Applied Deep Learning Using Health Data

3 units

Data are now available to social scientists in a way and quantity that has never existed before, presenting unprecedented opportunities for advancing social research and practices through state-of-the-art data analytics. On the other hand, dealing with extensive, complex, unconventional big data (e.g., free text, image, video/audio recording) requires revolutionary analytic tools only made available during the past decade. Artificial intelligence (AI), characterized by machine and deep learning, has become increasingly recognized as an indispensable tool in modern social and behavioral sciences. For example, AI methodologies have been applied to enhance the effectiveness of diagnosis and prediction of disease conditions, advance understanding of human development and functioning, and improve the effectiveness of data management in various social and human services. As a subdomain of AI, deep learning is based on artificial neural networks in which multiple (deep) layers of processing are used to extract higher-level features progressively from data. This layered representation enables modeling more complex, dynamic patterns than the traditional machine learning (which sometimes are called shallow learning as in contrast to deep learning), which finds its utility in analyzing the big data-data massive in scale and messy to work with (e.g., unstructured texts, images, audios, and videos). This course contributes to the overarching goal of training next-generation researchers in modern data analytics. It aims to equip students with the core knowledge and essential skills to apply deep learning models to address real-world problems. Through the course, students will familiarize themselves with computer programming in data science, learn state-of-the-art deep learning models, and apply them to social and behavioral questions. In addition, one essential field of deep learning applications is assisting decision-making through identifying patterns and trends, improving prediction precision, and automating evidence collection, synthetization, and dissemination. Students who master deep learning tools will be at the frontier to leverage the power of AI in analytics and practices.