1. Data Mining Using Health Data
Ensuring good health is crucial for a high-quality daily life. Fortunately, advancements in wearable devices and information technology have made it easier to collect personal health-related data such as step counts, blood pressure, and heart rate. The next challenge lies in figuring out how to effectively utilize such data and extract valuable insights.
In traditional research, individuals were often classified based solely on factors like gender and age, without considering differences in lifestyle or individual health risks. Therefore, our focus is on exploring methods for providing more personalized health services, taking into account factors such as anomaly detection considering potential factors causing individual diseases and diagnostic estimates considering lifestyle differences.
2. Sentiment Analysis on Social Media (SNSs)
Social media platforms host a diverse range of emotions, serving as a reflection of society and potentially influencing others through shared content. Effectively quantifying the emotions expressed in textual content is crucial. Especially in the case of irony or subtly nuanced expressions common in the Japanese language, it requires a sophisticated interpretation beyond simple keyword matching. Additionally, understanding the extent to which these expressions impact other aspects is intriguing. For instance, individuals with extremely negative thoughts or aggressive behavior may face social consequences, experiencing a decrease in online friendships, similar to real-life scenarios. Therefore, our research explores methodologies for accurately quantifying emotions and investigates how emotions influence various aspects.
3. Secure and Empowering Individual-Led Data Utilization
Health data, including diagnostic and screening results, is commonly stored in hospitals or clinics, making it challenging for individuals to manage their own data. Enabling individuals to control and selectively share their health data in response to requests aligns with personal preferences for data management. Therefore, we are dedicated to establishing mechanisms that ensure the security of health data and facilitate easy sharing. This involves efforts such as combining blockchain and public-key encryption methods to enhance security, visualizing anonymized data for self-comparison, and exploring methods for utilizing data.