Prediction Algoritnm And System Implementation Of College Entrance Examination Professional Score Line Based On Parameter Optimization Artificial Neural Network

The college entrance examination volunteer filling contains many factors that need to be considered by the candidates.However,because candidates and parents do not know much about the process of volunteering and related knowledge,there will often be problems such as high scores and low scores.In recent years,in response to the problems encountered by candidates,a lot of apps and webpages on the college entrance examination volunteer filling has been developed.But these apps and webpages have the problem of download trouble and the accuracy of prediction is relatively low.Additionally,few webpages and apps take into account the many information that candidates need on the page.This paper focus on accurately forecasting the academic professional score line.The professional score line of the college entrance examination is predicted by the recurrent neural network and the multi-layer feed-forward artificial neural network,respectively.The recurrent neural network has the ability to memorize long-term and short-term information,essentially a notion of long short-term memory(LSTM).And the multi-layer feed-forward artificial neural network optimizes the parameters of the back-propagation(BP)deep neuron network.The accuracy of the prediction is verified and analyzed by the K-Fold Cross-Validation.At the same time,the WeChat Mini Program is used to display the professional score line of the past years as a carrier.The main research contents are as follows:(1)Aiming at the problem that the score line of major admission in colleges and universities over the years is determined by many factors,a prediction method of the score line of college entrance examination major based on the LSTM model is proposed.The dataset of college entrance examination scores from 2016 to 2019 is divided into test sets and training sets according to a 9:1 ratio.And the data of continuous two years are entered into the LSTM model.Then,the admission score line of the college major is predicted.Compared with the prediction results of the gray prediction method on the admission score line of the major,the maximum accuracy rate of the same major increased from 96.15%to 98.90%.Moreover,the average accuracy rate increased from 93.75%to 98.29%,and the overall prediction accuracy was greatly improved.(2)The LSTM is focused on the problems with the relationship between time series.For the problem of the weak causal relationship between the data sets required for prediction with the data of straight two years,a method based on BP deep neural network is proposed.The four years from 2016 to 2019 are grouped according to a 9:1 ratio based on the K-fold cross-validation method.The dataset affecting the professional score line for any straight two years is input as an input neuron.The factors related to the relevant parameters of the institution for any straight two years are input as another input neuron.The prediction results of the professional score line are output.Compared with the LSTM prediction,the average accuracy increased from 98.29%to 99.06%,and the minimum absolute value of the error decreased from 7.06 to 5.31.In the end,it is found that under the premise of increasing the constraint range and the input influencing factors,the accuracy of the professional score line prediction is significantly higher than that of the LSTM model.Therefore,the learning ability of the model is enhanced,and the generalization ability is the best.(3)To realize the use of the parameter-based optimization of the BP deep neural network prediction scheme,the WeChat Mini Program is designed to predict data based on the score line of the college entrance examination.The expected score line of the college and the professional enrollment score line of the previous years are updated.For college entrance examination score prediction,the WeChat Mini Program platform has been built to facilitate students to view the score lines of the target school and the corresponding majors.At the same time,users can also retrieve the provincial control line,professional score line,relevant information of target colleges and professional information in Shaanxi Province over the years.It is more convenient for students to grasp the relevant information of the target school and the future employment status of the major on time,thus more comprehensively improving the accuracy of students’application for the exam.