic display page, a dynamic card-style layout is adopted. According to factors such as the popularity of the event, the approaching betting or transaction deadline, important information is highlighted through animation effects (such as card zooming, color gradients, etc.) to attract user attention. During the betting or trading operation, real-time feedback animations are introduced, such as the ripple effect after clicking a button, and the fireworks celebration animation after a successful bet or transaction, etc., to enhance the fun and sense of accomplishment of the user's operation and increase the enthusiasm of user participation. At the same time, the page switching and loading animations are optimized, and transition animations and preloading technology are used to make page transitions more natural and smooth, reduce user waiting time, and improve the overall user experience.

[0186] Personalized interface customization innovation. In order to meet the personalized needs of employees in different companies, the personalized customization function of the front-end interface is realized. Through the analysis of corporate employee portraits and behavioral data, personalized theme styles (such as color themes, font styles, etc.) and interface layouts (such as position adjustment of commonly used function modules, priority display of events of concern, etc.) are automatically generated for corporate employees. Corporate employees can also manually adjust the interface settings according to their preferences, such as choosing favorite color combinations, hiding infrequently used function modules, etc. This kind of personalized customization not only improves the user's sense of identity and belonging to the system, but also improves the user's operating efficiency, allowing them to focus more on the predicted events and functions of interest.

[0187] (2) Technical architecture innovation that balances performance and maintainability

[0188] Micro-frontend architecture application innovation. In the front-end technical architecture, the concept of micro-frontend architecture is introduced to split the front-end application into multiple independent sub-applications, each of which is responsible for a specific functional module (such as user login and registration, prediction event browsing and betting or trading, account management, etc.). This architectural innovation realizes the independent development, deployment and updating of modules, improves development efficiency and system maintainability. For example, when the betting or trading function module needs to be upgraded, the sub-application can be developed and deployed independently without affecting the normal operation of other functional modules. At the same time, the micro-frontend architecture facilitates integration with other internal systems of the enterprise. Through unified interface specifications and routing mechanisms, seamless switching and data interaction between front-end pages of different systems can be achieved, thereby improving the synergy of the enterprise's overall information system.

[0189] Innovation in componentization and code reuse. We emphasize front-end component development and have established a rich library of reusable components. These components cover all levels from basic UI elements (such as buttons, input boxes, tables, etc.) to complex business components (such as forecast event cards, betting or trading operation panels, market data analysis charts, etc.). Through componentized development, not only is the reusability of the code improved and the development workload reduced, but the consistency and stability of the front-end interface are also guaranteed. In the component design process, the versatility and scalability of the components are fully considered, and flexible properties and event mechanisms are adopted, so that the components can easily adapt to different business scenarios and changes in demand. For example, the forecast event card component can display different information layouts and styles according to different types of forecast events (such as market trend forecasts, project progress forecasts, etc.), which can be achieved through simple property configuration.

[0190] (3) Technology selection and integration innovation with existing enterprise systems

[0191] Integration innovation with the enterprise's unified identity authentication system. During the selection of the front-end technology stack, full consideration was given to the integration requirements with the enterprise's existing unified identity authentication system. Authentication protocols (such as OAuth 2.0 or SAML) that comply with enterprise security standards and technical specifications were adopted to implement single sign-on (SSO) functionality for users within the enterprise's internal prediction market system. Through deep integration with the unified identity authentication system, users no longer need to repeatedly register and log in on the prediction market platform and can directly access the system using their internal enterprise account, improving the convenience and security of user operations. At the same time, the system can obtain the user's identity information and permission data within the enterprise in real time, and dynamically display the corresponding functional modules and data content based on the user's role and permission, ensuring the security and compliance of system access.

[0192] Integration and innovation of front-end performance monitoring and enterprise operation and maintenance system. In order to ensure the stable operation of front-end applications, front-end performance monitoring is organically integrated with the existing operation and maintenance system of the enterprise. Front-end performance monitoring tools (such as Sentry, New Relic, etc.) that support integration with commonly used enterprise operation and maintenance tools (such as ELKStack, Prometheus, etc.) are selected to collect performance indicators of front-end applications (such as page loading time, interface request response time, user operation behavior path, etc.) and error information in real time. These data are transmitted to the enterprise operation and maintenance platform in real time, and after analysis and processing, they provide the operation and maintenance team with comprehensive front-end performance insights and fault warnings. The operation and maintenance team can promptly discover and resolve front-end performance problems based on monitoring data, optimize system configuration and resource allocation, ensure that front-end applications always maintain good performance status, and provide users with a stable and efficient service experience.

