I have a project, where i have to analyse student progression data. The data set consists of all students and their exams in a bachelor degree course (informatics), under a specific exam regulation (2006).
1. I take the progression data from one student and i want to do a prediction of the chance for this student to reach their degree successfully.
2. Optinal: I would like to say when it would be best to repeat the exam if it wasnt passed the first/second time.
I'm new to datamining and knime and i would like to know which nodes would be best to approach this problem.
About question 1, the most traditional way is to use a regression (for a similar project I used a polynomial regression) in the Statistics category. The input data rows must contain the n past samples of the time series and the current value that you want to predict. So it might be that you need to change the layout of your data. You might want to smooth the data with a moving average in the time series category first.
Keeping the same layout of the data (data rows contain n past samples and the desired sample to predict), another possibility is to use a neural network in the Mining category. Here the time series values must be normalized first.
I hope this helps