We wanted to be part of the solution towards removing the prevalence of opinionated information presented as fact, hopefully our solution will allow people to become better informed on their decisions by giving them a tool that can reliably distinguish subjective and objective texts.
What it does:
Our program takes in text and classifies it as either subjective or objective.
How we built it:
We used machine learning python to read in articles and trained the AI to distinguish between subjective or objective articles.
Challenges we ran into:
We had to learn machine learning in python given the constraints of the hackathon. We had no knowledge of machine learning and little on python. We also struggled with obtaining a categorised data set to train our machine learning algorithm.
Accomplishments that we're proud of:
We were able to create a machine learning algorithm that is capable of classifying input text despite having no background experience in the machine learning field.
What we learned:
Through the progression of this hackathon, we learned about different types of machine learning models, their best use cases and how to manipulate their structures to fit the training data. We also learned advanced python operations along with basic python syntax.
What's next for Classification of Subjective and Objective Texts Using RNN:
In the future we want to implement a more user friendly interface, such as a google chrome extension so users can easily interact with the system we have created.