• 

• GIF

Emojize Project

Inspiration

The inspiration behind the creation of Emojize was the growing awareness of the negative impact of social media on mental health, especially in the context of the COVID-19 pandemic. The team recognized the increasing demand for digital health platforms that address mental health issues, and we aimed to contribute to this important area. Social media platforms often exacerbate mental illnesses like depression and anxiety, and Emojize sought to provide a solution that promotes positive communication while leveraging deep learning to analyze and improve online interactions.

Learning Experience

Throughout the development of Emojize, we gained valuable insights into the intersection of technology and mental health. We learned about the importance of responsible AI and the ethical considerations surrounding content analysis, moderation, and recommendation systems. Additionally, we deepened our understanding of deep learning, specifically LSTM networks, and their applications in understanding the semantic meaning of user posts.

Project Development

Emojize was developed collaboratively by a team of three individuals, with a primary focus on creating a front-facing model of the platform. We utilized deep learning algorithms, particularly LSTM networks, to analyze user-generated content, flag inappropriate posts, and recommend "Emoji Lobbies" based on the connotations of the text. The platform encourages users to express themselves through memes and pre-written responses, fostering a supportive and non-toxic environment.

Key Features

• Content Analysis: LSTM networks analyze user-generated content to assess its emotional tone and context, allowing the platform to understand the semantic meaning of posts.
• Inappropriate Content Detection: Emojize employs AI algorithms to detect and flag inappropriate posts, ensuring a safe and respectful online environment.
• Emoji Chat Rooms: Users are recommended Emoji Lobbies based on the emotional connotations of their posts, fostering empathetic and relevant conversations.
• Meme and Response Integration: Emojize encourages expressive communication through memes and suggested pre-written responses, enabling users to connect on a personal level.
• Impact on Mental Health: Our platform addresses the negative impact of social media on mental health, offering a space for positive interactions and support.

Challenges Faced

While developing Emojize, we encountered several challenges:

• Giphy API Integration: Connecting the Giphy API to create an infinite and searchable gif library, similar to the one on iPhones, required overcoming technical hurdles.
• Personal Chatbox Feature: Implementing a personal chatbox feature within the front-facing website proved to be a complex task and was not successfully integrated.
• Connecting Frontend and Backend: Given our individual skillsets in frontend and backend development, we faced struggles in combining our Frontend website to our data on the Backend. Succeeding the final judging of the hackathon, there are plans to finalize the development of the platform.
• Optimizing LSTM Network: Changes to both model structure and hyperparameters can greatly affect the accuracy of the model. In addition, overfitting is an issue. Time is a big challenge given the time it takes to train the model over long epochs.

Positive Impact on Health Industry

Emojize aims to make a positive impact on the health industry by providing a platform that reduces the negative effects of social media on mental health. By encouraging expressive and empathetic communication through memes and emojis, and by leveraging AI to detect and address inappropriate content, Emojize can contribute to decreasing the number of individuals seeking medical attention for mental health issues. Our research supports the potential of Emojize as a valuable tool for improving mental well-being and reducing the burden on medical institutions.

In conclusion, Emojize represents a novel approach to addressing mental health concerns exacerbated by social media. With a focus on responsible AI, positive communication, and user centered design, our platform has the potential to make a meaningful impact on the digital health landscape.

Built With

• css
• html
• javascript
• keras
• lstm-network
• python
• tensorflow
• vector-embedding