Behavior data evaluation system and method based on hierarchical architecture, and program product

Abstract

The invention provides a behavior data evaluation system based on a hierarchical architecture, and the system comprises a data collection module which is used for collecting the behavior data and business contribution data of business personnel in real time; the data processing module is used for cleaning, integrating and analyzing the behavior data and the business contribution data based on a big data hierarchical architecture; the evaluation index calculation module is used for obtaining an evaluation index according to the cleaned, integrated and analyzed data and generating an evaluation result; and the feedback module is used for feeding back the evaluation index, the evaluation result and the suggested learning path to the business personnel in real time. According to the invention, comprehensive and accurate real-time evaluation of the behavior data of the business personnel can be realized, and the efficiency and accuracy of data processing are improved. Through flexible evaluation index calculation and a personalized feedback mechanism, the system not only can effectively promote learning and development of business personnel, but also can provide a scientific decision basis for a management layer.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing and behavior evaluation, and in particular to a behavior data evaluation system, method and program product based on a layered architecture. Background Art

[0002] With the development of information technology and big data technology, more and more companies and organizations have introduced digital platforms in employee training and performance management. These platforms can record business personnel's learning behavior data (such as course participation, test scores, etc.) and business contribution data (such as project participation, sales performance, etc.), and process and evaluate these data through relevant analysis technologies to achieve quantitative assessment of business personnel's performance. Existing systems are often limited to a single data source or data processing method, lack of deep integration and analysis of multi-dimensional behavioral data, and cannot fully reflect the performance of business personnel at different levels. At the same time, due to the singleness of the data processing architecture, the existing system cannot flexibly respond to the access and analysis needs of multi-source data, resulting in delayed evaluation results and failure to meet the requirements of real-time feedback and dynamic adjustment.

[0003] In addition, existing performance evaluation systems usually only provide basic performance scores, without fully considering the relationship between learning behavior and business contribution, and without combining personalized suggestions with learning path planning, making it difficult to provide business personnel with comprehensive and targeted feedback and improvement suggestions. The incentive mechanisms in existing systems are also mostly based on fixed rules, lacking the ability to dynamically adjust according to changes in business needs, and unable to flexibly adapt to the ever-changing business environment and training goals.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] In response to the problems in the prior art, the present invention aims to provide a big data behavior evaluation system based on a layered architecture, which can conduct a comprehensive evaluation of the learning behavior and business contribution data of business personnel through multi-source data access, cleaning, integration and in-depth analysis, and generate personalized evaluation results and feedback suggestions based on actual business needs, so as to further improve the learning efficiency and business performance of business personnel.

[0006] An embodiment of the present invention provides a behavior data evaluation system based on a layered architecture, characterized by comprising:

[0007] Data collection module, used to collect business personnel's behavior data and business contribution data in real time;

[0008] The data processing module cleans, integrates and analyzes the behavior data and business contribution data of the behavior data evaluation system based on the layered architecture based on the big data layered architecture;

[0009] An evaluation index calculation module is used to obtain evaluation indicators based on the data cleaned, integrated and analyzed by the behavior data evaluation system based on the hierarchical architecture, and generate evaluation results;

[0010] The feedback module is used to provide real-time feedback to business personnel on their evaluation indicators, evaluation results, and recommended learning paths.

[0011] In some optional embodiments, the big data hierarchical architecture includes:

[0012] The enterprise data warehouse layer (EDW) is used to centrally store data from different data sources;

[0013] The Common Data Mart layer (CDM) is used to integrate and standardize the data stored in the enterprise data warehouse layer of the behavioral data evaluation system based on the layered architecture;

[0014] The application data mart layer (ADM) is used to summarize and analyze the data integrated and standardized by the public data mart layer according to specific business needs.

[0015] In some optional embodiments, the data processing module further includes:

[0016] Data cleaning unit, used to detect outliers and fill missing values ​​in raw data;

[0017] A data integration unit, used to match and merge data from different data sources of a behavior data evaluation system based on a hierarchical architecture;

[0018] The data analysis unit is used to apply statistical analysis and machine learning algorithms to conduct in-depth analysis of the data integrated and standardized by the behavioral data evaluation system based on a hierarchical architecture.

[0019] In some optional embodiments, the statistical analysis method includes one or more of regression analysis, descriptive statistical analysis and variance analysis, and the machine learning algorithm includes one or more of decision tree, random forest and support vector machine.

[0020] In some optional embodiments, the evaluation index calculation module includes:

[0021] A multi-dimensional evaluation index unit, used to calculate evaluation indexes based on basic factors and bonus factors recorded in the enterprise data warehouse layer of the behavior data evaluation system based on a hierarchical architecture;

[0022] The dynamic adjustment unit is used to adjust the calculation rules of evaluation indicators according to the behavioral performance of business personnel and changes in business needs.

[0023] In some optional embodiments, the evaluation index calculation unit further includes:

[0024] A weight setting unit for adjusting the calculation weights for the evaluation indicators according to factors of different factors recorded in the enterprise data warehouse layer of the behavior data evaluation system based on the hierarchical architecture;

[0025] The cumulative evaluation index unit is used to generate the cumulative evaluation index of the business personnel according to their business contribution data and generate historical records.

[0026] In some optional embodiments, the data acquisition module further includes:

[0027] Multi-source data access unit, used to collect data from online learning platforms, training management systems and business management systems;

[0028] The data preprocessing unit is used to perform preliminary cleaning and format conversion on the collected data.

[0029] In some optional embodiments, the feedback module further includes:

[0030] The notification unit is used to notify business personnel of their evaluation indicators, evaluation results and recommended learning paths through one or more methods including email, text message or mobile application.

[0031] In some optional embodiments, the feedback module further includes:

[0032] User behavior monitoring module, used to track the learning behavior and participation of business personnel in real time;

[0033] The personalized suggestion module generates customized learning paths and improvement suggestions based on the business personnel's evaluation indicators.

