Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/76982
Title: Machine learning for mathematical question difficulty classification
Authors: Pang, Jarald Qi Kai
Keywords: DRNTU::Engineering::Computer science and engineering
Issue Date: 2019
Abstract: This project is an experimental study on how machine learning models can be used for classification of GCE ‘A’ Level mathematical questions. Two levels of classification are carried out. First, the classification of questions to their respective topics and second, the classification of the questions to their difficulty level. The report will contain detailed explanations of the steps gone through during the experiment. The grading metrics used in this experiment are F1 Score, Precision, Recall and Accuracy. For data pre-processing three text vectorization methods, count vector, word level TF-IDF and N-gram level TF-IDF were used and tested. Four machine learning methods, Support Vector Machines, Naïve Bayes, Random Forest and Extreme Gradient Boosting, were then used to classify the data to their respective topic. Analysis was then done on the models’ performance on each topic. The same 4 machine learning methods were then again used to classify the difficulty of each question using the vectorized question and predicted topic. A final analysis was then done on the performance of the models in difficulty classification.
URI: http://hdl.handle.net/10356/76982
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP_Jarald_Pang.pdf
  Restricted Access
2.28 MBAdobe PDFView/Open

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.