System and method for assessing feasibility of scientific and technological achievement transformation based on multi-source data

Abstract

The application provides a technology achievement transformation feasibility evaluation system and method based on multi-source data, relates to the technical field of technology achievement transformation, and comprises a server for processing data, a system construction method for constructing a risk evaluation system and a prediction method for optimizing the risk evaluation system. The modules used in the application are designed for the data types to be processed, and internet data is accessed in the data analysis process to construct a risk index system and a risk evaluation model in a big data assisted verification manner. According to the design of calculating other data according to the change data, the difficulty of data acquisition can be greatly reduced, the workload of the management team is reduced, the time of risk evaluation is greatly saved, the risk evaluation result is more time-efficient than the risk evaluation result obtained in the general manner, and the success of technology achievement transformation is positively promoted.

Description

Technical Field

[0001] This invention relates to the field of technology transfer technology, and in particular to a system and method for feasibility assessment of technology transfer based on multi-source data. Background Technology

[0002] Technology transfer refers to the process of transforming practically valuable scientific and technological achievements generated from scientific research and technological development into new technologies, processes, materials, and products, and ultimately, new industries, through subsequent experimentation, development, application, and promotion. Technology transfer is a high-investment, high-risk process, and the effective control and mitigation of these risks are crucial factors limiting its successful transformation. Therefore, conducting comprehensive and systematic risk assessments is essential. This not only helps improve the success rate of technology transfer but also provides a scientific basis for decision-making. Technology transfer risk assessment involves a systematic analysis and evaluation of uncertainties related to technology, market, investment, and intellectual property that may be encountered during the process of transforming scientific and technological achievements into actual products or services. It is a systematic task involving the assessment of uncertainties across multiple aspects.

[0003] By using scientific evaluation methods, various risks can be effectively identified and prevented, thereby improving conversion rates.

[0004] Current technology transfer risk assessments mostly rely on manual data collection and analysis using various deep learning models. This process demands significant intervention from the management team. Furthermore, the sheer variety of data items involved in technology transfer risk assessments, coupled with potential changes in these items, further complicates the process. When data items change, the existing approach involves re-collecting each item, requiring re-screening, analysis, and evaluation – a highly time-consuming and labor-intensive process. In the rapidly evolving network economy, the sheer volume of collected data further exacerbates the difficulty of data acquisition for technology transfer, highlighting existing technical limitations. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies, alleviate the need for data acquisition and processing due to changes in data items, save time in risk assessment for technology transfer, and improve the efficiency of risk assessment.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a feasibility assessment system and method for the transformation of scientific and technological achievements based on multi-source data, comprising a server for processing data, a deep learning model, an assessment model construction module, a risk indicator system construction module, a data processing module, a data storage module, and an internet data acquisition module; the server is connected to a basic data collection module, a manual operation module, a risk assessment module, and a result reporting module; the data processing module includes a data cleaning module, a data conversion module, and a data integration module; the deep learning model includes a language analysis module, a data comparison module, a big data analysis module, a basic data analysis module, a prediction module, and a module for matching risk assessment items with data; the risk assessment module includes a technology risk assessment module, a market risk assessment module, an investment risk assessment module, a management risk assessment module, and an environmental risk assessment module; It also includes the construction of a risk assessment system based on servers, risk assessment modules, result reporting modules, basic data collection modules, and manual operation modules: first, basic data is collected, then language analysis is performed, then data cleaning and data transformation are carried out, then big data analysis and basic data analysis are performed, then an assessment model and risk indicator system are built, then data matching and risk assessment are performed, and finally the result report is generated.

[0007] In one preferred implementation, the basic data collection module collects data including technical characteristics data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, capital investment data, and management team data.

[0008] In a preferred embodiment, the results reporting module includes an evaluation report overview module, an evaluation indicator display module, a detailed risk display module, a response measure display module, and a data visualization module.

