DETOX

The Home Page of the Web Application
About the Web Application
Enter the text and youtube video id
Analysis of the Text and Youtube video comment Section
Chart.js data vis

Demo (with audio)

Inspiration

Since the theme of the event was Ethical Hack, we wanted to make something that could help reduce the negativity one finds around the internet. People from different parts of the world have the opportunity to connect through the internet, yet sometimes it is misused. We wanted to prevent that, so we build a multi-lingual text analysis tool that can do sentiment analysis of text as well as give a summary of the comment section of a youtube video

What it does

Our Web application performs Sentiment analysis on the text entered by a User and tells them the levels of Toxicity, threat, Insult, and Profanity present in the text. The user can also enter the Youtube id of a video and receive a summary of the toxicity levels present in the comment section of the video.

How we built it

The final application is built using Flask for the backend and HTML, CSS, and JavaScript for the frontend. For fetching the youtube comments, the Youtube API is used. For performing sentiment analysis, the Perspective API of google is used which returns the toxicity levels of various text and aids in sentiment analysis. The API supports multiple languages, thus helping us achieve the goal of helping people from different languages and regions.

Challenges we ran into

The team first made the skeleton of the Project using Express.Js and TyeScript and had hosted the site in the server as well but we faced issues in connecting the frontend skeleton made to the Flask Backend. We decided on using the Flask Backend because the APIs implementation was present in python and few team members were comfortable using Python for performing the Backend. In the end, we decided to complete the whole application using Flask and changed the framework used for Frontend. Interacting in the team was made a bit difficult due to the time difference but the team overcame it to come up with this application. Using Google's API system was a first for some team members, but it was learned using tutorials, blogs, and documentation. The APIs used in the project were being used for the first time by the team members, so the documentation was read in order to assist in easy application.

Accomplishments that we're proud of

We are glad that we are able to present a complete working web application that performs the functions as per our vision. We intend to continue working on it to make it better!

What we learned

The project surely had a lot of learning for the team members. There was an international collaboration, got to learn about new tech stacks, ran into problems, and arrived at timely solutions as well. Gained experience in reading the documentation for APIs and implementing them efficiently.

What's next for DETOX

In the future, we plan to improve the web application. Add more features for text and sentiment analysis along with a text summary. We intend to add the feature of generating a youtube video summary using the captions generated by videos. In the future, the team members are interested to get into Natural language processing, so we will be building our own models for sentiment analysis as well.

Looking forward to all the Learning