[0193] Through the innovative measures implemented during the selection and development of the front-end technology stack, the platform architecture proposed in this embodiment provides a front-end interface that is not only aesthetically pleasing and easy to use, but also excels in performance, maintainability, and integration with existing enterprise systems. This provides employees with an efficient, convenient, and personalized platform for interactive prediction markets, effectively supporting internal decision-making and broad employee participation. In future development, we will continue to monitor innovative trends in front-end technology and continuously optimize and improve the front-end technology stack to adapt to business growth and the changing needs of employees.

[0194] The embodiments are provided merely to illustrate the present invention and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications may be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the claims.

Claims

1. An enterprise internal decision-making method based on multi-source data information aggregation, characterized in that: The following steps are involved: Acquire multi-source data information and pre-process the multi-source data information; the multi-source data information includes internal enterprise data information and external enterprise data information; Creating topics, which are forecasted events related to the company's internal and external strategies, and are implemented using decentralized smart contracts based on blockchain technology; Analyze multi-source data information based on deep learning models and natural language processing technology, and generate matching recommended topics for corporate employees based on the analysis results; Presenting recommended topics to company employees for them to participate in topic prediction. The prediction process includes employees placing bets or trades on topics to form bet or trade prices. The prediction process is implemented based on decentralized smart contracts based on blockchain technology. Determine whether the prediction results of the topics predicted by the company's employees meet the results set internally by the company. If so, the decentralized smart contract based on blockchain technology settles and distributes rewards to the employees participating in the prediction, and assists the company's internal decision-making based on the prediction results of the topics predicted by the company's employees. Otherwise, conduct internal research and judgment.

2. The enterprise internal decision-making method based on multi-source data information aggregation according to claim 1 is characterized in that: The pre-processing of multi-source data information includes: multi-source data cleaning, multi-source data integration and multi-source data transformation; the decentralized smart contracts based on blockchain technology include: event creation contracts, betting contracts, transaction contracts, settlement contracts and governance contracts, and the decentralized smart contracts based on blockchain technology are all constructed using the smart contract development language; The internal data information of the enterprise includes basic data information of enterprise employees, behavioral data information of enterprise employees and enterprise business decision-making information. The basic data information of enterprise employees includes: education level, age, work experience, interests, expertise and work ability; the external data information of the enterprise includes historical market transaction data information, historical external environment impact data information and current external environment impact data information.

3. The enterprise internal decision-making method based on multi-source data information aggregation according to claim 2 is characterized in that: The topic is created by a creator with specific permissions. When the topic is created, a decentralized event creation contract based on blockchain technology is implemented. The event creation contract defines a data structure, which includes an event structure and a verification function. The verification function is used to authenticate the creator's identity and verify whether the creator's permissions are legal. At the same time, it performs format and logic checks on the created topic. The topic creation process is as follows: S21: The event creation contract is triggered by the creation request submitted by the creator; S22: The event creation contract calls the verification function to verify whether the creator's permissions are legal. If so, execute S23; Otherwise, the event creation contract terminates the topic creation process and returns an error message; S23: The event creation contract performs format and logic checks on the created topic. If the check fails, the event creation contract terminates the topic creation process and returns an error message, requiring the creator to correct the error and submit the topic. If the check passes, execute S24; S24: The event creation contract packages the created topics into topics for enterprise employees to participate in prediction and broadcasts them to the blockchain network. The nodes in the blockchain perform consensus verification. When the consensus verification passes, the topic is recorded in the blockchain ledger, and the topic creation is successful. All created topics constitute the topic library.

4. The enterprise internal decision-making method based on multi-source data information aggregation according to claim 3 is characterized in that: When analyzing multi-source data information based on deep learning models and natural language processing technology, we first use historical market transaction data, historical external environmental impact data, and historical internal corporate decision-making data to train and optimize the deep learning model, and dynamically adjust the model parameters of the deep learning model; Deep learning models and natural language processing technology are used to analyze the basic data information of corporate employees, corporate employee behavior data information, current market transaction data information, and current external environment impact data information, match topics for corporate employees from the topic library, and then generate final recommended topics. The process of generating the final recommended topics is: first filter out irrelevant topics, then perform similarity calculations and refine topics; after the topics are generated, recommend topics through the topic recommendation algorithm, then sort the topics, set topic weights, and obtain the final recommendations; corporate employees provide feedback on the final recommended topics.