[0009] As a preferred implementation, the construction of the risk assessment system specifically includes the following steps: S1. Basic Data Collection: Collect basic data such as technical characteristic data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, capital investment data, and management team data through manual operation modules and the Internet, and classify the basic data into the basic data collection module; S2, Language Analysis: Retrieves basic data from the basic data collection module and analyzes the data through the language analysis module to obtain the language-analyzed data; S3. Data Cleaning and Data Transformation: After language analysis, the data is processed through the data cleaning and data transformation modules to obtain usable basic data that has eliminated errors and redundant content and improved data usability. S4. Big Data Analysis and Basic Data Analysis: Use the basic data analysis module to analyze the correlation between available basic data and obtain the correlation between basic data. At the same time, based on the basic data items, obtain the corresponding Internet data through the big data analysis module. Based on the correlation between basic data, use the big data analysis module to analyze Internet data and basic data to obtain the correlation between data of corresponding basic data items in Internet data. S5. Construct an evaluation model and risk indicator system: Based on the correlation between the data of corresponding basic data items in Internet data, clarify the objectives of the risk assessment system, plan the different stages and evaluation objectives of the technology transfer process, select different evaluation indicators, allocate indicator weights, construct a technology transfer risk indicator system, and then construct a technology transfer risk assessment model based on the constructed technology transfer risk indicator system. S6. Data Matching and Risk Assessment: Using a risk assessment model, a risk assessment module is created based on a risk indicator system. Then, according to different risk assessment projects, the data matching module is used to match the corresponding data analysis model. Finally, based on the data matching results of the risk assessment project, the risk assessment module is used to derive the risk assessment content. S7. Results Report: The evaluation report overview module simplifies and summarizes the risk assessment content; the evaluation indicator display module displays the evaluation indicator system for the transformation of scientific and technological achievements; the detailed risk display module displays the detailed risk assessment content for the transformation of scientific and technological achievements; the response measures display module displays the risk response measures suggestions formulated by the management team; and the data visualization module visualizes the basic data and risk assessment content.

[0010] In a preferred embodiment, step S7, the time-axis-based data visualization completed by the data integration module and the result reporting module, specifically includes the following steps: S7.1 First, starting from the completion time of the scientific and technological achievement, establish a timeline that grows over time. Then, through the data integration module, aggregate the basic data and risk assessment content within the same time period into one location, and establish a dataset based on the corresponding time point. Mark the dataset at the corresponding position on the timeline. The dataset is named according to the batch of data entry. The dataset content includes basic data items covering basic data and risk assessment items covering risk assessment content. Basic data items are established as subsets based on technical characteristic data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, capital investment data, and management team data. Risk assessment items are established as subsets based on technical risk assessment, market risk assessment, investment risk assessment, management risk assessment, and environmental risk assessment. Subsets under the corresponding data items and subsets under the risk assessment items are then matched one-to-one to establish a subset of record change records. S7.2 When the basic data or risk assessment data changes for the first time, the prediction model infers the changes in other basic data based on the changed data, uses the inferred basic data to evaluate the transformation of scientific and technological achievements, then integrates the changed basic data and risk assessment content, and regenerates the second data entry dataset based on the first data entry format. Change record markers are added under the corresponding basic data item subset and risk assessment subset, and change content subsets are created under the corresponding change records. The change technical data and risk assessment content classification numbers are assigned to the corresponding items. Finally, the second data entry dataset is associated with the corresponding time axis node of the visualization dataset. S7.3 When the basic data or risk assessment changes again, repeat step S7.2, delete the data entry subset except for the first data entry subset, and regenerate a new data entry dataset.

[0011] As a preferred embodiment, the system also includes a prediction method for optimizing the risk assessment system, the prediction method comprising the following steps: 1) When the change data is obtained, the change items are first identified through the data comparison module, then the change items are analyzed through the language analysis model, and then the prediction module is used to combine the relationship between the data of the Internet data corresponding to the basic data items in the big data analysis module and the relationship between the basic data to obtain the estimated basic data. 2) While the prediction module extrapolates changes in other data items based on the changed basic data or risk assessment data, the basic data is re-collected and language analysis is performed. Then, usable basic data is obtained through data cleaning and data transformation. 3) Use the basic data analysis module to compare the differences between the re-collected basic data and the extrapolated basic data. If the difference of any basic data item is greater than the threshold, the risk assessment will be re-evaluated based on the re-collected basic data. If the difference of any basic data item is less than the threshold, the risk assessment content based on the extrapolated basic data will be used directly.

[0012] In a preferred embodiment, in step 3): 3.1) When the difference between the estimated basic data and the available basic data is less than the threshold: the re-collected data is retained, and the estimated risk assessment content is directly adopted. Combined with the estimated basic data, the data is visualized through the result reporting module and the data integration module to obtain the estimated visualized data. 3.2) When the difference between the estimated basic data and the available basic data exceeds a threshold, the available basic data is retained, the estimated changed data and the estimated risk assessment content are deleted, and the risk assessment and data visualization are re-performed using the available basic data. Since the basic data contains uncontrollable and unquantifiable data, it is necessary to re-analyze the correlation between the basic data. When the data correlation has not changed, the original risk assessment model is used for risk assessment. When the data correlation has changed, the risk indicator system and assessment model are reconstructed, and then the risk assessment is performed.

