Scientific and technological achievement prediction and evaluation system based on big data

Abstract

The invention discloses a scientific and technological achievement prediction and evaluation system based on big data, and relates to the technical field of scientific and technological achievement prediction and evaluation, and the system specifically comprises a research and development process data collection module, an information scientific and technological achievement prediction module, and an information scientific and technological achievement comprehensive evaluation module. Wherein the research and development process data collection module is used for collecting historical research data, feature data and network transmission performance data of scientific research projects in the research and development process in real time, and the information scientific and technological achievement prediction module is used for establishing a prediction model of the feature data of the scientific and technological achievement research process and setting research process state indexes based on the feature data. The LSTM prediction module is used for obtaining network transmission performance data in a stationary time range and screening out the network transmission performance data in the stationary time range for further training and optimizing the LSTM prediction model, and the information scientific and technological achievement comprehensive evaluation module is used for comprehensively evaluating the research process stability of the information scientific and technological achievements. And setting an information research fluctuation index to predict the stability of the research process.

Description

Technical Field

[0001] The present invention belongs to the technical field of prediction and evaluation of scientific and technological achievements, and particularly relates to a prediction and evaluation system for scientific and technological achievements based on big data. Background Art

[0002] Technology Foresight is a systematic and structured method aimed at identifying emerging technologies or technology combinations that may have a significant impact on the economy, society, and environment by predicting and analyzing future trends in technological development. It emphasizes systematicness, long-term nature, and strategic nature, involving comprehensive analysis in multiple fields such as economy, society, and environment, and aims to select strategic research fields and general new technologies that may generate the greatest economic and social benefits.

[0003] Previous systems mainly focused on static data such as the application and approval of scientific research projects, while ignoring dynamic data during the R & D process, such as laboratory equipment operation data, equipment status data, R & D progress data, etc. These data are crucial for comprehensively evaluating scientific and technological achievements. Traditional prediction methods often rely on simple statistical models or expert experience, lacking accurate mathematical models and big data support, resulting in inaccurate prediction results and being unable to provide strong support for scientific research decisions. Existing scientific and technological achievement evaluation systems often only focus on the final scientific research achievements, while ignoring indicators such as stability and data integrity during the research process. These indicators are equally important for ensuring the smooth progress of research work.

[0004] Therefore, there is an urgent need for a prediction and evaluation system for scientific and technological achievements based on big data, which can realize the prediction and evaluation of the research process of scientific and technological achievements by integrating laboratory equipment operation data, equipment status data, and R & D progress data. Summary of the Invention

[0005] The purpose of the present invention is to provide a prediction and evaluation system for scientific and technological achievements based on big data, which is used to solve the technical problem in the prior art that only focuses on the final scientific research achievements while ignoring the stability and data integrity during the research process.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solutions: A prediction and evaluation system for scientific and technological achievements based on big data, comprising: An R & D process data collection module, an information science and technology achievement prediction module, and an information science and technology achievement comprehensive evaluation module. The R & D process data collection module is used to collect historical research data, characteristic data, and network transmission performance data of scientific research projects during the R & D process in real time, and determine the characteristic data set corresponding to different research objectives; Among them, the historical research data includes research objectives and technical routes. The time period from the preparation to the end of each research objective is recorded as a research stage, and the characteristic data and network transmission performance data of each research stage in the R & D laboratory are monitored in real time; The characteristic data includes equipment operation data, equipment status data, and R & D progress data; The network transmission performance data includes the number of error bits, the total number of transmitted bits, and the number of lost data packets; The information technology achievement prediction module sets research process status indicators based on the characteristic data, screens out the network transmission performance data within the stable time range, establishes a prediction model for the characteristic data of the science and technology achievement research process based on the LSTM network structure, and uses the network transmission performance data within the stable time range to train and optimize the LSTM prediction model; The information technology achievement comprehensive evaluation module is used to comprehensively evaluate the stability of the research process of information technology achievements. By predicting the real-time network transmission performance data, an information research fluctuation index is set to achieve the prediction of the stability of the research process.

