Topic: Not all machines are created equal: Your next ‘#social’? machine
Presenter: Muhammad Abdul-Mageed
Muhammad Abdul-Mageed is a Visiting Assistant Professor in the School of Informatics and Computing at Indiana University (IU) where he has (created and) taught courses on Social Media Mining, Sentiment Analysis, Computer-Mediated Communication, and Python Programming. Muhammad is a also a Visiting Scholar in the Department of Computer Science, the George Washington University. He acquired a double Ph.D. in Computational Linguistics and Information Science from IU in 2015. Between 2010 and 2012, he was a Visiting Scholar in the Center for Computational Learning Systems, Columbia University. He also currently serves as a member of the standing reviewing committee for Transactions of the Association for Computational Linguistics. Muhammad's interests are at the intersection of machine learning, natural language processing, and social media. He is especially interested in creating more 'social' machines.
Abstract: With the increasing role social media platforms like Facebook, Twitter, YouTube, and Tumbler play in our lives today, the body of data generated by their users continues to grow phenomenally. Accordingly, searches and processing of social media data beyond the limiting level of surface words are becoming increasingly important to business and governmental bodies, as well as to lay web and mobile users. Detection of sentiment, emotion, deception, gender, sarcasm, age, perspective, topic, community, and personality are all valuable social meaning components that promise to be important elements of next generation search engines, web intelligence, and smart devices. The emerging area of extracting social meaning from social media data using computational methods is known as Social Media Mining (SMM). In this talk, I introduce the area of SMM, address practical issues related to learning from social data, and discuss some of the primary computational methods employed for modeling social meaning as occurring in these data.