[0013] In a preferred embodiment, in step 3.2), when the difference between the estimated basic data and the available basic data is greater than a threshold, the prediction module algorithm is changed, a new prediction model is constructed using the changed prediction model algorithm, and the new prediction model is used to estimate the changed data to obtain new estimated basic data. Then, the data comparison module is used to recompare the new estimated basic data and the available basic data. When the difference between the two data is less than the threshold, the process ends. When the difference between the two data is greater than the threshold, the operation of changing the prediction module algorithm is repeated.

[0014] As a preferred implementation, the threshold setting in step 3) is as follows: based on 5%, the specific error thresholds of each basic data are redistributed according to the weight of the risk assessment indicators corresponding to each basic data.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: 1. The constituent modules used in this invention are all designed for the data types they are to be processed, and Internet data is accessed during the data analysis process. A risk indicator system and risk assessment model are constructed in a way that is assisted by big data. The risk assessment content obtained under the verification of a large amount of data is also accurate.

[0016] 2. The present invention, by inferring other data based on changed data, can greatly reduce the difficulty of data acquisition, not only reducing the workload of the management team, but also saving a lot of time for risk assessment. This makes the risk assessment results more timely than those obtained by conventional methods, thus providing a positive impetus for the successful transformation of scientific and technological achievements.

[0017] 3. This invention verifies the accuracy of the predicted data by the prediction model by comparing the predicted changes with the re-collected basic data, thereby continuously optimizing the prediction model. When the accuracy and stability of the predicted data meet the requirements, the process of re-collecting and analyzing basic data can be saved when subsequent data changes occur. It is only necessary to re-verify the prediction model from time to time to confirm its accuracy and stability.

[0018] 4. The way this invention integrates and visualizes basic data and risk assessment content allows the management team to intuitively understand the relationship between basic data and risk assessment content, which can provide inspiration for adjusting risk assessment indicators to a certain extent. Attached Figure Description

[0019] Figure 1 This invention presents a module diagram of a feasibility assessment system and method for the commercialization of scientific and technological achievements based on multi-source data. Figure 2 The flowchart below shows the feasibility assessment system and method for the transformation of scientific and technological achievements based on multi-source data proposed in this invention. Figure 3 The present invention presents a system construction flowchart for a technology transfer feasibility assessment system and method based on multi-source data. Figure 4 This invention proposes a flowchart for change data prediction and system optimization of a technology transfer feasibility assessment system and method based on multi-source data. Figure 5 This invention proposes a time-axis-based data visualization architecture diagram for a technology transfer feasibility assessment system and method based on multi-source data. Figure 6 This is the first data entry content architecture diagram in the time-axis-based data visualization architecture of the technology transfer feasibility assessment system and method based on multi-source data proposed in this invention. Figure 7 This is the second data entry content architecture diagram in the time-axis-based data visualization architecture of the technology achievement transformation feasibility assessment system and method based on multi-source data proposed in this invention. Figure 8This invention presents the Nth data entry content architecture diagram within a time-axis-based data visualization architecture for a technology transfer feasibility assessment system and method based on multi-source data. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example 1

[0021] like Figure 1 As shown, a feasibility assessment system and method for the commercialization of scientific and technological achievements based on multi-source data includes methods for completing the data analysis. The server for processing includes a deep learning model built on a deep learning model, a data processing module for processing data, a data storage module for storing data, and an internet data acquisition module for acquiring internet data. The server is connected to a basic data collection module for basic data collection, a manual operation module, a risk assessment module for risk assessment, and a result reporting module for displaying assessment results. The basic data collection module collects data on the following types of data: technical characteristics data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, capital investment data, and management team data. Furthermore, in constructing the risk indicator system, the management team establishes the entire process of technology transfer, clarifies the objectives of the evaluation system, selects different evaluation indicators based on different stages of the technology transfer process and evaluation objectives, analyzes the impact of each evaluation indicator on technology transfer, and assigns different weights to each indicator to establish the risk indicator system. In constructing the evaluation model, the management team analyzes and selects appropriate evaluation models for different product types based on the evaluation indicators and indicator items with different weights. Evaluation model types include, but are not limited to, hierarchical analysis models and fuzzy comprehensive evaluation models.

