Dr. Jasy Liew Suet Yan is a Senior Lecturer at the School of Computer Sciences, Universiti Sains Malaysia specializing in sentiment analysis and text mining. Her broader research interests include natural language processing, machine learning, affective computing and human-computer interaction. Dr. Jasy completed her PhD from the School of Information Studies, Syracuse University, USA in 2016. Her dissertation titled “Fine-grained Emotion Detection in Microblog Text”, which explored building machine learning models to detect fine-grained emotions expressed in text was awarded the Syracuse University iSchool 2016 Doctoral Prize and recognized as the runner-up for the iSchools Doctoral Dissertation Award 2017. She has worked on many projects involving machine learning and data mining, particularly mining content from Twitter and other sources of user-generated content. Dr. Jasy was awarded the 2017 L’Oréal-UNESCO for Women in Science National Fellowship Award for her research in depression detection through social media. Being a champion of the data science program at her university, she is committed in training students both from academia and the industry to apply machine learning to extract insights from data. She is also actively mentoring students to build innovative systems for information analytics and visualizations, many of which have been recognized in top-notch competitions such as Innovate Malaysia, PECIPTA and Novel Research and Innovation Competition (NRIC). She is a member of the Academy of Sciences (ASM) Special Interest Group on Machine Learning (SIG ML).
SESSIONS
KEYNOTE
Extracting Emotion Insights from Textual Data
Tuesday, November 26 | 9.45am – 10.30am
Auditorium 2
With the prevalence of user-generated content on the Web especially in the form of text, there is a need for computational models that can automatically help turn plain data into interesting insights that can help drive business decisions as well as improve the quality of our lives. My talk will provide a lens to data science through the use of machine learning and text mining to make sense of emotions expressed in massive amounts of unstructured text. Emotion focuses on analyzing what people feel (e.g., happy, angry, sad, etc.) about certain entity or topic. Sentiment analysis and emotion detection in text have important applications in consumer analytics, health monitoring, stock market analysis, political analysis as well as in the development of more emotion-sensitive computers. As a crucial first step, I will discuss the different models of emotion and how to select a suitable classification scheme that will serve as the output of computational models used to detect emotions in text. Next, I will share some insights as to how we can use machine learning to build models that can detect emotions or sentiments. Finally, I will talk about more specific applications of sentiment analysis from my research projects on detecting emotions from Twitter and online customer reviews and highlight challenges that data scientists can help to solve going forward.
