STEM-In-Action Fall Follow-Up: Team Vision


We’ve caught up with almost all of our STEM-In-Action teams this Fall to see how their projects are progressing. The last one to follow up with is Team Vision from Colorado. This team of 8th graders was a National Finalist at our National Judging and Educational Event (NJ&EE) back in June and were also awarded the prestigious STEM-In-Action grant. If you haven’t followed along and are new to the SIA grant, each year the U.S. Army Educational Outreach Program (AEOP) STEM-In-Action Grant awards eCYBERMISSION teams up to $5,000 to develop their projects into mature and scalable solutions in their community. eCYBERMISSION itself is a STEM competition for students in 6th-9th grade that’s completely online and gives students the chance to win thousands in grant money. Using this opportunity, Team Vision set out to tackle a very serious topic that not only effects their community, the nation as a whole.

--

We are TEAM VISION from STEM School Highlands Ranch.  We are rising 9th graders and our team includes Gitanjali, Jatin, Azra, and Ceyda. Our team advisor is Mrs. Simi Basu, a Computer Engineer teacher in our school.


We are from Denver, CO. Originally, we are all from different parts of the world with different backgrounds. Gitanjali moved from Tennessee, Jatin is from India, and Azra & Ceyda are from Turkey. This introduced diversity in background and caused different experiences resulting in different thought processes and a wider thinking span.

In 2018, our computer science teacher introduced us to this competition. Gitanjali and Ceyda competed last year with 2 other different teammates. We wanted to pursue and try this back in 2019 and hence Gitanjali and Ceyda reached out to two different team members, Jatin and Azra, who were really interested to compete. We became TEAM VISION.


We are addressing the issue of cyberbullying, which is not a problem only in our community but nationally. Nationally, nearly 1 in 3 students get bullied every day and approximate 34% of the students mention they are cyberbullied their lifetime. Youth suicide rates have been a 40 year high as per Center for Disease Control and Prevention in the United States. This is a national problem. Cyber-bullying seems to be the prevalent reason for most suicides. The current solutions available today are more counseling and social structure rather than technology. There are few technologies available, but most of them just indicate a bullying event and ask the bully to rethink and has just a bank of words to rely on. In addition, these technology solutions are available just for one medium such as either an exclusive chat app. Some of the newer ones have started using AI models, but they are elementary AI models and do not use an External service with language interpretation. This makes these solutions very inefficient and ineffective.

With research, expert interviews about cyberbullying, and feedback on the need for a solution, we decided that we would create a service that would help the bully and the victim in a cyberbullying event irrespective of the platform they are using.


We came up with a complete solution where the service “KINDLY” (our proposed solution)  is available on a variety of different platforms such as application, SMS, chat apps, and other social media. It is more accurate than any other app out in the market since it uses the latest machine learning-based algorithms to be able to identify the most commonly used words. KINDLY uses the latest AI technology and Natural Language Processing Algorithm and sentiment analysis to identify a bullying event.  The messages are scored not only based on their category, but also on the “likeliness” of being in the category and whether it was a negative or positive connotation. It is intended to be a solution for others to learn from and not necessarily as a punishment for students or those who bully.

In addition, the solution notifies a trusted adult who can come to the rescue, while at the same helping the bully understand and take the appropriate reactions to the actions. We came up with "KINDLY" as an extension that trains itself based on the messages and to adapt from person to person, as we understand that every student is different, Kindly is also intended to judge differently based on the relationship between the two users, which will be learned from observing the conversation and learning from the interactions.

We have created an initial prototype of a standalone app called "Kindly" and took initial feedback on the functionality. We have also created a service that can be called from any social media platforms and have tested it.

We have identified a four step plan to implement or roll-out KINDLY with the STEM-In-Action that was awarded to us. They are:

1) Enhancement of the solution and continuous improvement : The goal of this phase is to enhance the service to full functional solution for users along with documentation.

2) Solution Access to the community:  The goal of this phase is to provide an easy way for users to download the solution and start using it

3) Awareness and Knowledge: The goal of this phase is to increase the use of the solution, reduce cyber-bullying events and to spread awareness on cyberbullying and impact to our lives.

4) Collection of feedback and measurement: The goal of this phase is to find out how the solution is able to reduce cyber-bullying instance and iteratively improve the efficiency of the solution.

--

Team Vision has taken an important topic is today’s society and tried to help the cause by changing the narrative for kids that are victims of bullying, as well as the bully alike. They’ve approached this serious issue from a new perspective and we know they will continue to be successful in helping lots of people alleviate this problem in their communities across the country.

We can’t wait to see what’s next for Team Vision.

-Mission Control

Comments

Popular Posts