Research on the Prediction Method of the College Professional Admission Scores

This article is based on the web crawler technology to obtain the admission scores of some colleges and universities in various provinces, and integrate it with the existing college professional admission scores. The edit distance algorithm and the mean difference interpolation method are used to clean the data of similar duplicate records and missing records respectively. Using BP neural network theory and support vector machine algorithm to construct a network model with 9 characteristics of school, major, province, batch, discipline, average score, highest score, provincial control line and year as input, and the lowest score as output. Using labeling coding to encode non-numerical data to normalize the data. Preset 1% and 2% minimum score error bands and compare the pass rates predicted by two models in different preset error bands. Obtained that both models meet the preset 2% error, But within the 1% error band, the network model constructed based on the SVM algorithm has higher accuracy and the forecasting effect is better.