[0007] Furthermore, the characteristic data sets corresponding to different research objectives are determined. The specific method is as follows: Sort according to the research order of scientific and technological achievements, and record the set of research objectives in the research process of the obtained scientific and technological achievements as , where Y(u) represents the set of research objectives of scientific and technological achievement u, i represents the research objective numbered i, n represents that scientific and technological achievement u needs to complete n research objectives in total, and the set of characteristic data of the research stage where research objective i is located is recorded as , where yx(i) represents the set of equipment operation data of the R & D laboratory in the research stage represented by the i-th research objective, xh(i) represents the set of material consumption data of the R & D laboratory in the research stage represented by the i-th research objective, and zt(i) represents the set of equipment status data of the R & D laboratory in the research stage represented by the i-th research objective; Among them, the equipment operation data includes the actual operation time of the equipment within a specific time period and the energy consumption of the equipment during operation; The equipment status data includes the temperature change of different equipment during operation in the research process; The R & D progress data represents the data processing volume during the R & D process.

[0008] Furthermore, research process status indicators are set based on the characteristic data. The specific method is as follows: Based on the set of characteristic data of the research stage where research objective i is located, determine the actual operation time of the equipment during the research process. Each continuous actual operation time is recorded as the time period of a research stage. Using the formula by synthesizing the characteristic data of each time period Indicates the status index of the research process. Here, m represents the m-th time period, t represents time, A(m, t) represents the status index of the research process that changes with time t in the m-th time period, sj(m, t) represents the data processing volume of the research process that changes with time t within the m-th time period, which is directly obtained by recording the data processing task volume (such as calculation volume, storage volume) completed within a specific time period through device logs or data processing systems. wd(m, t) represents the change value of the local temperature of the device that changes with time t within the m-th time period, which is obtained by collecting the temperature data of the device during operation in real time through a temperature sensor and processing it by a data collection module. ny(m, t) represents the energy consumption value that changes with time t within the m-th time period, and the electrical energy consumption of the device is measured in real time using an electric energy meter or an energy monitoring device. Represents the weight coefficient of the data processing volume. Represents the weight coefficient of the temperature change value. Represents the weight coefficient of the energy consumption value.

[0009] Furthermore, filter out the network transmission performance data within the stable time range. The specific method is as follows: Set the stable range of the research process status index based on historical data, denoted as (a, b). Calculate the research process status index of each research objective in each time period, compare it with the stable range of the research process status index, and filter out the time interval in which the research process status index of each time period is within the stable range of the research process status index, denoted as the stable time range. Retain the network transmission performance data during the process of researching information technology achievements of each research objective within the stable time range of each time period, and denote the retained network transmission performance data as stable transmission performance data.

[0010] Furthermore, establish a prediction model for the characteristic data of the scientific and technological achievements research process. The specific method is as follows: Construct a prediction model based on the LSTM network. The LSTM network structure includes an input unit, an output unit, a forgetting unit, a state unit, and a recurrent unit. The new input information is retained in the state unit after being determined by the input unit. The state unit is connected to the recurrent unit, enabling the information in the state unit to perform a linear self-recurrence through the recurrent unit. By presetting the weight value of the forgetting unit, the output unit controls whether to output. Determine the output value function of each unit based on the sigmoid function, the input vector, the hidden layer vector, and different weight coefficients. The weight coefficients include: different bias weight coefficients for each unit, input weight coefficients, and recurrent weight coefficients of the input unit, the forgetting unit, and the output unit. Determine the update method of the internal state of the LSTM model based on the output value functions of the input unit, the forgetting unit, and the output unit.

[0011] Further, determine the update method of the internal state of the LSTM model based on the output value functions of the input unit, forget unit, and output unit. The specific method is as follows: Calculate the output values of the input unit, forget unit, and output unit. Use the forget unit to control the degree of information retention of the state unit at the previous time step, use the input unit to control the degree of new information flowing into the state unit, update the value of the state unit, and use the output unit to control the information output of the state unit to the hidden state. Among them, through, use the formula represents the update method of the state unit, where t represents time, represents the value of the state unit at time t, represents the output of the input unit at time t, represents the output of the forget unit at time t, represents element-wise multiplication, and tanh represents the hyperbolic tangent function, represents the input value at time t, represents the input weight coefficient, represents the hidden state at time t, represents the hidden state weight coefficient, represents the bias weight coefficient of the state unit, and these weight coefficients can be determined according to specific situations and requirements, represents the value of the state unit at the previous moment of time t.

