Speaker: Stephen Wan, Data61 Seminar Date: Tuesday May 30 12:00pm Brief abstract: Public social media data can complement other data sources, such as interview and questionnaire approaches, to produce a richer data set for understanding community needs, opinions and sentiment. This presentation will provide an overview of CSIRO Data61 research on social media analytics, with a focus on business decision applications and applications in computational social science. These applications help motivate our Natural Language Processing and Machine Learning work in text analysis for social media, sometimes referred to as “noisy” or user-generated text, vertex labelling, and explorations of the degree to which social media is representative of the wider population. Short Bio: Stephen Wan is a computer scientist specialising in computational linguistics, an inter-disciplinary field that draws on both linguistics and computer science. His research employs natural language processing and text mining methods, in conjunction with machine learning, information retrieval and human computer interaction techniques. With over 15 years of experience in computational linguistics, he leads a team of software engineers and researchers at CSIRO Data61 to build information management systems for a variety of text data types. A common element in these systems is the need to extract actionable nuggets of information that will help with a user’s task. His team has developed systems used by both the government and private sector for areas such as business intelligence, social media monitoring, and information systems for scientific literature.