碩士 / 國立高雄應用科技大學 / 電子工程系碩士班 / 102 / Purpose: To analyze the influence factors for the English learning performance of the elementary school students.
Materials and methods: Questionnaire statistics for elementary school students were used to analyze the influence factors of the English learning performance. The passing score for the lower graders and the middle graders are 90 points; 80 for the higher graders, respectively. The missing data were corrected by using the expectation maximization (EM) method. The least absolute shrinkage and selection operator (LASSO) were used to select the predictive factors to assist the logistic regression model and artificial neural network (ANN) to predict the influence of elementary school students in English learning performance. Statistical analysis was performed using the R², ROC curve, scaled Brier score (S-BS), root mean squared error (RMSE) and mean absolute percentage error (MAPE) to assess the model efficacy.
Results: The results showed that the influence factors selected by LASSO are family monthly income and mother's education for lower grade students; family monthly income, years of learning English for middle grade students; English learning environment in the family, years of learning English and tutorial for the higher grade students, respectively. System performances for both models were satisfactory with the expected values of three different grade students.
Conclusions: We found that the logistic regression model with LASSO has more excellent performance, which can be used to predict the influence factors for the English learning performance of the elementary school students.
| Identifer | oai:union.ndltd.org:TW/102KUAS0393023 |
| Date | January 2014 |
| Creators | Yea-Wen Kuo, 郭雅雯 |
| Contributors | Dr. Tsair-Fwu Lee, 李財福 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 60 |
Page generated in 0.0019 seconds