[0022] In the above process, data related to scientific and technological achievements and products are obtained through the basic data collection module. Based on this data, after data processing by the data processing module, a deep learning model is used to perform data analysis and data matching. Then, the risk assessment module is used to assess the risks of scientific and technological achievements transformation, and the risk assessment results are reported by the results reporting module. Example 2

[0023] like Figure 1 As shown, the data processing module includes a data cleaning module for removing errors and redundant content from the data, a data transformation module for improving the usability of the data, and a data integration module for aggregating data from different sources into one location; As mentioned above, due to the diversity of the sources of basic data and the complexity of the data itself, unprocessed data often has problems such as missing values, outliers, duplicate records, and inconsistent formats. Using data cleaning and data transformation modules to clean the data can filter out useful information from a large amount of raw data, remove errors and redundant content, and thus lay a solid foundation for subsequent data analysis and model building. The data integration module is used to aggregate data from different sources into one location to provide a unified data view and support data analysis and decision-making. Example 3

[0024] like Figure 1 As shown, the deep learning model includes a language analysis module for analyzing language in the basic data, a data comparison module for comparing newly acquired basic data with the original basic data, a big data analysis module for analyzing basic data based on continuously acquired Internet data, a basic data analysis module for analyzing basic data, a prediction module for predicting other basic data based on changed basic data, and a data matching module for matching risk assessment projects with basic data. The language analysis module, as described above, aims to analyze the language and mathematical content in the data. It can employ a large language model to accurately analyze the meaning of written language. The primary purpose of the basic data analysis module is to correlate various basic data points and analyze the relationships between them. Since the sources of basic data are diverse, this module can be built upon a multimodal learning model to analyze data from different sources. The big data analysis module provides the functionality to analyze basic data based on big data analytics. To analyze large amounts of data and extract causal relationships between various data points, the big data analysis module can use the Kumtz model. The Kumtz model contains a deep learning model layer. Through this deep learning model layer, abstract high-level features are extracted from the raw data using convolutional neural networks or recurrent neural networks. These features can include unquantifiable management team data, competitive environment data, policy and regulatory data, etc. The data matching module provides the function of matching risk assessment projects and basic data projects for different risk assessment contents. For the basic data related to different types of risk assessment projects, it is necessary to use targeted data analysis models one by one. Therefore, the data matching module is based on deep learning models and automatically matches the corresponding data analysis models for different risk assessment projects and different project basic data, and then performs the corresponding data matching. The prediction module is used in conjunction with the big data analysis module and the basic data analysis module when the basic data changes. It uses the data relationship between the big data analysis module and the basic data analysis module as the calculation logic, and uses the changed basic data as the basis to make calculations on other basic data. Example 4

[0025] like Figure 1 As shown, the risk assessment module includes a technology risk assessment module for assessing the technology risks during the technology transfer process, a market risk assessment module for assessing the market risks after the technology transfer, an investment risk assessment module for assessing the investment situation during the technology transfer process, a management risk assessment module for assessing the management situation during the technology transfer, and an environmental risk assessment module for assessing the market environment. In the above content, the risk assessment module is based on the results of the data comparison module, big data analysis module, basic data analysis module and data matching module in the deep learning model, the risk assessment model for technology transfer constructed by the assessment model construction module, and the risk indicator system for technology transfer constructed by the risk indicator system construction module, and assesses the technology risk, market risk, investment risk, management risk and environmental risk respectively. Example 5

[0026] like Figure 1 As shown, the results reporting module includes an evaluation report overview module for simplifying the evaluation report, an evaluation indicator display module for displaying risk assessment indicators for technology transfer, a detailed risk display module for displaying detailed risk information for technology transfer, a response measure display module for displaying risk response measures for technology transfer, and a data visualization module that visualizes and integrates data based on the data integration module. The above results are displayed in different formats. Therefore, corresponding display windows are designed based on the risk assessment indicators for technology transfer, detailed risk information for technology transfer, risk response measures for technology transfer, and visualization data, and displayed on the same display device. The visualization module generates visualization data by integrating basic data and risk assessment content through the data integration module. The integrated dataset is then displayed on a timeline. Furthermore, new datasets are generated based on changes in the basic data and risk assessment content and displayed on the same timeline. This design allows for a clear and intuitive observation of the changes in basic data and risk assessment content, facilitating the review of risk assessment results for technology transfer. Example 6