5. The enterprise internal decision-making method based on multi-source data information aggregation according to claim 4 is characterized in that: The process of enterprise employees participating in topic prediction is implemented based on a decentralized betting contract or transaction contract based on blockchain technology. The betting contract includes an enterprise employee betting structure, a price calculation function, a balance check function, and a betting record function. The enterprise employee betting structure includes an enterprise employee identity record, the enterprise employee's selected topic, the topic prediction betting result option, the topic prediction betting quantity, and the time when the enterprise employee participated in the topic prediction betting; the transaction contract includes: an enterprise employee transaction structure, a price calculation function, a balance check function, and a transaction record function. The enterprise employee transaction structure includes an enterprise employee identity record, the enterprise employee's selected topic, the topic prediction transaction result option, the topic prediction transaction quantity, and the time when the enterprise employee participated in the topic prediction transaction. It indicates that enterprise employees buy and sell contracts through a bidding model. The bidding model refers to a trading method in which the contract price is determined by the quotes of the buyer and seller during the contract transaction process; When enterprise employees choose to participate in topic prediction by betting, the process of enterprise employees participating in topic prediction is as follows: S31: The betting contract is triggered after the company employee selects the topic prediction and topic prediction betting result options and determines the topic prediction betting amount; S32: The betting contract calls the balance check function to check whether the token or points balance in the enterprise employee account is sufficient to pay the bet amount. If so, execute step S33; otherwise, the balance is insufficient, the betting contract is terminated, the event prediction process ends, and the enterprise employee is prompted to recharge; S33: The betting contract calls the price calculation function to calculate the latest betting price of the betting result option based on the current betting situation of the predicted event; S34: After the latest betting price is determined, the contract deducts the corresponding tokens or points from the enterprise employee's account and calls the betting record function to record the enterprise employee's betting information in the blockchain ledger; at the same time, the betting contract updates the betting ratio and betting amount information of the predicted event in the market data; When enterprise employees choose to participate in topic forecasting by means of transactions, the process of enterprise employees participating in topic forecasting is as follows: S301: The transaction contract is triggered after the enterprise employee selects the topic prediction and topic prediction transaction result options and determines the topic prediction transaction quantity; S302: The transaction contract calls the balance check function to check whether the token or points balance in the enterprise employee's account is sufficient to pay the transaction amount. If so, step S33 is executed; otherwise, the balance is insufficient, the transaction contract is terminated, the event prediction process ends, and the enterprise employee is prompted to recharge; S303: The trading contract calls the price calculation function to calculate the final transaction price of the transaction result option based on the current predicted event transaction status. The transaction price is determined by matching the quotes of the buyer and seller. S304: After the final transaction price is determined, the transaction contract deducts the corresponding tokens or points from the enterprise employee's account and calls the transaction record function to record the enterprise employee's transaction information in the blockchain ledger. At the same time, the transaction contract updates the transaction ratio and transaction amount information of the predicted event in the market data. When settling accounts and distributing rewards to employees of companies that participated in the prediction, a decentralized settlement contract based on blockchain technology is implemented. The settlement contract includes: a result confirmation function, a valid bet or transaction share calculation function, a reward distribution function, and an account balance update function; The process of settling accounts and distributing rewards to employees who participated in the forecast is as follows: S41: The result of enterprise employees participating in event prediction triggers the settlement contract; S42: The settlement contract calls the valid bet or transaction share calculation function, traverses the enterprise employee bet or transaction information records in the blockchain ledger, filters out valid bets or transactions according to preset rules, and calculates the valid bet or transaction share for each result option; S43: Based on the calculated valid bet or transaction share and the total amount of tokens or points in the reward pool, the settlement contract calls the reward allocation function and calculates the reward amount that each participating enterprise employee should receive according to the set settlement algorithm; S44: The settlement contract calls the account balance update function to add the calculated reward amount to the account balance of the enterprise employee who placed the correct bet or transaction, and records the settlement process and results in detail in the blockchain ledger, including the event results, user bets or transactions, and reward distribution details.

6. The enterprise internal decision-making method based on multi-source information data aggregation according to claim 5 is characterized in that: The method also includes a governance process. When the balance of an enterprise employee's account meets the set conditions, the enterprise employee can participate in the formulation of internal market rules and the review of forecast issues. The governance process is based on the decentralized governance contract of blockchain technology. The governance contract includes a governance structure, a voting mechanism function, a proposal management function, and an authority control function. The process of enterprise employees participating in governance is as follows: When an enterprise employee submits a governance proposal, the governance contract calls the proposal management function, records the proposal content into the governance structure, and displays it for other enterprise employees to review and discuss; The governance contract calls the voting mechanism function and starts the voting process according to the set voting method. Employees vote on the proposal according to their own wishes, and the voting records are stored in the blockchain ledger. After the voting is completed, the governance decision is executed according to the rules determined by the proposal management function and the permission control function.