[0034] Another aspect of the present invention further provides a behavior data evaluation method based on a hierarchical architecture, comprising the following steps:

[0035] Collect business personnel's learning behavior data and business contribution data in real time;

[0036] Clean, integrate and analyze learning behavior data and business contribution data;

[0037] Calculate the integral using the cleaned, integrated and analyzed data and generate integral evaluation results;

[0038] Provide feedback to business personnel on their points status and recommended learning paths.

[0039] Another aspect of the present invention also provides a behavior data evaluation program product based on a hierarchical architecture. The program product includes computer instructions. When the computer instructions are executed by a processor, the functions of the above-mentioned behavior data evaluation system based on a hierarchical architecture are realized.

[0040] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure.

[0041] The behavior data evaluation system, method and program product based on the layered architecture of the present invention have the following beneficial effects:

[0042] The present invention can achieve a comprehensive and accurate real-time evaluation of business personnel behavior data, and improve the efficiency and accuracy of data processing. Through flexible evaluation index calculation and personalized feedback mechanism, the system can not only effectively promote the learning and development of business personnel, but also provide a scientific decision-making basis for management. In addition, the system design is highly adaptable and can be flexibly adjusted according to the actual needs of the enterprise and market changes, ensuring the continued effectiveness and foresight of performance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Other features, objectives and advantages of the present invention will become more apparent from a reading of the detailed description of non-limiting embodiments made with reference to the following accompanying drawings.

[0044] Figure 1 is a schematic diagram of a behavior data evaluation system based on a layered architecture according to an embodiment of the present invention;

[0045] Figure 2 It is a flow chart of a behavior data evaluation method based on a layered architecture according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0047] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0048] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the steps. For example, some steps may be decomposed, while some steps may be combined or partially combined. Therefore, the actual execution order may be changed according to the actual situation.

[0049] The present invention provides a behavior data evaluation system based on a hierarchical architecture, including a data acquisition module, a data processing module, an evaluation index calculation module and a feedback module. The data acquisition module is used to collect the behavior data and business contribution data of business personnel in real time. Behavior data refers to the performance information of business personnel in the learning and training process, such as course participation, test scores, etc.; business contribution data refers to the performance of business personnel in actual work, such as project participation, sales performance, etc. This module realizes the real-time collection of multi-source data, laying a foundation for subsequent data processing and analysis. The data processing module is based on the big data hierarchical data processing architecture, and is responsible for cleaning, integrating and analyzing the collected behavior data and business contribution data. The hierarchical data processing architecture classifies and processes the data according to different processing stages to ensure the consistency and availability of the data. Through this architecture, the system can efficiently process data from multiple sources, ensure the quality of the data, and provide reliable data support for the subsequent evaluation index calculation. The evaluation index calculation module is not only used to generate evaluation indicators based on the business personnel behavior data and business contribution data after data processing, but also can include the evaluation of the lecturer. Through the learning behavior, training results, and feedback on courses and lecturers of business personnel, the system can comprehensively evaluate the teaching effect and teaching quality of lecturers. The evaluation indicator calculation module conducts multi-dimensional analysis on the data of business personnel and lecturers according to different business needs and rules, so as to obtain evaluation indicators that reflect their overall performance. The evaluation rules and indicators can be flexibly configured according to the actual business scenario. The feedback module is used to provide real-time feedback on evaluation results and recommended learning paths to business personnel and relevant managers. In addition to providing feedback on evaluation indicators, evaluation results, and learning path recommendations for business personnel, the feedback module can also provide feedback on the evaluation results of lecturers to relevant personnel, such as training supervisors or teaching management, to help optimize training content and teaching quality. Feedback can be sent via email, SMS, or mobile applications to help business personnel understand their current performance and provide personalized learning path recommendations based on the evaluation results, so that they can improve their skills in a targeted manner, while providing managers with a basis for improving training and teaching. The various modules of the present invention cooperate with each other. The data provided by the data acquisition module is cleaned and analyzed by the data processing module, and then passed to the evaluation index calculation module for calculation. Finally, the feedback module feeds back the evaluation results and learning suggestions to the business personnel and management in real time, realizing a complete behavior data and lecturer evaluation process. Through the behavior data evaluation system based on the hierarchical architecture, it is possible to effectively realize the multi-dimensional processing and evaluation of business personnel behavior data, business contribution data and lecturer teaching quality, ensure the accuracy of data processing and the timeliness of feedback, provide personalized learning improvement suggestions for business personnel, and provide targeted support for the optimization of training and teaching content for the management.

[0050] like Figure 1As shown, the embodiment of the present invention provides a behavior data evaluation system based on a layered architecture, including a data collection module M100, a data processing module M200, an evaluation index calculation module M300 and a feedback module M400. The design and functions of these modules are intended to effectively collect, process and evaluate the learning behavior data and business contribution data of business personnel, thereby generating actionable evaluation results and providing personalized feedback.

[0051] The main function of the data collection module M100 is to collect the behavior data and business contribution data of business personnel in real time. Data is obtained in real time from multiple data sources (such as online learning platforms, training management systems, business management systems, etc.) through multi-source data access units. The collected behavior data includes the participation of business personnel in learning activities, test scores, etc., while the business contribution data includes project completion, sales performance, etc. The data collection module M100 can use API interfaces or real-time data streams (such as Apache Kafka) to integrate data from different sources into the system to ensure the timeliness and accuracy of the data.

[0052] The data processing module M200 cleans, integrates and analyzes the collected behavioral data and business contribution data based on the big data layered architecture. The data processing module M200 contains multiple sub-units, among which the data cleaning unit is responsible for processing outliers, missing data, and ensuring data quality; the data integration unit matches and merges data from different data sources to ensure consistent data formats; the data analysis unit applies statistical analysis and machine learning algorithms to conduct in-depth analysis of the integrated data to identify potential trends and patterns. The cleaning unit can use tools (such as Talend) for data processing, the integration unit ensures the consistency of multi-source data through field mapping and merging strategies, and the analysis unit can use big data processing frameworks (such as Apache Spark) for complex data analysis.