[0027] like Figure 2 and Figure 3 As shown, the construction process of a big data-based technology transfer risk assessment system is as follows: S1. Basic Data Collection: Collect technical characteristic data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, and capital investment data through manual operation modules and the Internet. Basic data, such as data from the data management team, will be categorized under the basic data collection module. Specifically, technical characteristic data is obtained by the management team through basic parameter collection and performance data testing of the product. Combined with intellectual property data, it can reflect the basic technical characteristics of the product. Market demand data is obtained by analyzing market demand based on the performance data in the technical characteristic data. Competitive environment data, economic environment data, and similar case data are collected from relevant data of similar products in the market based on market demand data. Combined with policy and regulatory data and investment data, it can provide data support for marketing planning. Management team data can reflect the stability of the product's production, marketing, and other aspects. Through comprehensive analysis, the above data can provide strong data support for risk assessment in the manufacturing and marketing of scientific and technological products. S2, Language Analysis: Retrieves basic data from the basic data collection module and analyzes the data through the language analysis module to obtain the language-analyzed data; Specifically, due to differences in the sources and forms of representation of basic data, the language analysis module needs to meet the requirements for understanding and analyzing various types of language, such as written language, mathematical language, and programming language. Using a large language model can analyze the specific meaning of different language types in the basic data and provide data support for subsequent data processing. S3. Data Cleaning and Data Transformation: After language analysis, the data is processed through the data cleaning and data transformation modules to obtain usable basic data that has eliminated errors and redundant content and improved data usability. Specifically, the data after language analysis can undergo a certain degree of data cleaning through the data transformation module, but its main function is still to transform the data to ensure the usability of data from different sources. The data cleaning module can further remove errors and erroneous content from the data, thereby further improving the usability of the data. S4. Big Data Analysis and Basic Data Analysis: Use the basic data analysis module to analyze the correlation between available basic data and obtain the correlation between basic data. At the same time, based on the basic data items, obtain the corresponding Internet data through the big data analysis module. Based on the correlation between basic data, use the big data analysis module to analyze Internet data and basic data to obtain the correlation between data of corresponding basic data items in Internet data. Specifically, the relationships between basic data derived from the basic data analysis module built on the multimodal model can not only reflect the causal relationships between basic data, but also serve as a foundation for the big data analysis module to establish relationships between corresponding basic data items in the Internet data during the big data analysis process. S5. Construct an evaluation model and risk indicator system: Based on the correlation between the data of corresponding basic data items in Internet data, clarify the objectives of the risk assessment system, plan the different stages and evaluation objectives of the technology transfer process, select different evaluation indicators, allocate indicator weights, construct a technology transfer risk indicator system, and then construct a technology transfer risk assessment model based on the constructed technology transfer risk indicator system. Specifically, the risk assessment system aims to plan for various risk projects. This planning process requires basic data support, and there are also correlations between the objectives of different risk assessment projects. Therefore, it is necessary to determine the correlations between basic data projects and the correlations between corresponding basic data projects in internet data to identify the relationships between risk assessment projects. This clarifies the objectives of the risk assessment system, enabling the planning of different stages and assessment objectives in the technology transfer process. Furthermore, based on the varying importance of the basic data, different assessment indicators are selected and weights are assigned to these indicators, thus constructing a technology transfer risk indicator system. Based on the different plans for risk projects within the risk indicator system, the assessment model construction module selects the corresponding assessment algorithm for each risk assessment project, thereby constructing a technology transfer risk assessment model. S6. Data Matching and Risk Assessment: Using a risk assessment model, a risk assessment module is created based on a risk indicator system. Then, according to different risk assessment projects, the data matching module is used to match the corresponding data analysis model. Finally, based on the data matching results of the risk assessment project, the risk assessment module is used to derive the risk assessment content. Specifically, the data matching module, built on a deep learning model, can automatically match the corresponding data analysis model based on the basic data and risk assessment projects of different projects. This serves as preparation for the risk assessment module to conduct risk assessments. The matched risk assessment projects and available basic data are then evaluated by the risk assessment module to derive the risk assessment content. Among them, the risk assessment model serves as the foundation, and the risk indicator system serves as the standard for risk assessment. The resulting risk assessment module can accurately assess the risks of technology transfer based on the risk indicators. S7. Results Report: The evaluation report overview module simplifies and summarizes the risk assessment content; the evaluation indicator display module displays the evaluation indicator system for the transformation of scientific and technological achievements; the detailed risk display module displays the detailed risk assessment content for the transformation of scientific and technological achievements; the response measures display module displays the risk response measures suggestions formulated by the management team; and the data visualization module visualizes the basic data and risk assessment content. Specifically, during the results reporting process, the risk assessment content and basic data need to be visualized together. The risk assessment content is divided and displayed under the categories of the results reporting module, and the exact risk assessment project content is integrated with the available basic data through the data integration module. The time-axis-based presentation created in this way can also be displayed together with data changes, making it convenient for the management team to view. The above steps can obtain the relationships between the basic data components, the relationships between the data of the corresponding basic data items in the Internet data, the risk indicator system, the risk assessment model, the risk assessment content, and the visualization data, thereby determining the composition of the system of the present invention. Example 7

