Research Paper: Battling cultural bias within Hate Speech Detection | AI, ML Project
Lead : Shlok Bhattacharya
Project Description : Battling cultural bias within hate speech detection: An experimental correlation analysis
Status : Approved and awaiting publication in the peer-reviewed Journal of Emerging Investigators
i2Vibes – A Social Media App (iPhone & Android)
Lead : Shlok Bhattacharya
Project Description : Social Media App – designed to enhance individual focus and elevate the happiness index
Status : i2Vibes has been officially released in iPhone and Android app store
Podcast with Dr. Abhinav Singh – Sleep heals teen anxiety and depression
Lead : Shlok Bhattacharya
Project Description : In depth conversation with Dr. Abhinav Singh, Medical Director of Indiana Sleep Center talking about how to tackle anxiety and depression in young adults by improving sleep quality
Status : Podcast released on multiple platforms including Spotify, Apple Podcasts, iHeart Radio, Google Podcasts, Amazon Music, and more
Research Paper: Battling cultural bias within Hate Speech Detection | AI, ML Project
Summary: Hate speech detection systems have become essential in the advancing digital world. They limit the dissemination of hateful and offensive language online. However, the machine learning algorithms that provide the basis for these systems struggle to identify hate speech versus clean speech within a cultural context, allowing the growth of cultural bias. Though previous methods had aimed to mitigate the cultural bias of a machine learning model, we attempted to find a new understanding with regard to cultural bias. This study seeks to determine a correlation between increasing the amount of cultural speech used to train the machine learning model and the model’s cultural bias when classifying hate speech and clean speech. Additionally, we hypothesized that increasing the cultural weight of a training dataset will mitigate the cultural bias. To test this hypothesis, we created a unique method named Categorial Weighted Training (CaWT), derived from multiple other methods of previous researchers, to identify a correlation. CaWT involved the creation of multiple culturally weighted training datasets and training a machine learning algorithm against them. From this, the results illustrated minimal correlation between the cultural weight of a training dataset and the model’s performance on cultural speech. However, a significant correlation and negative relationship exist between the cultural weight and the model’s performance on non-cultural speech. This implies that increasing the cultural weight does not affect the model’s cultural bias but decreases the model’s performance on non-cultural speech, proving that a lower cultural weight is ideal within the limitations of our research.
Status: Approved and awaiting publication in the peer-reviewed Journal of Emerging Investigators
Research Paper Link: Click here to read the Research Paper
i2Vibes – A Social Media App (iPhone & Android)
i2Vibes is a transformative social media platform that challenges the societal flaw of conformity. Rather than seeking validation from others, i2Vibes encourages users to embrace their authentic selves. By celebrating individuality and personal growth, it provides a space where users can genuinely express who they are without societal pressures. With a focus on self-discovery and connection with like-minded individuals, i2Vibes aims to reshape our approach to social media, fostering a sense of belonging and acceptance for all.
Status: i2Vibes has been officially released in iPhone and Android app store
Application Link: Click here to download application
Podcast with Dr. Abhinav Singh – Sleep heals teen anxiety and depression
Sleep Heals Teen Anxiety and Depression - Podcast with Dr. Abhinav Singh, The Indiana Pacers Sleep Physician - interviewed by Shlok Bhattacharya
Status: Podcast released on multiple platforms including Spotify, Apple Podcasts, iHeart Radio, Google Podcasts, Amazon Music, and more
Podcast Link: Click here to listen to the Podcast - Spotify