[0053] The evaluation indicator calculation module M300 generates evaluation indicators and evaluation results for business personnel based on the cleaned, integrated and analyzed data. The evaluation indicator calculation module M300 combines different evaluation dimensions (such as learning participation and business contribution) to form a comprehensive scoring system. Specifically, the module can calculate evaluation indicators through a multi-dimensional evaluation indicator unit, and use a dynamic adjustment unit to dynamically modify the evaluation rules based on changes in business personnel's performance and adjustments to business needs. The evaluation indicator calculation module M300 can set a variety of calculation rules (such as weighted average) and allow managers to adjust the parameters of the evaluation rules through a graphical user interface (GUI).

[0054] The feedback module M400 is used to provide real-time feedback to business personnel on their evaluation indicators, evaluation results, and recommended learning paths. This module ensures that business personnel can understand their performance in a timely manner and receive personalized learning suggestions. Feedback can be sent through multiple channels (such as email, SMS, or mobile applications) to help business personnel improve their learning and work performance in a targeted manner. The feedback module M400 integrates notification services (such as Twilio and SendGrid) to generate personalized feedback content based on the evaluation results and send it to business personnel through the corresponding notification method.

[0055] Through the close cooperation of the data acquisition module M100, data processing module M200, evaluation index calculation module M300 and feedback module M400 of the behavior data evaluation system based on the layered architecture, the system can efficiently collect, process and analyze the behavior data and business contribution data of business personnel. The data acquisition module M100 provides real-time data input, the data processing module M200 ensures the quality and consistency of the data, the evaluation index calculation module M300 generates evaluation results based on multi-dimensional data, and the feedback module M400 promptly communicates these results to the business personnel. The overall architecture of the invention ensures the real-time, accuracy and personalization of the behavior data evaluation, thereby effectively supporting the learning and performance improvement of business personnel. This embodiment is some specific embodiments of the present invention for the data acquisition module M100, the data processing module M200, the evaluation index calculation module M300 and the feedback module M400. In practical applications, the data collection and processing methods can be flexibly adjusted according to different business needs and technical platforms. The calculation and feedback methods of the evaluation indicators can also use a variety of different implementation methods. The invention is not limited to a specific technical solution and can be expanded and adjusted according to specific circumstances.

[0056] In some embodiments, the big data layered architecture of the present invention includes an enterprise data warehouse layer (EDW), a common data mart layer (CDM) and an application data mart layer (ADM).

[0057] The enterprise data warehouse layer (EDW) is used to centrally store raw data from different data sources. Since the behavior data and business contribution data of business personnel may come from different systems (such as learning management systems, performance management systems, etc.), EDW provides a unified storage space for these data and ensures the integrity and consistency of the data. Through this layer, all relevant data can be centrally managed, providing a basis for subsequent data processing and analysis. The implementation of the EDW layer can build a centralized data storage system based on a big data platform (such as Hadoop HDFS, Amazon S3). The enterprise data warehouse layer (EDW) is responsible for collecting the behavior data and business contribution data of business personnel from multiple different data sources (such as learning management systems, CRM, ERP, financial management systems, etc.). In order to ensure the efficiency and real-time nature of data collection, data extraction, transformation, loading (ETL) tools or streaming data processing systems (such as Apache Kafka, Apache Flume) can be used to receive data from different data sources. Each data source stores the collected data in EDW through regular batch processing or real-time data streaming. The data is stored in the original format to maintain the integrity of the data for subsequent processing.

[0058] The common data mart layer (CDM) is used to integrate and standardize the data stored in the enterprise data warehouse layer. Because the data formats from different sources may be inconsistent, the CDM layer integrates data of different formats and structures into a unified standard format through cleaning, conversion and standardization. This layer ensures that the data is consistent and operational before entering the analysis phase, thereby providing high-quality input data for data processing and evaluation of different modules in the system. The technical implementation of the CDM layer focuses on cleaning, integrating and standardizing the data in the enterprise data warehouse. Because the business data from different data sources may have different formats and duplicate or inconsistent data, at this layer, data cleaning tools (such as Trifacta and Talend) are used to deduplicate the data and standardize the format. During the standardization process, predefined data specifications (such as ISO and internal enterprise standards) can be used to ensure that all data fields are processed consistently. Integrate data from various sources into a consistent table structure based on the data dictionary and metadata management system.

[0059] The application data mart layer (ADM) performs further aggregation and analysis based on the integrated and standardized data. The ADM layer performs targeted analysis based on specific business needs, such as the evaluation of business personnel and the evaluation of lecturers. Through this layer, the system can filter, calculate and aggregate the standardized data at the application level to generate evaluation results that meet business needs. The ADM layer is used to further aggregate and deeply analyze the standardized data to meet specific business needs. The ADM layer can perform data calculation and aggregation based on data analysis tools (such as Apache Spark, Hive, and Impala). The implementation of this layer focuses on customized data processing and result generation based on business rules or specific needs (such as the evaluation of business personnel performance and the evaluation of lecturer teaching effectiveness). The data is grouped and aggregated according to business needs at the ADM layer (such as grouping and analyzing the performance of business personnel by region and department), and then the data is calculated using the preset analysis model. For example, when evaluating the participation and lecturer performance of a course, historical data can be used for regression analysis to obtain specific evaluation results. The system can integrate multi-dimensional data analysis models (such as OLAP and machine learning algorithms) to achieve complex analysis under specific business needs, such as evaluating the learning effects of business personnel through classification algorithms, or identifying outstanding teams or individuals through clustering algorithms.

[0060] In an embodiment of the present invention, the hierarchical architecture realizes efficient management of data through multiple levels of processing steps. The enterprise data warehouse layer ensures the centralization and integrity of data, the public data mart layer solves the heterogeneity problem of multi-source data through data cleaning and standardization, and the application data mart layer performs customized analysis and processing according to actual business needs. The synergy between the layers enables the entire big data hierarchical architecture to efficiently cope with complex data processing requirements and provide strong support for subsequent evaluation index calculation and feedback. The enterprise data warehouse layer, public data mart layer and application data mart layer described in the present invention are only an example of implementing a hierarchical data processing architecture, and other hierarchical architectures can also be applied to the present invention. The division of specific levels, data processing methods and architecture implementations can be adjusted according to actual needs and specific business scenarios. The embodiments of the present invention are not limited to the specific technical details described. Through the big data hierarchical architecture of this embodiment, efficient processing and analysis of business personnel behavior data and business contribution data can be achieved, ensuring the consistency and integrity of data, and being able to flexibly meet business needs, thereby providing strong support for the calculation and feedback of evaluation indicators.

