SafeConnections

Links

• No demo was used for this project. All ideas were original. 

• GITHUB LINK

Inspiration:

The inspiration for SafeConnections stems from a profound desire to combat human trafficking, particularly in regions where it's most prevalent. Southern Africa, where millions of children fall victim to forced labor and trafficking, served as a crucial motivator. Our team aimed to leverage technology to create a safer online environment for children, providing parents with tools to monitor and protect their kids from online predators and traffickers.

What it Does:

SafeConnections is a secure messaging platform designed to prevent human trafficking. It uses unsupervised machine learning to detect and flag suspicious conversations that may indicate trafficking attempts. Parents receive notifications about these potential threats, helping them take swift action to protect their children. The platform ensures a safer online space for kids to communicate without fear.

How We Built It:

We utilized a combination of technologies and tools to bring SafeConnections to life. The front-end was built with HTML, CSS, Java, React.js, and Apps Script, with design assistance from Figma for the user interface. The back-end was developed in Python, incorporating several libraries for data handling and machine learning. Specifically, we used pandas for data manipulation, TfidfVectorizer for converting text data into numerical format suitable for machine learning algorithms, joblib for model persistence, and oauth2client for handling OAuth 2.0 authentication. Our machine learning pipeline included training models to analyze and predict user data, enhancing the app's functionality and user experience. For integration, Google Sheets served as an intermediary to connect the front-end with the Python backend, allowing seamless data flow and processing. Additionally, we implemented the Twilio API for notification services. The project is deployed and hosted on GitHub Pages.

Challenges We Ran Into:

Building SafeConnections presented several challenges. Ensuring seamless communication between the front-end and back-end via Google Sheets required meticulous coordination. Implementing and fine-tuning the unsupervised machine learning model to accurately detect suspicious activity was complex and time-consuming. Additionally, designing a user-friendly and intuitive interface that met the needs of both children and parents necessitated significant iteration and feedback.

Accomplishments That We're Proud Of:

Functional Prototype: Developing a working prototype within a limited timeframe that effectively demonstrates our concept.

Machine Learning Integration: Successfully incorporating machine learning to enhance the platform's safety features.

Social Impact: Creating a tool that has the potential to make a real difference in the fight against human trafficking and online exploitation.

What We Learned:

Collaboration: Effective teamwork and communication are crucial, especially when integrating diverse technologies.

Technical Skills: Enhanced our skills in front-end and back-end development, machine learning, and API integration.

Problem-Solving: How to approach and solve complex problems under time constraints, improving our ability to innovate and adapt quickly.

What's Next?:

Looking ahead, we plan to…

Enhance Detection Accuracy: Further refine the machine learning model to improve accuracy in detecting suspicious activities.

Expand Features: Introduce additional safety features and tools for parents and children.

Broader Deployment: Work towards deploying SafeConnections in regions most affected by human trafficking, starting with Southern Africa.