[0012] Further, the hidden state specifically includes: Use the formula represents the hidden state, where t represents time, represents the hidden state at time t, represents the value of the state unit at time t, represents the input value at time t, represents the sigmoid function, represents element-wise multiplication, and tanh represents the hyperbolic tangent function, represents the input weight coefficient of the state unit, input weight coefficient, represents the hidden state weight coefficient, represents the bias weight coefficient of the output unit, and these weight coefficients can be determined according to specific situations and requirements, represents the hidden state at the previous moment of time t.

[0013] Further, train and optimize the LSTM prediction model using the network transmission performance data within a stable time range. The specific method is as follows: Extract part of the stable transmission performance data as training data, use the remaining stable transmission performance data as control data, use the training data as the input data of the LSTM prediction model to train the LSTM prediction model, compare the output value of the LSTM prediction model with the control data to determine the actual prediction error value, preset an error threshold R, and when the average error value between the output value of the LSTM prediction model and the control data within the time range j is less than R, it is determined that the training of the LSTM prediction model is completed.

[0014] Furthermore, set an information research fluctuation index to predict the stability of the research process. The specific method is as follows: Use the formula to represent the information research fluctuation index, where p represents the information research fluctuation index, j represents the prediction duration of j, t represents time, g(t) represents the number of error bits at time t, which is obtained by counting the number of bits with verification failures during transmission through a network monitoring tool (such as Wireshark) or a hardware device (such as a router), f(t) represents the number of lost data packets at time t, which is obtained by comparing the data packet sequence numbers at the sender and receiver and counting the number of unacknowledged or lost data packets, v(t) represents the total number of transmitted bits at time t, which is obtained by using a network traffic counter (such as the SNMP protocol) to count the total number of successfully transmitted bits in real time. represents the weight coefficient of the number of error bits, represents the weight coefficient of the number of lost data packets, represents the weight coefficient of the total number of transmitted bits; Set an information research fluctuation index threshold P. When the predicted information research fluctuation index p is less than the threshold P, it means that the research process of the current information technology achievement is proceeding stably. When the predicted information research fluctuation index p is greater than or equal to the threshold P, it indicates that there may be large fluctuations or instability in future network transmission performance.

[0015] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are as follows: 1. By establishing a prediction model for the characteristic data of the research process of scientific and technological achievements, the present invention extracts part of the stable transmission performance data as training data, and the remaining data as control data to test the prediction accuracy of the prediction model, ensuring the reliability of the prediction results. By setting an error threshold and a time range to judge the completion of the training of the LSTM prediction model, the problems of overtraining or under-training are avoided, the efficiency and effect of model training are improved, and effective prediction of the characteristic data of the research process of scientific and technological achievements is realized; 2. Based on the characteristic data set at the research stage of the research objective, the present invention determines the actual running time of the device and sets the research process status indicators, which helps to more intuitively understand the status of the research process. By setting the stable range of the research process status indicators based on historical data and screening out the stable time range, key network transmission performance data can be retained, providing a basis for subsequent decision-making. 3. By predicting the network transmission performance data within the future time period and obtaining the predicted value of the information research fluctuation index by synthesizing the predicted value of the network transmission performance data, the present invention can quantitatively evaluate the stability of the information technology achievement research process, helping researchers determine whether to continuously promote the research work. According to the prediction results, it is helpful to more accurately evaluate the demand for research process resources by future scientific and technological achievements, so as to optimize the resource allocation in advance, improve the stability and efficiency of the research process, and promote the transformation of information technology achievements. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 Shows a program block diagram of a scientific and technological achievement prediction and evaluation system based on big data; Figure 2 Shows a method step diagram of a scientific and technological achievement prediction and evaluation method based on big data; Figure 3 Shows a method step diagram for screening the stable time range. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0019] As Figure 1 、 Figure 2 、 Figure 3 shown, a scientific and technological achievement prediction and evaluation system based on big data specifically includes the following: a research and development process data collection module, an information technology achievement prediction module, and an information technology achievement comprehensive evaluation module; Among them, the R & D process data collection module realizes the automatic synchronization and update of information technology achievement data by establishing a data interface with the scientific research project management system. The historical research data of information technology achievements is determined through the scientific research project application data, scientific research project approval data, and research process data of historical information technology achievements. The historical research data includes the key elements of the research plan for big data scientific research achievements: research objectives and technical routes. The research objectives to be achieved for information technology achievements are determined, the technical routes designed to achieve different research objectives are determined, and the time period from preparation to completion of each research objective is recorded as a research stage. Using Internet of Things technology, the characteristic data and network transmission performance data of each research stage in the R & D laboratory are monitored in real time.

