Tutorials

Free tutorials on educational and social science technology, targeted at the beginner, and practitioner audiences will be open to the general public, but require advanced registration to cater for workshop’s logistics.

tiny.cc/wessttutorials is an easy to remember shortcut URL, to access the tutorials and schedule pages.

Register for the workshop

You will be contacted later to confirm your participation in any workshops or events that you are interested in.  All tutorials will take place at iCube Level 3.  Floor layout and directions here.

Morning Tutorials

  1. LearnSphere, by Dr John Stamper (Carnegie Mellon University; 2 hours; 9:15-11:00, STMI Classroom)
  2. Basics of Deep Learning, by Mr Animesh Prasad and Mr Muthu Kumar Chandrasekaran  (National University of Singapore; 1 1/2 hours; 9:15-11:00, CHI Classroom)
  3. Leveraging Technology for Collaborative, Dynamic, Personalized Experimentation, by Dr Joseph Jay Williams (National University of Singapore; 1 1/2 hours; 11:15-13:00; CHI Classroom)
  4. Econometrics and Social Science Methods, by Dr Tuan Q Phan (National University of Singapore; 1 1/2 hours;11:15-13:00; STMI Classroom)
  5. Sentiment Analysis on Social Media, by Mr Wenqiang Lei and Mr Kishaloy Halder  (National University of Singapore; 1 1/2 hours; 11:15-13:00; Meeting Room 8)

Afternoon Tutorials

  1. Learning Analytics Data Policy and Adoption Strategies, by Prof. Dragan Gašević (University of Edinburgh; 1 1/2 hours; 14:00-15:30; Meeting Room 8)
  2. Getting Started on Computational Social Science Cluster, by Dr Tuan Q Phan and Dr Xuesong Lu (National University of Singapore; 2 hours; 14:00-15:30; STMI Classroom)
  3. AutoTutor, an implementation of Conversation-Based Intelligent Tutoring Systems (ITS), by Prof Xiangen Hu (University of Memphis; 2 hours; 16:00-18:00; CHI Classroom)
  4. From Data to Design of Dynamic Support for Collaborative Learningby Carolyn Rosé (Carnegie Mellon University, Language Technologies Institute and HCI Institute; 1.5 hours;14:00-15:30; STMI Classroom, I3 03-44)

1 LearnSphere

About the Instructor: John Stamper is an Assistant Professor at the Human-Computer Interaction Institute at Carnegie Mellon University. He is also the Technical Director of the Pittsburgh Science of Learning Center DataShop. His primary areas of research include Educational Data Mining and Intelligent Tutoring Systems. As Technical Director, John oversees the DataShop, which is the largest open data repository of transactional educational data and set of associated visualization and analysis tools for researchers in the learning sciences. John received his PhD in Information Technology from the University of North Carolina at Charlotte, holds an MBA from the University of Cincinnati, and a BS in Systems Analysis from Miami University.  Prior to returning to academia, John spent over ten years in the software industry including working with several start-ups.

 


2 Basics of Deep Learning

This tutorial will touch upon basics and the recent advancements in the field of deep learning. This prime focus would be make the audience aware of the landscape of machine learning and the application of deep learning to solve some classic problems in the machine learning. The tutorial would include information on datasets, toolkits, resources etc. with some simple hands on to familiarise with simple neural architectures. Main focus of the tutorial would be on unstructured data like text, with covering the basics of structured data like industrial table-style data. We would discuss word vector representations, feed-forward neural networks, recurrent neural networks, long-short- term-memory models, convolutional neural networks etc.

Intended Audience: This tutorial is primarily for beginners who would like to learn more about recent paradigm in machine learning and data analytics via deep learning. Some understanding of machine learning would be ideal but not required. Bringing a laptop is not compulsory, however, you are welcomed to bring it in case you want to login to our remote server to run the demonstrations during the session.

About the Instructor: Animesh Prasad is a third year PhD candidate at School of Computing, National University of Singapore. He is associated with the Web Information Retrieval / Natural Language Processing Group (WING) and advised by associate professor Dr. Min-Yen Kan. His research interest lies in application of deep learning to text, specifically scientific documents. He is working on application of deep learning to solve some cornerstone tasks for digital libraries like reference string parsing, scientific information extraction, multi- document summarisation etc. Prior to joining NUS, he graduated with a B.Tech. in Computer Science and Engineering from Indian Institute of Technology, Patna. He has worked for an year at Cisco Systems as Software Development Engineer in anti-spam email server team and did internships at Mentor Graphics and Indian Railways.