[0061] In some embodiments, the data processing module M200 of the present invention further includes a data cleaning unit, a data integration unit and a data analysis unit, aiming to ensure that data collected from multiple sources can be efficiently cleaned, integrated and analyzed, thereby providing high-quality data support for subsequent evaluation index calculation and feedback.

[0062] The data cleaning unit is used to detect outliers and fill missing values ​​in the raw data. In practical applications, the raw data may contain noise, outliers, and incomplete data records. Through outlier detection technology, for example, using statistical methods such as box plots or Z-score methods, data that deviate from the normal range can be identified and processed. For missing data, a variety of filling strategies can be used, such as mean filling, nearest neighbor filling (KNN) or prediction filling based on regression analysis. The role of the data cleaning unit is to ensure that all data are standardized before entering the subsequent processing stage and have sufficient completeness and accuracy.

[0063] The data integration unit is used to match and merge data from different data sources. Since the data comes from multiple heterogeneous systems (such as learning management systems, business management systems, ERP, etc.), their data structures and field naming may be different, and even the same type of data may have naming conflicts or inconsistencies in different systems. Through the data integration unit, the system can unify multi-source data into a consistent format and merge them according to predefined rules (for example, field mapping or matching rules based on unique identifiers). The working principle of this unit includes field mapping, data integration, and data deduplication to ensure that data from different data sources are consistent in structure and content, providing a unified data view for subsequent analysis.

[0064] The data analysis unit is used to conduct in-depth analysis of the integrated and standardized data. Statistical analysis methods include one or more of regression analysis, descriptive statistics analysis, and variance analysis. The data analysis unit can select appropriate analysis algorithms according to different business needs to help identify the learning behavior patterns of business personnel or trends in business contributions. Machine learning algorithms include one or more of decision trees, random forests, and support vector machines. This unit plays a role in converting basic data into valuable business decision information in the architecture. It can effectively analyze the behavioral data of business personnel, explore potential performance patterns, and provide solid data support for the calculation of evaluation indicators. This multi-level data analysis strategy ensures the comprehensiveness and accuracy of the evaluation process, enabling enterprises to make more informed training and management decisions based on data.

[0065] Regression analysis is used to establish a relationship model between business personnel's behavior and their business contribution. Linear regression or multiple regression analysis can be used to predict the future performance of business personnel. Through regression analysis of historical data, key factors affecting business performance can be identified. Data analysis tools (such as R and Python's statsmodels library) are used to train the model, optimize parameters through the least squares method, and evaluate the R 2 The p-value and p-value were used to verify the validity of the model.

[0066] Descriptive statistical analysis is used to aggregate and summarize the behavior data of business personnel to help identify overall trends and patterns. This includes the calculation of statistical quantities such as mean, standard deviation, and median, so that management can have a preliminary understanding of the learning participation and performance of business personnel. Through data processing tools (such as Pandas), statistical calculations are performed on behavioral data and business contribution data to generate reports and visualization charts (such as bar charts and box plots).

[0067] Variance analysis is used to evaluate the impact of different courses or training programs on the performance of business personnel. By comparing the means of different groups, it is determined whether there are significant differences, thereby providing a basis for training improvements. The ANOVA (Analysis of Variance) method is used in combination with Python's scipy library to compare groups, and the F value and P value are output to evaluate the significance of the difference.

[0068] Decision trees are used to build a classification model for business personnel performance. The model can predict the performance category (such as excellent, qualified, and unqualified) of business personnel based on input behavioral data (such as learning participation and test scores). Use decision tree algorithms (such as CART) for model training, build and train models through the Scikit-learn library, use cross-validation to evaluate the accuracy of the model, and display the decision tree structure through visualization tools (such as Graphviz).

[0069] Random forest, as a method of ensemble learning, improves accuracy by building multiple decision trees and aggregating their prediction results. This method can reduce overfitting and enhance the robustness of the model. Use Scikit-learn's RandomForestClassifier, set appropriate number of trees and depth parameters for training, and evaluate model performance through OOB (out-of-bag estimation).

[0070] Support vector machines (SVMs) are used to classify business personnel performance and are particularly suitable for processing high-dimensional data. SVMs distinguish different categories by constructing the best segmentation hyperplane. Using the SVC class in the Scikit-learn library, select a suitable kernel (such as a linear kernel or an RBF kernel), train the model, and optimize the hyperparameters through grid search (GridSearchCV) to improve classification accuracy.

[0071] Through the cooperation of the data cleaning unit, the data integration unit and the data analysis unit, the data processing module M200 can ensure that the raw data obtained from the data acquisition module M100 is fully processed and analyzed, and provide high-quality and accurate data input for the subsequent evaluation index calculation module M300. The mutual cooperation of the data cleaning unit, the data integration unit and the data analysis unit ensures the complete process of data from the original state to the deep analysis, reducing the possibility of errors and data loss. The data cleaning unit, the data integration unit and the data analysis unit are specific implementations in some embodiments of the present invention, but the present invention is not limited to this. The technical implementation of data cleaning, integration and analysis can be adjusted according to specific business scenarios, technical platforms and data processing requirements. For example, different outlier detection technologies, data matching methods and analysis algorithms can be used, or processing steps can be increased or decreased according to specific application needs. The present invention is not limited to any specific technical implementation or platform, and can be flexibly adjusted without deviating from the essence of the invention. Through this embodiment, multi-source data can be efficiently processed to ensure that data maintains high quality and consistency at each stage of cleaning, integration and analysis, thereby providing accurate and reliable data support for evaluation index calculation and personalized feedback.