[0020] The characteristic data of each research stage includes the equipment operation data, equipment status data, and R & D progress data of the R & D laboratory. Among them, the equipment operation data includes the actual operation time of the equipment within a specific time period and the energy consumption of the equipment during operation. In this embodiment, the energy consumption only considers the consumption of electric energy. The equipment status data includes the temperature change of different equipment during operation in the research process, and the R & D progress data represents the data processing volume in the R & D process.

[0021] Sort the research objective set according to the research order of scientific and technological achievements, and denote the research objective set in the research process of the obtained scientific and technological achievements as , where Y(u) represents the research objective set of scientific and technological achievement u, i represents the research objective numbered i, and n represents that scientific and technological achievement u needs to complete n research objectives in total. Denote the characteristic data set of the research stage where research objective i is located as , where yx(i) represents the equipment operation data set of the R & D laboratory in the research stage represented by the i-th research objective, xh(i) represents the material consumption data set of the R & D laboratory in the research stage represented by the i-th research objective, and zt(i) represents the equipment status data set of the R & D laboratory in the research stage represented by the i-th research objective; During the research process of information technology achievements, network transmission performance data that appears during the transmission process is recorded through network monitoring tools or device logs. The network transmission performance data includes the number of error bits, the total number of transmitted bits, and the number of lost data packets.

[0022] The information technology achievement prediction module is used to establish a prediction model for the characteristic data of the scientific and technological achievement research process. The historical characteristic data collected in different research stages is used as the input data set of the prediction model. Through the prediction model, the prediction of the characteristic data of the scientific and technological achievement research process is realized. The real-time characteristic data collected in different research stages is used as the control data set to test the prediction accuracy of the prediction model; Based on the characteristic data set of the research stage where research objective i is located, determine the actual operating time of the equipment during the research process. Record each continuous actual operating time as a time period of a research stage. Set the research process status indicators by integrating the characteristic data of each time period. The specific formula for the research process status indicators within each time period of research objective i is as follows: ; Among them, m represents the m-th time period, t represents time, A(m, t) represents the research process status indicator that changes with time t in the m-th time period, sj(m, t) represents the data processing volume of the research process that changes with time t within the m-th time period, which is directly obtained by recording the data processing task volume (such as calculation volume, storage volume) completed within a specific time period through the device log or data processing system. wd(m, t) represents the local temperature change value of the device that changes with time t within the m-th time period, which is obtained by collecting the temperature data of the device during operation in real time through a temperature sensor and processing it by the data collection module. ny(m, t) represents the energy consumption value that changes with time t within the m-th time period, and the electrical energy consumption of the device is measured in real time using an electric energy meter or energy monitoring device. represents the weight coefficient of the data processing volume, represents the weight coefficient of the temperature change value, represents the weight coefficient of the energy consumption value.

[0023] Set the stable range of the research process status indicators by integrating historical data, denoted as (a, b). Calculate the research process status indicators of each research objective in each time period, compare with the stable range of the research process status indicators, and screen out the time intervals where the research process status indicators in each time period are within the stable range of the research process status indicators, denoted as the stable time range. Retain the network transmission performance data during the process of researching information technology achievements for each research objective within the stable time range of each time period. Denote the retained network transmission performance data as stable transmission performance data.