About the Instructor: Muthu Kumar Chandrasekaran is a fourth year Ph.D. student at the Web IR /NLP Group, NUS advised by Prof. Min-Yen Kan. He is broadly interested in natural language processing and its applications to information retrieval; specifically, in retrieving and organising information from asynchronous conversation media such as scholarly publications, discussion forums. He has been working on these topics for 6 years and has publishing in premier AI, NLP and IR venues. His Ph.D. research is on “Investigating Instructor Intervention in MOOC Discussion Forums”. He also Chairs the BIRNDL workshop series and scientific document summarization, CL-SciSumm, Shared Task series. He has served on the Programme Committee of ACL, EMNLP and BIR@ECIR.  He holds a Master’s degree from NUS and a Bachelors in Computer Science from Anna University, India. He has also worked as an application software developer for two years in India.


3 Leveraging Technology for Collaborative, Dynamic, Personalized Experimentation

About the Instructor:  Dr Joseph Jay William’s  research agenda is to create intelligent self-improving systems that conduct dynamic experiments to discover how to optimize and personalize technology, helping people learn new concepts and change habitual behavior. This requires using computational cognitive science and Bayesian statistics to bridge human-computer interaction with machine learning, with applications to education and health behavior change.  He was previously a postdoc at Stanford University in the Graduate School of Education and Lytics Lab, working with the Office of the Vice Provost for Online Learning and Candace Thille’s Open Learning Initiative. He received my PhD in Computational Cognitive Science from UC Berkeley’s Psychology Department. He worked with Tania Lombrozo to investigate why prompting people to explain “why?” helps learning, and with Tom Griffiths on using Bayesian statistics and methods from machine learning to characterize learning, reasoning, and judgment.


5 Econometrics and Social Science Methods

Social scientists are interested to use data to reveal underlying behavior of mechanisms which are sometimes hidden, missing, or biased in data.  This session introduces the econometrics framework to thinking about data for social science explorations.  We will start with fundamental data generating processes, introduce regression methods, and discuss modeling techniques to address some limitations in data to draw causal analysis in different scenarios.  The lecturer will use examples in social media, and cover methods such as difference-in-difference, propensity score matching, regression discontinuity, and instrumental variables.  The target audience are those who are new social science including computer science researchers.

About the Instructor:  Tuan Q Phan is currently an Assistant Professor at NUS in the Department of Information System. His research brings together social sciences, computer science, and statistics to investigate social networks, social media, Big Data, product diffusion, word-of-mouth, and web and mobile commerce. He has looked at a number of topics related to online and offline social networks such as the effects of natural disasters on human social networks, how individuals react to privacy changes, how some individuals become influential, and how firms can leverage social media.

Before going for his graduate studies, he has also started a company providing 3-D computer graphics for mobile devices. He continues to be consult for a number of firms dealing with Big Data issues in marketing and advertising, retail and e-commerce, IT, publishing, logistics, healthcare and consumer finance. He is currently the co-founder for Key Insights which provides privacy-preserving analytics solutions for retailers, telecom, and healthcare industries.

He has also been in the top 100 competitive ballroom dancer in the world, and top 10 dancer in the USA. Dr. Phan likes to build robots and tinker with technology in his spare time.

He holds a Doctor of Business Administration from Harvard Business School in Marketing, and a Bachelor of Science from MIT in Computer Science & Electrical Engineering with concentrations in business and economics.


4 Sentiment Analysis on Social Media
This talk will introduce basic sentiment analysis techniques in NLP and its application in social platforms – in particular Twitter. In this talk, Wenqiang will start with the general introduction of NLP technology and then followed by standard technologies of sentiment analysis in formal (well written) texts. The later half of the talk would focus on the application domains. We would discuss state-of-the-art techniques for analysing sentiment from tweets.

About the Instructor: Wenqiang Lei is a third year Ph.D candidate under the supervision of Prof. Min-Yen Kan. Before that, he successfully got his B.S. degree from East China Normal University (ECNU), being the first student ever graduating from the univeristy within 3.5 years. His PhD dissertation focuses on implicit discourse relation recognition while his research interest ranges from deep learning, machine learning to linguistic aspect of natural language processing (NLP). Apart from being a young researcher in computer science, he is also quite versatile, being the sixth generation descendant of Yang style Taichi and an Erhu player in NUS Chinese Orchestra. He has long been interested in incorporating computational methods in arts and social science study.

About the Instructor: Kishaloy is a PhD candidate at School of Computing, working under Prof. Min-Yen Kan. He is interested in Recommendation Systems, Information Retrieval, and NLP. He is currently working on building a Recommendation System for Online Health Forums as part of his PhD thesis. Prior to this, he has done his Masters in Computer Science from IIT Bombay, India. He also has work experience of two years in the industry as Software Engineer.