[0028] like Figure 4 As shown, the prediction method includes the following steps: 1) When data changes, the changed data is first processed, and then the prediction module is used to estimate the remaining basic data based on the relationship between the changed data and the basic data. 2) While the forecasting module extrapolates changes in other data items based on the changed basic data or risk assessment data, the basic data is re-collected; 3) Use the basic data analysis module to compare the differences between the re-collected basic data and the extrapolated basic data. If the difference of any basic data item is greater than the threshold, the risk assessment will be re-evaluated based on the re-collected basic data. If the difference of any basic data item is less than the threshold, the risk assessment content based on the extrapolated basic data will be used directly. Specifically, when the change data is obtained, the change items are first identified through the data comparison module, then the change items are analyzed through the language analysis model, and then the prediction module is used to combine the correlation between the data of the Internet data corresponding to the basic data items in the big data analysis module and the correlation between the basic data to obtain the estimated basic data. Then, the risk analysis module and the risk assessment module are used to obtain the estimated risk assessment content based on the estimated basic data. In this process, while predicting and extrapolating the basic data based on the changed data, the basic data is re-collected and language analysis is performed. Then, usable basic data is obtained through data cleaning and data transformation. Finally, the data comparison module is used to compare the extrapolated basic data with the usable basic data and calculate the difference between the two. Furthermore, a difference threshold is set: based on a 5% baseline, and determined according to the risk assessment indicators corresponding to each baseline data point. The target weights are redistributed based on a 5% threshold for specific errors in each basic data item. When the difference between the estimated basic data and the available basic data is less than the threshold, the newly collected data is retained, and the estimated risk assessment content is directly adopted. Combined with the estimated basic data, the estimated visual data is obtained through the results reporting module and the data integration module. Furthermore, when the difference between the estimated baseline data and the available baseline data exceeds a threshold, the available baseline data is retained, the estimated changed data and the estimated risk assessment content are deleted, and the risk assessment and data visualization are re-performed using the available baseline data. Since the baseline data contains uncontrollable and unquantifiable data, it is necessary to re-analyze the correlations between the available baseline data. If the data correlations remain unchanged, the original risk assessment model is used for risk assessment. If the data correlations change, the risk indicator system and assessment model are reconstructed before conducting a risk assessment. Furthermore, when the difference between the estimated base data and the available base data exceeds a threshold, the prediction module algorithm is changed, a new prediction model is constructed using the changed prediction model algorithm, and the new prediction model is used to estimate the changed data to obtain new estimated base data. Then, the data comparison module is used to recompare the new estimated base data and the available base data. The process ends when the difference between the two data is less than the threshold, and repeats the operation of changing the prediction module algorithm when the difference between the two data is greater than the threshold. As mentioned above, the newly collected basic data requires reprocessing and analysis during risk assessment, consuming significant computer resources and time, resulting in low efficiency in obtaining risk assessment results. In contrast, the basic data extrapolated by the prediction module does not require data processing and analysis and can be directly generated in the processed data format, thus saving substantial computer resources and time. By setting a threshold for differences in basic data and continuously training the prediction module with new basic data or risk assessment content, the accuracy of predictions can be improved, gradually reducing the number of verifications, saving costs, and increasing the efficiency of risk assessment. Example 8

[0029] like Figure 5 , Figure 6 , Figure 7 and Figure 8 As shown, in step S7, the time-axis-based data visualization completed by the data integration module and the result reporting module specifically includes the following steps: S7.1 First, starting from the completion time of the scientific and technological achievement, establish a timeline that grows over time. Then, through the data integration module, aggregate the basic data and risk assessment content within the same time period into one location, and establish a dataset based on the corresponding time point. Mark the dataset at the corresponding position on the timeline. The dataset is named according to the batch of data entry. The dataset content includes basic data items covering basic data and risk assessment items covering risk assessment content. Basic data items are established as subsets based on technical characteristic data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, capital investment data, and management team data. Risk assessment items are established as subsets based on technical risk assessment, market risk assessment, investment risk assessment, management risk assessment, and environmental risk assessment. Subsets under the corresponding data items and subsets under the risk assessment items are then matched one-to-one to establish a subset of record change records. S7.2 When the basic data or risk assessment data changes for the first time, the prediction model infers the changes in other basic data based on the changed data, uses the inferred basic data to evaluate the transformation of scientific and technological achievements, then integrates the changed basic data and risk assessment content, and regenerates the second data entry dataset based on the first data entry format. Change record markers are added under the corresponding basic data item subset and risk assessment subset, and change content subsets are created under the corresponding change records. The change technical data and risk assessment content classification numbers are assigned to the corresponding items. Finally, the second data entry dataset is associated with the corresponding time axis node of the visualization dataset. S7.3 When the basic data or risk assessment changes again, repeat step S7.2 and delete everything except the first one. A new data entry dataset is generated from a subset of data entries outside the previous subset. In the above content, the data items in the timeline-based data visualization only include basic data items and risk assessment items. The former is data that can be directly obtained and viewed, while the latter is the risk assessment result analyzed based on the basic data. When both are on the same interface, the relationship between the basic data and the risk assessment content can be seen intuitively, which provides great help to the management team in planning the transformation and commercialization of scientific and technological achievements.