[0024] Construct a prediction model based on the LSTM network. The LSTM network structure includes an input unit, an output unit, a forgetting unit, a state unit, and a recurrent unit. The new input information is retained in the state unit after being determined by the input unit. The state unit is connected to the recurrent unit, enabling the information in the state unit to perform a linear self-recurrence through the recurrent unit. By presetting the weight value of the forgetting unit, the output unit controls whether to output. Determine the output value function of each unit based on the sigmoid function, input vector, hidden layer vector, and different weight coefficients. The weight coefficients include: different bias weight coefficients for each unit, input weight coefficients, and recurrent weight coefficients of the input unit, forgetting unit, and output unit; Determine the update method of the internal state of the LSTM model based on the output value functions of the input unit, forget unit, and output unit. The specific method includes calculating the output values of the input unit, forget unit, and output unit, using the forget unit to control the degree of information retention of the state unit at the previous time step, using the input unit to control the degree of new information flowing into the state unit, updating the value of the state unit, and using the output unit to control the information output of the state unit to the hidden state. Among them, the update method of the state unit is represented by the following formula: ; where t represents time, represents the value of the state unit at time t, represents the output of the input unit at time t, represents the output of the forget unit at time t, represents element-wise multiplication, and tanh represents the hyperbolic tangent function, represents the input value at time t, represents the hidden state at time t, input weight coefficient, represents the hidden state weight coefficient, represents the bias weight coefficient of the state unit, and these weight coefficients can be determined according to specific situations and requirements, represents the value of the state unit at the previous moment of time t. In this embodiment, each moment is set to be 0.4 seconds apart.

[0025] Furthermore, the hidden state is obtained from the output value of the output unit. The specific calculation formula is: ; where t represents time, represents the hidden state at time t, represents the value of the state unit at time t, represents the input value at time t, represents the sigmoid function, represents element-wise multiplication, and tanh represents the hyperbolic tangent function, represents the input weight coefficient of the state unit, input weight coefficient, represents the hidden state weight coefficient, represents the bias weight coefficient of the output unit, and these weight coefficients can be determined according to specific situations and requirements, represents the hidden state at the previous moment of time t.

[0026] Extract part of the stable transmission performance data as training data, use the remaining stable transmission performance data as control data, use the training data as the input data of the LSTM prediction model to train the LSTM prediction model, compare the output value of the LSTM prediction model with the control data to determine the actual prediction error value, preset an error threshold R. When the average error value between the output value of the LSTM prediction model and the control data within the time range j is less than R, it is determined that the training of the LSTM prediction model is completed. In this embodiment, the length of the time range j is set to 60 seconds.

[0027] The comprehensive evaluation module for information technology achievements uses the trained LSTM prediction model to predict the collected real-time network transmission performance data. Taking the current time point as the end time point, the network transmission performance data included within the time period of y is used to predict the network transmission performance data within the future j duration, and the predicted value of the network transmission performance data for the j duration is obtained. Set the information research fluctuation index based on the predicted value of the network transmission performance data. The specific formula for the information research fluctuation index is: ; where p represents the information research fluctuation index, j represents the prediction duration of j, t represents time, g(t) represents the number of error bits at time t, which is obtained by counting the number of bits with failed checks during transmission through a network monitoring tool (such as Wireshark) or a hardware device (such as a router), f(t) represents the number of lost packets at time t, which is obtained by comparing the packet sequence numbers at the sender and receiver and counting the number of unacknowledged or lost packets, v(t) represents the total number of transmitted bits at time t, which is obtained by using a network traffic counter (such as the SNMP protocol) to count the total number of successfully transmitted bits in real time, represents the weight coefficient of the number of error bits, represents the weight coefficient of the number of lost packets, represents the weight coefficient of the total number of transmitted bits.

[0028] Set the information research fluctuation index threshold P. When the predicted information research fluctuation index p is less than the threshold P, it means that the research process of the current information technology achievement is proceeding stably. When the predicted information research fluctuation index p is greater than or equal to the threshold P, it indicates that there may be large fluctuations or instability in the future network transmission performance, which may have an adverse impact on the data integrity, accuracy, real-time nature, and overall research progress of the information technology achievement research. At this time, the system issues a warning signal, prompting that corresponding measures should be taken, such as optimizing the network configuration, increasing the network bandwidth, and using a more reliable data transmission protocol, to ensure the smooth progress of the research work.