7. The enterprise internal decision-making method based on multi-source information data aggregation according to claim 4 is characterized in that: The rewards also adopt a multi-level incentive mechanism, which includes material rewards, career development and spiritual recognition. The material rewards include "bonuses and coupons", the career development includes "training opportunities and job promotions", and the spiritual recognition includes "senior level recognition and exclusive titles". The method also includes risk management and control, wherein the risk judgment preset criteria of the risk management and control are: abnormal betting or transaction amount, abnormal betting or transaction frequency, abnormal betting or transaction behavior pattern, and abnormal source and flow of funds; Preset risk response strategies for risks with different preset standards, including: betting or trading restrictions, investigation and warning strategies, and market intervention strategies.

8. The enterprise internal decision-making method based on multi-source information data aggregation according to any one of claims 1 to 7, characterized in that: Decentralized smart contracts based on blockchain technology use multiple dimensions to select blockchain platforms. These dimensions include transaction processing speed, throughput metrics, and: The prediction event processing capability indicator per unit time is used to measure the processing efficiency of the blockchain platform during operations such as the creation, betting, trading, and settlement of highly concurrent prediction events; Dynamic node expansion adaptability indicators are used to measure the stability of system performance, data consistency, and transaction processing capabilities during the dynamic increase or decrease of nodes on the blockchain platform; Smart contract vulnerability detection coverage indicator, used to evaluate the blockchain platform's ability to detect and prevent smart contract vulnerabilities; Data desensitization and authorized access granularity indicators are used to measure the blockchain platform's ability to desensitize sensitive information during data storage and transmission, as well as its flexibility in implementing fine-grained authorized access control; The smart contract development language and enterprise technology stack compatibility index evaluation indicator is used to measure the degree of integration between the smart contract development language supported by the blockchain platform and the enterprise's existing technology system; The blockchain platform introduces a node trust weighted consensus mechanism, assigns different voting weights according to the credibility of internal nodes of the enterprise, and adopts an optimized index structure for storage to improve data query efficiency. At the same time, an event tracing data storage model is designed to store the complete life cycle data of each predicted event in a chain structure.

9. An enterprise internal decision-making system based on multi-source data information aggregation, characterized by: The system is deployed in a distributed network environment and includes: A multi-source data information acquisition and processing module is used to acquire multi-source data information and pre-process the multi-source data information; the multi-source data information includes internal enterprise data information and external enterprise data information; The topic creation module is used to create topics, which are predicted events related to the internal and external strategies of the enterprise. The topic creation is realized by decentralized smart contracts based on blockchain technology; The topic recommendation module analyzes multi-source data information based on deep learning models and natural language processing technology, and generates matching recommended topics for enterprise employees based on the analysis results; The prediction participation module displays recommended topics to corporate employees so that they can participate in topic prediction. The prediction participation process is implemented based on the decentralized smart contract of blockchain technology. The prediction reward distribution module is used to determine whether the prediction results of the topics predicted by enterprise employees meet the results set within the enterprise. If so, the decentralized smart contract based on blockchain technology settles and distributes rewards to the employees participating in the prediction, and assists the internal decision-making of the enterprise based on the prediction results of the topics predicted by the employees. Otherwise, internal research and judgment will be conducted.

10. An enterprise internal decision-making platform architecture based on multi-source data information aggregation, characterized by: The platform architecture adopts a layered design, including: data acquisition and pre-processing layer, intelligent engine layer, predictive betting transaction rules business logic layer, enterprise employee interface layer and data storage layer; The data acquisition and preprocessing layer is used to deploy a multi-source data information acquisition and processing module, the intelligent engine layer is used to deploy a topic creation module and a topic recommendation module, the predicted betting and transaction rules business logic layer is used to deploy a prediction participation module and a prediction reward distribution module; the enterprise employee interface layer is used to provide a friendly interactive interface for enterprise employees; the data storage layer is used to save predicted betting or transaction records, enterprise employee information and other data. The data output by the multi-source data information acquisition and processing module flows through the data acquisition and processing layer and enters the intelligent engine layer. The intelligent engine layer includes deep learning models, natural language processing components and intelligent event generators, and is responsible for topic generation and topic recommendation; The business logic layer of the predicted betting transaction rules is equipped with a centralized smart contract platform, which is responsible for predicting transaction management, rule formulation and market monitoring; the transaction cycle on the decentralized smart contract platform includes: initiating contract creation, broadcasting, verification, storage, execution and settlement.

Images