[0030] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments for application in other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A feasibility assessment system and method for technology transfer based on multi-source data, characterized in that: The system includes a server for processing data, comprising a deep learning model, an evaluation model building module, a risk indicator system building module, a data processing module, a data storage module, and an internet data acquisition module. The server connects to a basic data collection module, a manual operation module, a risk assessment module, and a result reporting module. The data processing module includes a data cleaning module, a data transformation module, and a data integration module. The deep learning model includes a language analysis module, a data comparison module, a big data analysis module, a basic data analysis module, a prediction module, and a module for matching risk assessment items with data. The risk assessment module includes a technology risk assessment module, a market risk assessment module, an investment risk assessment module, a management risk assessment module, and an environmental risk assessment module. The system also includes the construction of a risk assessment system based on the server, risk assessment module, result reporting module, basic data collection module, and manual operation module: first, basic data is collected; then, language analysis is performed; then, data cleaning and data transformation are performed; then, big data analysis and basic data analysis are performed; then, an evaluation model and risk indicator system are built; then, data matching and risk assessment are performed; and finally, a result report is generated.

2. The technology transfer feasibility assessment system and method based on multi-source data according to claim 1, characterized in that: The basic data collection module collects data on the following types of data: technical characteristics, market demand, competitive environment, policy and regulations, intellectual property, economic environment, similar cases, capital investment, and management team.

3. The technology transfer feasibility assessment system and method based on multi-source data according to claim 1, characterized in that: The results reporting module includes an evaluation report overview module, an evaluation indicator display module, a detailed risk display module, a response measure display module, and a data visualization module.

4. The technology transfer feasibility assessment system and method based on multi-source data according to claim 1, characterized in that: The construction of the risk assessment system specifically includes the following steps: S1. Basic Data Collection: Collect basic data such as technical characteristic data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, capital investment data, and management team data through manual operation modules and the Internet, and classify the basic data into the basic data collection module; S2, Language Analysis: Retrieves basic data from the basic data collection module and analyzes the data through the language analysis module to obtain the language-analyzed data; S3. Data Cleaning and Data Transformation: After language analysis, the data is processed through the data cleaning and data transformation modules to obtain usable basic data that has eliminated errors and redundant content and improved data usability. S4. Big Data Analysis and Basic Data Analysis: Use the basic data analysis module to analyze the correlation between available basic data and obtain the correlation between basic data. At the same time, based on the basic data items, obtain the corresponding Internet data through the big data analysis module. Based on the correlation between basic data, use the big data analysis module to analyze Internet data and basic data to obtain the correlation between data of corresponding basic data items in Internet data. S5. Construct an evaluation model and risk indicator system: Based on the correlation between the data of corresponding basic data items in Internet data, clarify the objectives of the risk assessment system, plan the different stages and evaluation objectives of the technology transfer process, select different evaluation indicators, allocate indicator weights, construct a technology transfer risk indicator system, and then construct a technology transfer risk assessment model based on the constructed technology transfer risk indicator system. S6. Data Matching and Risk Assessment: Using a risk assessment model, a risk assessment module is created based on a risk indicator system. Then, according to different risk assessment projects, the data matching module is used to match the corresponding data analysis model. Finally, based on the data matching results of the risk assessment project, the risk assessment module is used to derive the risk assessment content. S7. Results Report: The evaluation report overview module simplifies and summarizes the risk assessment content; the evaluation indicator display module displays the evaluation indicator system for the transformation of scientific and technological achievements; the detailed risk display module displays the detailed risk assessment content for the transformation of scientific and technological achievements; the response measures display module displays the risk response measures suggestions formulated by the management team; and the data visualization module visualizes the basic data and risk assessment content.

