Inspiration

Nowadays, alot of explicit and adult content is available in terms of images as well as textual content. This type of content is available for anyone to see due to which it could have a bad effect on younger people.

Our inspiration was to create an application that hides/blurs the expicit content and also provide the tone of the content written so as to know before hand and refrain for reading such content if found unsuitable for the reader. This is extended with the accessibility features like Translation of the selected text and Content Recommendation which stores everything securely inside your system and do not track you. It's you who have the total ownership of the saved data.

What it does

A Chrome Extension which allows you to moderate and filter the explicit content present in the webpage. It also provides the sentiment of the text present i.e it identifies the tone of the text and classifies it into Positive , Negative and Neutral.

Through this, a person can ensure that explicit contents will be hidden by default. This can be useful in various situations when you are browsing websites on the public space or using this as a child protection tool. This will also help you understand the sentiment of the articles on the webpage you are browsing. This identifies the tone of the content and then put a badge on the extension icon based on the content.

Apart from this, two extended functionalities help you to get more accessibilities. You can select any text to see it's sentiment score as well as the translation of that text on the page itself. Interestingly, you can also see from the Artificial Intelligence, if the opened content is a recommended article for you or not. The A.I. only see what you are browsing and completely anonymously. It doesn't have any unique identifier to trace you. All content related to the recommendation is store on your system itself with your ownership and is not sent to server anytime.

How we built it

The extension consumes several APIs from Amazon Web Services.

  1. Amazon Rekognition API for detecting the explicit images present in the webpage
  2. Amazon comprehend API for identifying the sentiment of the text content and to provide the recommendation to the pages you are browsing.
  3. Amazon Translate API for providing translation service within the text of a webpage.

Challenges we ran into

The challenges were not big for us. We tried these APIs for the first time, therefore, had to learn from documentation about usage of these APIs. Thanks to Amazon there are plenty of resources available on the internet for this.

Accomplishments that we're proud of

After a long haul of coding trying to make sense of data through various API endpoints of Amazon Rekognition and Amazon Comprehend, we were finally able to connect them in the way we wanted and present the user with a useful summary. And Now, We have working extension.

What we learned

This was the first experience developing for Amazon Rekognition and Amazon Comprehend, and we enjoyed learning developing extensions for Chrome and solving everyday issues with the help of it.

What's next for Mod-Ranger

The next lined-up are several features which include monitoring video ads. This would be a challenging task as we will have to somehow stream the videos to Rekognition.

Built With

Share this project:
×

Updates