1 Learning Analytics Data Policy and Adoption Strategies

About the Instructor:


2 Getting Started on Computational Social Science Cluster

The Computational Social Science cluster, “Musang,” is a strategic initiative by the School of Computing to promote social science studies using computational and large-scale analytics methods.  Started in January 2017, the cluster will eventually grow to nearly 100 nodes, and can host large-scale datasets for analysis.  This shared resource will enable researchers and principal investigators to host sensitive datasets, provide tools and cloud computing services to analyze datasets, and to support social science research.  Initially, Musang will be a free service for PIs and their team, but will eventually move to a yearly subscription model to help sustain infrastructure and manpower support.
This session will guide researchers through getting started on the cluster.  It will include basics in command line, connecting to the cluster, basic linux and manipulation, and submitting jobs to the cluster.  The tutorial will go a simple example in the WESST context.  This is suitable for any researchers who work large datasets and who want to look for support for their research.
Interested participants need to register by emailing Xuesong with the subject “WESST CSS registration”.  Provide your desired username.  You will receive a confirmation and temporary password to use during this session.  Please bring your own laptop.

About the Instructors:  Tuan Q Phan is currently an Assistant Professor at NUS in the Department of Information System. His research brings together social sciences, computer science, and statistics to investigate social networks, social media, Big Data, product diffusion, word-of-mouth, and web and mobile commerce. He has looked at a number of topics related to online and offline social networks such as the effects of natural disasters on human social networks, how individuals react to privacy changes, how some individuals become influential, and how firms can leverage social media.

Before going for his graduate studies, he has also started a company providing 3-D computer graphics for mobile devices. He continues to be consult for a number of firms dealing with Big Data issues in marketing and advertising, retail and e-commerce, IT, publishing, logistics, healthcare and consumer finance. He is currently the co-founder for Key Insights which provides privacy-preserving analytics solutions for retailers, telecom, and healthcare industries.

He has also been in the top 100 competitive ballroom dancer in the world, and top 10 dancer in the USA. Dr. Phan likes to build robots and tinker with technology in his spare time.

He holds a Doctor of Business Administration from Harvard Business School in Marketing, and a Bachelor of Science from MIT in Computer Science & Electrical Engineering with concentrations in business and economics.

Dr. LU Xuesong is a research fellow of information systems at School of Comput- ing (SoC) at National University of Singapore (NUS). He received his Bachelor degree in computer science from Fudan University, China, in 2008, and his PhD in computer science from National University of Singapore in 2013.

During his PhD candidate, LU Xuesong was working on random sampling and generation over data streams and graphs. The work aimed to extract rep- resentative sub-datasets of manageable sizes from large-scale datasets. The categories of the investigated datasets included data streams, large graphs and social networks. After graduating from NUS, he joined the DIAS Lab at EPFL in Switzerland and worked as a research Fellow. During this period, he was working on spatial data management under the Human Brain Project (HBP), which is a European Commission Future and Emerging Technologies Flagship that aims to accelerate people’s understanding of the human brain, make ad- vances in defining and diagnosing brain disorders and develop new brain-like technologies. In particular, his research was focused on developing new spatial indexing techniques to speed up the building, analysis and simulation of brain models.

Currently, LU Xuesong’s work is focusing on big data analytics using data mining/machine learning techniques.


3 AutoTutor, an implementation of Conversation-Based Intelligent Tutoring Systems (ITS)

AutoTutor started as a flagship application at the Institute of Intelligent Systems (IIS) of the University of Memphis 20 years ago. In the past 20 years, there have been major research funding on AutoTutor Research and Development (Over $35 Million US Federal Funding).

In this tutorial, I will talk about three aspects of AutoTutor: 1) selected cognitive theories of learning that served as theoretical foundations of AutoTutor, 2) Enabling technologies that make AutoTutor work, 3) Example applications of AutoTutor in different domains.  At the end, it is expected the audiences can create his/her own autotutor.

About the Instructor: Dr. Xiangen Hu is a professor in the Department of Psychology and Department of Electrical and Computer Engineering at The University of Memphis (UofM) and senior researcher at the Institute for Intelligent Systems (IIS) at the UofM and visiting professor at Central China Normal University (CCNU). Dr. Hu received his MS in applied mathematics from Huazhong University of Science and Technology, MA in social sciences and Ph.D. in Cognitive Sciences from the University of California, Irvine. Dr. Hu is the Director of Advanced Distributed Learning (ADL) center for Intelligent Tutoring Systems (ITS) Research & Development, and senior researcher in the Chinese Ministry of Education’s Key Laboratory of Adolescent Cyberpsychology and Behavior.