[0029] As described above, it is only the preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and should be covered within the protection scope of the present invention.

[0030] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific embodiments. Obviously, according to the content of this specification, many modifications and changes can be made. This specification selects and specifically describes these embodiments in order to better explain the principle and practical application of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A scientific and technological achievement prediction and evaluation system based on big data, characterized in that: include: R&D process data collection module, information technology achievement prediction module, information technology achievement comprehensive evaluation module. The R&D process data collection module is used to collect historical research data, feature data and network transmission performance data of scientific research projects in real time during the R&D process, and determine the feature data sets corresponding to different research objectives. The historical research data includes research objectives and technical routes. The time period from preparation to completion of each research objective is recorded as a research phase. The characteristic data and network transmission performance data of each research phase of the R&D laboratory are monitored in real time. Characteristic data includes equipment operation data, equipment status data and R&D progress data; Network transmission performance data includes the number of error bits, total number of transmitted bits and number of packet losses; The information technology achievement prediction module sets the research process status indicators based on the characteristic data, and selects the network transmission performance data within the stable time range. It establishes a prediction model for the characteristic data of the scientific and technological achievement research process based on the LSTM network structure, and uses the network transmission performance data within the stable time range to train and optimize the LSTM prediction model. The comprehensive evaluation module of information technology achievements is used to comprehensively evaluate the stability of the research process of information technology achievements. By predicting the real-time network transmission performance data and setting the information research fluctuation index, the stability of the research process can be predicted.

2. According to the big data-based scientific and technological achievement prediction and evaluation system of claim 1, it is characterized in that: Determine the feature data set corresponding to different research objectives. The specific method is: Sort by the research order of scientific and technological achievements, and record the research target set obtained in the process of scientific and technological achievements research as , where Y(u) represents the research target set of scientific and technological achievement u, i represents the research target numbered i, and n represents that scientific and technological achievement u needs to complete a total of n research targets. The characteristic data set of the research stage of research target i is recorded as , where yx(i) represents the R&D laboratory equipment operation data set of the research stage represented by the i-th research objective, xh(i) represents the R&D laboratory material consumption data set of the research stage represented by the i-th research objective, and zt(i) represents the R&D laboratory equipment status data set of the research stage represented by the i-th research objective; The equipment operation data includes the actual operation time of the equipment in a specific time period and the energy consumption of the equipment during operation; Equipment status data include temperature changes of different equipment under operating conditions during the study; R&D progress data represents the amount of data processed during the R&D process.

3. According to the big data-based scientific and technological achievement prediction and evaluation system of claim 1, it is characterized in that: The research process status indicators are set based on the characteristic data. The specific method is: Based on the characteristic data set of the research phase of the research target i, the actual operation time of the equipment in the research process is determined, and each continuous actual operation time is recorded as a time period of the research phase. The characteristic data of each time period is combined using the formula represents the research process status indicator, where m represents the mth time period, t represents time, A(m, t) represents the research process status indicator that changes with time t in the mth time period, sj(m, t) represents the data processing volume of the research process that changes with time t in the mth time period, wd(m, t) represents the local temperature change value of the equipment that changes with time t in the mth time period, ny(m, t) represents the energy consumption value that changes with time t in the mth time period, The weight coefficient representing the amount of data processing, The weight coefficient representing the temperature change value, Indicates the weight coefficient of energy consumption value.

4. According to the big data-based scientific and technological achievement prediction and evaluation system of claim 1, it is characterized in that: Filter out the network transmission performance data within the stable time range. The specific method is as follows: Based on the comprehensive historical data, the stable range of the research process status indicator is set, denoted as (a, b), and the research process status indicator of each research target in each time period is calculated. The stable range of the research process status indicator is compared, and the time interval within the stable range of the research process status indicator of each time period is screened out, which is recorded as the stable time range. The network transmission performance data of each research target in the process of researching information technology achievements within the stable time range of each time period is retained, and the retained network transmission performance data is recorded as the stable transmission performance data.