5. The technology transfer feasibility assessment system and method based on multi-source data according to claim 4, characterized in that: In step S7, the time-axis-based data visualization completed by the data integration module and the result reporting module specifically includes the following steps: S7.1 First, starting from the completion time of the scientific and technological achievement, establish a timeline that grows over time. Then, through the data integration module, aggregate the basic data and risk assessment content within the same time period into one location, and establish a dataset based on the corresponding time point. Mark the dataset at the corresponding position on the timeline. The dataset is named according to the batch of data entry. The dataset content includes basic data items covering basic data and risk assessment items covering risk assessment content. Basic data items are established as subsets based on technical characteristic data, market demand data, competitive environment data, policy and regulatory data, intellectual property data, economic environment data, similar case data, capital investment data, and management team data. Risk assessment items are established as subsets based on technical risk assessment, market risk assessment, investment risk assessment, management risk assessment, and environmental risk assessment. Subsets under the corresponding data items and subsets under the risk assessment items are then matched one-to-one to establish a subset of record change records. S7.2 When the basic data or risk assessment data changes for the first time, the prediction model infers the changes in other basic data based on the changed data, uses the inferred basic data to evaluate the transformation of scientific and technological achievements, then integrates the changed basic data and risk assessment content, and regenerates the second data entry dataset based on the first data entry format. Change record markers are added under the corresponding basic data item subset and risk assessment subset, and change content subsets are created under the corresponding change records. The change technical data and risk assessment content classification numbers are assigned to the corresponding items. Finally, the second data entry dataset is associated with the corresponding time axis node of the visualization dataset. S7.3 When the basic data or risk assessment changes again, repeat step S7.2, delete the data entry subset except for the first data entry subset, and regenerate a new data entry dataset.

6. The technology transfer feasibility assessment system and method based on multi-source data according to claim 1, characterized in that: It also includes a prediction method for optimizing the risk assessment system, the prediction method comprising the following steps: 1) When the change data is obtained, the change items are first identified through the data comparison module, then the change items are analyzed through the language analysis model, and then the prediction module is used to combine the relationship between the data of the Internet data corresponding to the basic data items in the big data analysis module and the relationship between the basic data to obtain the estimated basic data. 2) While the prediction module extrapolates changes in other data items based on the changed basic data or risk assessment data, the basic data is re-collected and language analysis is performed. Then, usable basic data is obtained through data cleaning and data transformation. 3) Use the basic data analysis module to compare the differences between the re-collected basic data and the extrapolated basic data. If the difference of any basic data item is greater than the threshold, the risk assessment will be re-evaluated based on the re-collected basic data. If the difference of any basic data item is less than the threshold, the risk assessment content based on the extrapolated basic data will be used directly.

7. The technology transfer feasibility assessment system and method based on multi-source data according to claim 6, characterized in that: In step 3): 3.1) When the difference between the estimated basic data and the available basic data is less than the threshold: the re-collected data is retained, and the estimated risk assessment content is directly adopted. Combined with the estimated basic data, the data is visualized through the result reporting module and the data integration module to obtain the estimated visualized data. 3.2) When the difference between the estimated baseline data and the available baseline data exceeds a threshold, the available baseline data is retained, the estimated changed data and the estimated risk assessment content are deleted, and the risk assessment and data visualization are re-performed using the available baseline data. Since the baseline data contains uncontrollable and unquantifiable data, it is necessary to re-analyze the correlation between the baseline data and the available baseline data. When the data correlation has not changed, the original risk assessment model is used for risk assessment. When the correlation between data changes, the risk indicator system and assessment model are reconstructed, and then the risk assessment is carried out again.

8. The technology transfer feasibility assessment system and method based on multi-source data according to claim 7, characterized in that: In step 3.2), when the difference between the estimated basic data and the available basic data is greater than a threshold, the prediction module algorithm is changed, a new prediction model is constructed using the changed prediction model algorithm, and the new prediction model is used to estimate the changed data to obtain new estimated basic data. Then, the data comparison module is used to recompare the new estimated basic data and the available basic data. When the difference between the two data is less than the threshold, the process ends. When the difference between the two data is greater than the threshold, the operation of changing the prediction module algorithm is repeated.

9. The technology transfer feasibility assessment system and method based on multi-source data according to claim 6, characterized in that: In step 3), the threshold setting is as follows: based on 5%, the specific error thresholds of each basic data are redistributed according to the weight of the risk assessment indicators corresponding to each basic data.

Images