Data-Mining

During my undergrad, I implemented a music recommendation system based on music digital track analysis. However, it's time for me to use text mining technology on lyrics to upgrade that project. Goals: (1)build a music mood(happy or sad) classifier based on lyrics analysis (2)what words and their distributions are in different mood categories? (3)How are the key words change in songs for the recent years? Project evaluation: (1)data collection: the training data and validation data will be collected from the largest lyric database on Lyricwiki.org (2)feature selection: the most common feature type to consider are BOW(bag of word) and POS(part of speech) combined with stemming using word-net (3)Training model : SVM, Naive Bayes using grid search method. (4)data visualization for goal two and three This project will be done using python on jupyter notebook. reference: Hu, X. (2010). Improving music mood classification using lyrics, audio and social tags (Doctoral dissertation, University of Arizona).

My Final Year Project (Using software such as Pycharm, Git and Jupyter Notebook; and also programming languages of Python, HTML, CSS and Javascript)

This repo houses the jupyter notebook presentation for my workshop on Twitter Data Mining, At PyCon NG, Abuja.

This collection of Jupyter notebooks documents my learning as I progress through the Data Mining and Predictive Analytics book.

My basic data mining container

Just my Jupyter Notebook!