What it does
This program can take in the input of a news article title and analyse the words in it to determine whether the article has positive or negative news.
How I built it
The program is made with Python, with the help of two borrowed data sets [Minqing Hu and Bing Liu. "Mining and Summarizing Customer Reviews." Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2004), Aug 22-25, 2004, Seattle, Washington, USA.]
Challenges we ran into
One challenge was to omit the apostrophes that some headlines have before analysing it. So, to fix that, a function was created to detect apostrophes and eliminate them before move to the analysing.
What I learned
I have learnt about how to incorporate data sets into my programs and utilise them to obtain desired results.
What's next for ArticleAnalyser
The first step would be to improve the accuracy of the analyser. Certain article titles that do have negative content doesn't get recognized as a negative article since the words of the title are not included in the pre-determined list of negative words.
Log in or sign up for Devpost to join the conversation.