DKE kaggle competition, a competition among students about predicting weather a link is clickbait or not. This competition was part of a year 3 course named Data Analysis at Maastricht University. This was a collaboration between Xavier Weber and Jeroen Vermazeren. Final grade 7.
Original Description below:
This is the official Kaggle page for the DKE/KEN3450 "Data Analysis" competition. Overview
In this competition, you will work with real data from some "clickbait" articles. We are interested in predicting what drives the popularity and clicking rates of such articles. Data are supposed to be clean (but you know how to guarantee that right?). We will be using features related to the following aspects:
words: number of words in title, content, word length, etc.
links: number of links to other articles
digital media: number of images, videos
publication time: weekend or not
keywords: number of keywords, article category, ...
NLP features: title/text polarity/subjectivity, ...
Our goal is to predict the number of shares of each article. Basic Info
Use tab "Data" to find relevant information about the dataset you are going to be using.
While you can submit a solution based on what we learned during the course, feel free to use your favorite programming language and/or your preferred machine learning methods.
Files
- train.csv - the training set : contains 29734 rows (1 header row + 29733 data points)
- test.csv - the test set: contains 9912 rows (1 header row + 9911 data points)
- sample_sol.csv - a sample submission file in the correct format
- derived.csv - the released solution after assignment was graded