[0072] In some embodiments, the evaluation index calculation module M300 includes a multi-dimensional evaluation index unit and a dynamic adjustment unit. The design and function of these units are intended to ensure that the evaluation index calculation process of the business personnel can reflect the performance of multiple aspects, and the evaluation rules can be flexibly adjusted according to the changes in actual business needs. The multi-dimensional evaluation index unit is used to calculate the evaluation index according to a variety of factors, including the basic item factors and bonus item factors recorded in the enterprise data warehouse layer (EDW). The basic item factors are usually the basic tasks and activities completed by business personnel in their daily work, such as course participation, project task completion, etc. The bonus item factors refer to the additional performance or contribution beyond the basic items, such as particularly excellent test scores, innovative solutions in projects, etc. By integrating these different factors into the evaluation system, the system can evaluate the overall performance of business personnel more comprehensively and three-dimensionally. In specific implementation, the system assigns different weight values ​​to factors of different dimensions through weight settings, and calculates based on these weight values. For example, the weight of the basic item at a certain stage may account for 60%, while the bonus item accounts for 40%. In some optional embodiments, the unit can also support multiple evaluation models, and the system can select different evaluation methods according to business needs. For example, a simple scoring model based on weighted average can be used, or a more complex evaluation can be performed in combination with machine learning algorithms such as decision trees and random forests. In addition, the unit can process evaluation factors from different data sources and integrate the performance of business personnel on different platforms (such as training participation and work contribution) into a unified evaluation result. The function of the dynamic adjustment unit is to dynamically adjust the calculation rules of the evaluation indicators according to the behavioral performance of business personnel and changes in business needs. In actual applications, the work tasks and performance of business personnel may vary due to changes in business goals, market environment or company strategy. In order to adapt to these changes, the dynamic adjustment unit allows the weight or calculation method of the evaluation indicators to be adjusted according to real-time data or preset rules. For example, when an enterprise adds a new business category, in order to ensure that business personnel master the corresponding knowledge of the new business category as soon as possible, the weight of the relevant course contribution may be increased, while in the training stage, the weight of learning participation is increased accordingly.

[0073] The evaluation index calculation module M300 includes a multi-dimensional evaluation index unit and a dynamic adjustment unit. The specific technical implementation can be achieved in the following ways:

[0074] Real-time monitoring and feedback: The system can dynamically adjust the parameters of the evaluation model according to the set thresholds or business rules by monitoring the real-time behavioral data of business personnel (such as course participation and sales performance).

[0075] The dynamic adjustment unit can automatically apply preset business rules and adjustment strategies by integrating rule engines (such as Drools). Managers can also manually set or modify evaluation rules through the system management interface.

[0076] The adaptive learning mechanism system can also learn the behavioral patterns of business personnel by analyzing historical data and predict future performance based on these patterns to adjust weights and evaluation rules.

[0077] Through the combination of the multidimensional evaluation index unit and the dynamic adjustment unit, the system can flexibly adapt to changes in the business environment, ensuring that the evaluation results are both comprehensive, real-time and flexible. The mutual cooperation of these units enables the evaluation process to be adjusted at any time according to the needs of the enterprise to reflect the actual performance of the business personnel. The multidimensional evaluation index unit and the dynamic adjustment unit are only some embodiments of the present invention. The specific implementation of the present invention is not limited to a specific calculation method or weight setting, nor is it limited to a specific rule adjustment method. In practical applications, the calculation method, dimension selection and dynamic adjustment strategy of the evaluation index can be flexibly adjusted according to specific business scenarios, industry needs and data characteristics. The present invention allows free configuration and expansion of evaluation models and rules, and the specific implementation can be customized according to changes in demand. Through the multidimensional evaluation index unit and the dynamic adjustment unit of this embodiment, the system can perform a comprehensive, flexible and real-time evaluation of the performance of business personnel, ensuring that the evaluation results can not only reflect the current performance of business personnel, but also can be dynamically adjusted according to changes in business needs, providing managers with more accurate decision-making basis.

[0078] In some embodiments, the evaluation index calculation unit also includes a weight setting unit and a cumulative evaluation index unit. These units are designed to enable the system to flexibly adjust the weights of various factors in the evaluation process according to different business needs, and to perform cumulative evaluation and historical record generation on the long-term performance of business personnel.

[0079] The weight setting unit is used to adjust the calculation weight of the evaluation index according to the different factors recorded in the enterprise data warehouse layer. The enterprise data warehouse layer (EDW) stores the behavior data and business contribution data of business personnel, which may involve multiple dimensions, such as learning participation, project completion, sales performance, etc. Each dimension may have different effects on the evaluation results. The weight setting unit ensures that the evaluation model can reflect the different effects of different factors on the overall performance of business personnel by assigning different weight values ​​to each dimension or factor. For example, at a certain stage, learning behavior may be the main assessment factor, and its weight can be set to 60%, while business contribution accounts for 40%; at another stage, when the business goal changes, the weight of business contribution can be increased to 70%. Through the weight setting unit, the system can flexibly respond to the strategic adjustment of the enterprise and adjust the calculation rules of the evaluation index in real time. In specific implementation, the weight setting unit can be configured through the system administrator interface. The administrator can adjust the weight value of each factor according to actual needs, or set a predefined weight configuration scheme. In addition, the unit can also dynamically adjust the weight according to the analysis results of historical data or the rule engine. The data analysis module can help identify which factors have the greatest impact on business performance in a specific stage or task, thereby automatically adjusting the weights of these factors.

[0080] The cumulative evaluation indicator unit is used to generate the cumulative evaluation indicators and historical records based on the business contribution data of the business personnel. The role of this unit is to track and accumulate the performance of business personnel in the long term, rather than just a single evaluation based on the current performance. For example, the system will generate corresponding evaluation indicators based on each contribution of the business personnel (such as completing a project, achieving a sales target), and accumulate these evaluation results to form a long-term performance evaluation. At the same time, the cumulative evaluation indicator unit can also record the historical performance of each business personnel and generate historical records for long-term trend analysis or annual performance evaluation. This can not only help companies to evaluate employees more objectively, but also provide managers with detailed historical data to help them make long-term incentive decisions or promotion decisions. In specific implementation, the cumulative evaluation indicator unit can record each evaluation result through a database table and accumulate this data in a time series manner. The system can also display the historical evaluation results of business personnel through a graphical interface, and managers can clearly see the performance changes of business personnel in different time periods. The implementation of this unit can rely on a relational database (such as MySQL, PostgreSQL) or a NoSQL database (such as MongoDB), and regularly generate updated evaluation results and historical records through stored procedures or scripts.