5. According to the big data-based scientific and technological achievement prediction and evaluation system of claim 1, it is characterized in that: Establish a prediction model for characteristic data of scientific and technological achievements research process. The specific method is as follows: A prediction model is constructed based on an LSTM network. The LSTM network structure includes an input unit, an output unit, a forgetting unit, a state unit, and a loop unit. New input information is maintained in the state unit after being determined by the input unit. The state unit is connected to the loop unit so that the information in the state unit undergoes a linear self-circulation through the loop unit. The output unit controls whether to output by presetting the weight value of the forgetting unit. The output value function of each unit is determined based on the sigmoid function, the input vector, the hidden layer vector, and different weight coefficients. The weight coefficients include: different bias weight coefficients for each unit, input weight coefficients, and loop weight coefficients of the input unit, the forgetting unit, and the output unit. The updating method of the internal state of the LSTM model is determined based on the output value functions of the input unit, the forgetting unit, and the output unit.

6. The scientific and technological achievement prediction and evaluation system based on big data according to claim 5 is characterized in that: The update method of the internal state of the LSTM model is determined based on the output value function of the input unit, the forgetting unit, and the output unit. The specific method is: Calculate the output values ​​of the input unit, the forgetting unit, and the output unit, use the forgetting unit to control the degree of information retention of the state unit in the previous time step, use the input unit to control the degree of new information flowing into the state unit, update the value of the state unit, and use the output unit to control the information of the state unit to be output to the hidden state, where, by, using the formula Represents the update mode of the state unit, where t represents time, represents the state cell value at time t, represents the output of the input unit at time t, represents the output of the forgetting unit at time t, represents element-by-element multiplication, tanh represents the hyperbolic tangent function, represents the input value at time t, represents the input weight coefficient, represents the hidden state at time t, Enter the weight coefficient, represents the hidden state weight coefficient, Represents the bias weight coefficient of the state unit, Indicates the state unit value at the moment before time t.

7. The scientific and technological achievement prediction and evaluation system based on big data according to claim 6 is characterized in that: Hidden states, including: Using the formula represents the hidden state, where t represents the time, represents the hidden state at time t, represents the state cell value at time t, represents the input value at time t, represents the sigmoid function, represents element-by-element multiplication, tanh represents the hyperbolic tangent function, Represents the input weight coefficient of the state unit, Enter the weight coefficient, represents the hidden state weight coefficient, represents the bias weight coefficient of the output unit, Represents the hidden state at the moment before time t.

8. The scientific and technological achievement prediction and evaluation system based on big data according to claim 1 is characterized in that: Use network transmission performance data within a stable time range to train and optimize the LSTM prediction model. The specific method is as follows: Extract part of the smooth transmission performance data as training data, use the remaining smooth transmission performance data as control data, use the training data as the input data of the LSTM prediction model to train the LSTM prediction model, compare the output value of the LSTM prediction model with the control data, determine the actual prediction error value, preset the error threshold R, and when the average error value between the output value of the LSTM prediction model and the control data within the time range j is less than R, the LSTM prediction model training is judged to be completed.

9. The scientific and technological achievement prediction and evaluation system based on big data according to claim 1 is characterized in that: Setting the information research fluctuation index can predict the stability of the research process. The specific method is as follows: Using the formula represents the information research volatility index, where p represents the information research volatility index, j represents the prediction time j, t represents time, g(t) represents the number of error bits at time t, f(t) represents the number of packet losses at time t, v(t) represents the total number of transmitted bits at time t, The weight coefficient representing the number of error bits, The weight coefficient representing the number of packet losses, The weight coefficient representing the total number of transmitted bits; The information research fluctuation index threshold P is set. When the predicted information research fluctuation index p is less than the threshold P, it means that the research process of the current information technology achievements is progressing stably. When the predicted information research fluctuation index p is greater than or equal to the threshold P, it indicates that the future network transmission performance may fluctuate greatly or be unstable.

Images