This repository is about a lecture which is named "Web Mining" at Mugla Sitki Kocman University. There are 3 sections:
- Apriori
- Midterm
- Final
In this section, there is an algorithm which is called Apriori and it was written in Java. The algorithm is tested with 3 different transaction lists and also there is a report by reported in a file with txt extention.
The output image of my Apriori algorithm:
The midterm has 2 questions and ones is about Apriori algorithm evaluation and another one is about naive Bayesian classifier. In this repository, there are both questions and my answers in PDF files.
Actually, the final exam is a course project. I've joined a partnership with my classmate and we worked on sentiment analysis. To make an analysis, we created a web crawler for eksisozluk which is a reddit-like website. Then we mined opinions about a Turkish series on Netflix. Finally, we reached up to 73% accuracy rate, approximately. The mean accuracy was 61%, too.