[0081] Through the combination of the weight setting unit and the cumulative evaluation index unit, the system can not only flexibly adjust the evaluation weight according to the business needs of the enterprise, but also continuously track and evaluate the long-term performance of the business personnel, thereby realizing flexibility and long-term combination in the calculation of the evaluation index. The weight setting unit and the cumulative evaluation index unit are only some embodiments of the present invention. In practical applications, the weight setting method can be customized according to specific business scenarios and technical requirements, and is not limited to fixed weight value settings. Similarly, the implementation of the cumulative evaluation index unit can also use different data storage and processing technologies, such as distributed databases or memory databases. The present invention allows the implementation of these units to be adjusted and expanded without deviating from the core invention ideas to meet different enterprise needs and business goals. Through the weight setting unit and the cumulative evaluation index unit, the present invention can flexibly adjust the weights of each factor during the evaluation process, ensure that the evaluation results can reflect the changes in business needs in real time, and support the evaluation of the long-term performance of business personnel and the accumulation of historical records, providing long-term data support and decision-making basis for the performance management of the enterprise.

[0082] In some embodiments, the data acquisition module M100 further includes a multi-source data access unit and a data preprocessing unit. The functional design of these units is to solve the problems of data acquisition, cleaning and conversion from different systems, thereby providing standardized and high-quality data input for subsequent data processing and analysis.

[0083] The multi-source data access unit is used to collect data from multiple data sources (such as online learning platforms, training management systems, and business management systems). Since the behavior data and business contribution data of business personnel may be stored in different systems, for example, the online learning platform saves course participation and learning progress information, while the business management system records project participation, performance and other data. In order to enable the system to conduct a comprehensive evaluation, the data collection module M100 needs to integrate and access these data sources. Through the multi-source data access unit, the system can receive data input from different sources and aggregate it into a central storage or enterprise data warehouse. This unit can achieve data collection through customized APIs, database connectors, or file interfaces (such as CSV, JSON), which may include delivering data through batch processing (such as periodic import) or real-time stream processing (such as using Apache Kafka or Flink).

[0084] The role of the data preprocessing unit is to perform preliminary cleaning and format conversion on the collected data. After the data enters the central system from multiple source systems, the preprocessing unit needs to standardize the data because there may be differences in data format, encoding or structure (for example, different date formats or field names in different systems). The preliminary cleaning process includes removing invalid data, handling outliers, and dealing with missing data, while format conversion ensures that the format of all data (such as time format, numerical unit) is consistent so that subsequent processing modules can correctly read and process the data. The specific technical implementation can use ETL tools (such as Apache NiFi, Talend) or write data conversion scripts to standardize the data using regular expressions and data mapping tables.

[0085] In an embodiment of the present invention, a multi-source data access unit and a data preprocessing unit work together to ensure that the system can efficiently collect, clean and format data, providing a basis for subsequent data analysis and evaluation. The multi-source data access unit ensures the system's compatibility with multiple data sources and the extensiveness of data collection, while the data preprocessing unit ensures that the data is consistent and operable when entering the enterprise data warehouse or other processing units by preliminarily cleaning and standardizing the data. The multi-source data access unit and the data preprocessing unit are for some specific embodiments of the present invention. In practical applications, the data access and preprocessing methods can be adjusted according to specific business scenarios, technical platforms and data source characteristics. The multi-source data access unit can be implemented through a variety of protocols, file formats or real-time stream processing tools, and the data preprocessing unit can also use different data cleaning and format conversion technologies to meet specific needs. The present invention is not limited to a specific implementation method, and can be flexibly adjusted and expanded without deviating from the core idea of ​​the invention. Through the multi-source data access unit and data preprocessing unit in the data acquisition module M100, the present invention can effectively solve the problem of data acquisition and preliminary processing from multiple data sources, ensuring that data from different sources can be processed consistently before entering subsequent processing modules, laying the foundation for the overall data processing and analysis of the system.

[0086] In some embodiments, the feedback module M400 further includes a user behavior monitoring module, a personalized suggestion module, and a notification unit. These units work together to ensure that the system can monitor the behavior of business personnel in real time, generate personalized learning suggestions, and notify business personnel of relevant evaluation results in a timely manner, forming a closed-loop feedback mechanism.

[0087] The function of the user behavior monitoring module is to track the learning behavior and participation of business personnel in real time. Specifically, this module can obtain the learning activity data of business personnel from the online learning platform or training management system, such as course viewing time, chapter completion, test scores, etc. By monitoring these behavioral data in real time, the system can accurately record the learning progress of business personnel and conduct in-depth analysis of their learning habits and participation enthusiasm. The implementation of this module can rely on data stream processing tools (such as Apache Kafka, Flink), which can capture the activities of business personnel on the learning platform in real time and transmit the data to the central processing system. The behavioral data of business personnel is collected in real time through the user behavior monitoring module, providing a real-time and accurate data basis for subsequent evaluation and personalized recommendations.

[0088] The personalized suggestion module generates customized learning paths and improvement suggestions based on the evaluation indicators of business personnel. The core function of this module is to generate personalized learning paths for each business person based on evaluation indicators (such as learning participation, test scores, business contributions, etc.). This process involves the system analyzing the current performance of business personnel, combining their historical data, and providing suggestions to further improve learning results. For example, if a business person's test scores are low, the system may recommend that they re-study related courses or participate in additional training. The personalized suggestion module can generate recommended content through machine learning algorithms (such as collaborative filtering and content recommendation), so that each business person can obtain tailored learning path suggestions. The implementation of this module can integrate existing recommendation system algorithms, combined with historical data and real-time monitoring data, to ensure that the recommended content can truly match the actual needs and learning goals of business personnel.

