Project for machine learning classes at university.
Subject: Systems analysis and decision support methods in Computer Science.
The lecturers implemented unit tests and other files. The task was to implement missing machine learning alghoritms in content.py file, using only numpy library.
- to see results, run main.py script,
- to read more about Naive Bayes visit wikipedia,
- to read more about k-NN visit (no surprise...) wikipedia.
Classify each text into one of four topic-related groups based of words present in the text.
Since we assume that words in one text are unrelated the algorithm is called Naive Bayes.
In case of k-NN alghoritm we measure the distance as the number of words that two text differ by.
As the result of main.py you should see a buch of charts comparing the tested parameters and alghoritms.
Stay tuned for my upcoming projects!