[0089] The notification unit is responsible for notifying business personnel of their evaluation indicators, evaluation results, and recommended learning paths through one or more methods including email, SMS, or mobile applications. This unit ensures that the evaluation results and learning suggestions are communicated to business personnel in a timely manner so that they can understand their own performance and take corresponding learning or improvement measures. The notification unit can be implemented by integrating third-party notification services (such as Twilio and SendGrid) to send emails and SMS, or by sending mobile application notifications through the company's internal message push system. This module can select the most suitable notification method based on the business personnel's preferences to ensure that feedback information can be communicated efficiently and conveniently.

[0090] Through the close collaboration of the user behavior monitoring module, personalized suggestion module and notification unit, the feedback module M400 can achieve real-time tracking of business personnel's behavior, generate accurate learning suggestions, and provide timely feedback to business personnel, ensuring that the system's feedback mechanism can cover the entire process from behavior collection to suggestion delivery.

[0091] The user behavior monitoring module, the personalized suggestion module and the notification unit are one of the specific implementation modes of the present invention, and can be flexibly adjusted according to business needs in practical applications. For example, the user behavior monitoring module can be implemented by other real-time data stream tools, the personalized suggestion module can generate learning paths according to different algorithm optimizations, and the notification unit can also support multiple communication protocols or third-party notification services. The specific implementation method of the invention is not limited to the above description, and can be expanded and adjusted according to business scenarios. Through the cooperation of the user behavior monitoring module, the personalized suggestion module and the notification unit, the present invention can monitor the learning behavior of business personnel in real time, generate personalized learning paths according to their performance, and communicate the feedback results to the business personnel in a timely manner, thereby forming a complete and personalized feedback mechanism to ensure that business personnel can continuously and effectively improve their learning and business performance.

[0092] like Figure 2 As shown, an embodiment of the present invention provides a flowchart of a behavior data evaluation method based on a hierarchical architecture, aiming to provide a comprehensive and efficient evaluation process to ensure that the learning and performance of business personnel can be accurately evaluated and fed back, including the following steps:

[0093] S100, real-time collection of learning behavior data and business contribution data of business personnel. In this step, the data collection module M100 collects relevant data of business personnel from different systems (such as online learning platforms, training management systems, and business management systems) in real time through multi-source data access units. This process uses API or data interface to stream data into the system. This step ensures the timeliness and integrity of the data and provides a reliable data foundation for subsequent processing.

[0094] S200, clean, integrate and analyze learning behavior data and business contribution data. In the data processing module M200, the data cleaning unit is responsible for identifying and processing outliers, missing values ​​and other issues to ensure the accuracy of the data. The cleaned data will be passed to the data integration unit, which matches and merges data from different sources to form a unified data format. This process may include field mapping and data standardization operations to ensure that all data is consistent before entering the analysis phase. Finally, the integrated data is deeply analyzed by the data analysis unit, which uses statistical analysis and machine learning algorithms to mine potential information in the data to provide a basis for evaluation.

[0095] S300, using the cleaned, integrated and analyzed data to calculate the evaluation indicators and generate the evaluation results. In this step, the evaluation indicator calculation module M300 uses the cleaned and integrated data to calculate the evaluation indicators of the business personnel based on the preset evaluation rules. These evaluation indicators reflect the performance of business personnel in multiple dimensions such as learning behavior and work contribution. The multi-dimensional evaluation indicator unit and the dynamic adjustment unit in the evaluation indicator calculation module M300 work together to ensure that the evaluation process can flexibly adapt to changes in business needs and automatically adjust the calculation rules according to real-time data.

[0096] S400: Feedback the evaluation indicators and recommended learning paths to the business personnel. In the feedback module M400, the system promptly communicates the evaluation results to the business personnel through the notification unit and provides personalized learning suggestions based on their points. Feedback information can be sent through various means such as email, SMS or mobile applications to ensure that business personnel can easily obtain relevant information.

[0097] Through the mutual coordination of the above steps, the present invention realizes a complete process from data collection to calculation of evaluation indicators and then to feedback of results, ensuring that the learning and performance of business personnel can be accurately evaluated and timely fed back. Behavioral data evaluation methods are some embodiments of the present invention. In practical applications, the methods of data collection, processing, evaluation and feedback can be adjusted according to specific business needs and technical environment, and are not limited to the specific steps or implementation details described. For example, methods of data cleaning and integration, methods of calculating evaluation indicators, and means of feedback notification can be optimized and changed according to actual conditions. The present invention allows flexible adjustment of the implementation method according to different business scenarios and needs to adapt to changing environments and goals.

[0098] The embodiment of the present invention also provides a behavior data evaluation program product based on a hierarchical architecture, the program product includes computer instructions, and when the computer instructions are executed by a processor, the functions of the behavior data evaluation system based on the hierarchical architecture described above are implemented. The computer instruction functions of this embodiment can be implemented using the specific implementation of the behavior data evaluation system based on the hierarchical architecture described above, which will not be described in detail here.

[0099] Another embodiment of the present invention provides a behavior data evaluation system based on a layered architecture. The functions of each functional module are the same as the behavior data evaluation system based on a layered architecture in the above embodiment. This embodiment provides a special learning platform for business personnel and merchants according to different roles and regions. These platforms rely on detailed user behavior data to efficiently manage content, analyze and predict user behavior, and optimize the empowerment effect through data. As the market develops, overseas merchant homepages have been launched to further mine user behavior data, and apply it to subsequent personnel incentive systems to improve operational efficiency. With these data, global businesses and merchants can be effectively empowered and operating performance can be improved.

[0100] The courses and training provided can provide business user behavior data.

[0101]

[0102]

[0103] Through the EDW layer of spot checks on quality inspections, exams, and viewing of live classes or replayed classes, the basic items and bonus items are aggregated into the ADM layer, and finally the wide table output is summarized.

[0104] The training will open some training courses regularly. In order to supervise the business learning courses, basic points are set. The basic points include: live broadcast participation, random inspections, elective courses and monthly exams.

[0105] The evaluation of participation in live classes requires analysis of the duration of training participation, as well as analysis of the coverage and quality of watching live broadcasts. The calculation of the completion rate is based on a combination of live broadcasts and replays.

[0106] The course quality will be evaluated by spot checking all business personnel after the course, using stratified sampling. Finally, the basic score will be calculated by combining the courses, exams, and elective courses.

[0107] In this embodiment, the evaluation indicator selection is implemented in the form of points, but it is not limited to this and can also exist in other forms, such as awards, badges, etc.

[0108] In this embodiment, part of the data model constituting the data source of the behavior data evaluation system based on the layered architecture is as follows:

[0109] Course entity: defines the basic attributes of a course, such as course name, course number, credits, etc. Courses can be divided into different categories, such as required courses, elective courses, etc.

[0110] Lecturer entity: defines the basic attributes of a lecturer, such as name, work number, professional field, etc. A lecturer can be responsible for multiple courses.

[0111] Project entity: defines projects related to a course, such as labs, assignments, exams, etc. Projects can be associated with courses and instructors.

[0112] Points system entity: defines the basic properties of the points system, such as points rules, points level, etc. Points systems can be associated with courses, instructors, and projects.

[0113] In this embodiment, the data layers of the big data hierarchical architecture are:

[0114] Operational Data Layer: The ODS layer in the data warehouse. It contains operational data such as business, courses, lecturers, and projects, and is used to record business course selection, lecturer teaching, project completion, etc.

[0115] Integrated Data Layer: The EDW and CDM layers in the data warehouse. The data in the operational data layer is integrated and cleaned for subsequent analysis and decision support.

[0116] Analytical Data Layer: The ADM layer of the data warehouse. Based on the integrated data layer, data analysis and mining are performed to gain insights and understanding of business learning and instructor teaching.

[0117] Decision Data Layer: The ADM layer in data modeling. Based on the results of the analytical data layer, decisions and plans are made to optimize the effectiveness of business learning and instructor teaching.

[0118] In this embodiment, the behavior data evaluation system based on the layered architecture is implemented as follows:

[0119] Data collection and storage: Through the business course selection system, lecturer teaching system, etc., collect and store data related to business, courses, lecturers, projects, etc.

[0120] Data cleaning and integration: Clean and integrate the collected data to ensure data accuracy and consistency.

[0121] Data analysis and mining: Use data analysis and mining techniques to conduct statistical analysis, correlation analysis, etc. on the cleaned and integrated data to obtain information related to business learning and instructor teaching.

[0122] Decision support and optimization: Based on the results of data analysis, make corresponding decisions and plans to optimize the effectiveness of business learning and instructor teaching.

[0123] By designing and implementing data modeling and implementation of a behavioral data evaluation system based on a hierarchical architecture, we can effectively motivate business learning behaviors and improve the quality and effectiveness of education.

[0124] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A behavior data evaluation system based on a hierarchical architecture, characterized in that: include: Data collection module, used to collect business personnel's behavior data and business contribution data in real time; A data processing module, which cleans, integrates and analyzes the behavior data and business contribution data based on a big data hierarchical architecture; An evaluation index calculation module, used to obtain evaluation indicators based on the cleaned, integrated and analyzed data and generate evaluation results; The feedback module is used to provide real-time feedback to business personnel on their evaluation indicators, evaluation results, and recommended learning paths.

2. The behavior data evaluation system based on a hierarchical architecture according to claim 1, characterized in that: The big data layered architecture includes: The enterprise data warehouse layer is used to centrally store data from different data sources; A public data mart layer, used to integrate and standardize the data stored in the enterprise data warehouse layer; The application data mart layer is used to summarize and analyze the data integrated and standardized by the public data mart layer according to specific business needs.

3. The behavior data evaluation system based on a layered architecture according to claim 2, characterized in that: The data processing module further comprises: Data cleaning unit, used to detect outliers and fill missing values ​​in raw data; A data integration unit, used for matching and merging the data from different data sources; The data analysis unit is used to apply statistical analysis and machine learning algorithms to perform in-depth analysis on the integrated and standardized data.

4. The behavior data evaluation system based on a layered architecture according to claim 3 is characterized in that: The statistical analysis method includes one or more of regression analysis, descriptive statistical analysis and variance analysis, and the machine learning algorithm includes one or more of decision tree, random forest and support vector machine.

5. The behavior data evaluation system based on a layered architecture according to claim 2, characterized in that: The evaluation index calculation module includes: A multi-dimensional evaluation index unit, used to calculate evaluation indexes based on basic factors and bonus factors recorded in the enterprise data warehouse layer; The dynamic adjustment unit is used to adjust the calculation rules of evaluation indicators according to the behavioral performance of business personnel and changes in business needs.

6. The behavior data evaluation system based on a layered architecture according to claim 5, characterized in that: The evaluation index calculation module also includes: A weight setting unit, used to adjust the calculation weights used for the evaluation index according to the factors of different factors recorded in the enterprise data warehouse layer; The cumulative evaluation index unit is used to generate the cumulative evaluation index of the business personnel according to their business contribution data and generate historical records.

7. The behavior data evaluation system based on a layered architecture according to claim 1, characterized in that: The data acquisition module further comprises: Multi-source data access unit, used to collect data from online learning platforms, training management systems and business management systems; The data preprocessing unit is used to perform preliminary cleaning and format conversion on the collected data.

8. The behavior data evaluation system based on a layered architecture according to claim 1, characterized in that: The feedback module further comprises: User behavior monitoring module, used to track the learning behavior and participation of business personnel in real time; Personalized suggestion module, which generates customized learning paths and improvement suggestions based on the evaluation indicators of business personnel; The notification unit is used to notify business personnel of their evaluation indicators, evaluation results and recommended learning paths through one or more methods including email, text message or mobile application.

9. A behavior data evaluation method based on a hierarchical architecture, characterized in that: The following steps are involved: Collect business personnel's learning behavior data and business contribution data in real time; Clean, integrate and analyze learning behavior data and business contribution data; Calculate evaluation indicators using cleaned, integrated and analyzed data and generate evaluation results; Provide feedback to business personnel on their evaluation indicators and recommended learning paths.

10. A behavior data evaluation program product based on a hierarchical architecture, characterized in that: The program product includes computer instructions, which, when executed by a processor, implement the functions of the behavior data evaluation system based on a hierarchical architecture according to any one of claims 1 to 8.

Images