Monday, 27 July, 08:30 - 12:30 EDT (Eastern Daylight Time - Canada)
Prof. Dvijesh Shastri (short bio)
University of Houston-Downtown, USA
Modality
on-site
Room: TBA
Target Audience
Anyone interested in learning machine learning for data analytics
Requirements for participants
- Laptop
- Software: Orange Datamining tool
Abstract
Knowledge of machine learning (ML) is essential for building descriptive and predictive ML models. Multiple pre-built ML libraries are available for popular programming languages such as R and Python to build and evaluate ML models. However, their usage demands good programming knowledge and, therefore, is limited to only those who know how to program in Python or R. The proposed course relaxes the need for programming by offering model building and evaluation via visual programming tools such as Orange ML (https://orangedatamining.com/). The no-code data analytics approach is suitable for the HCI community as the community consists of researchers from various fields of study, including computer science, social science, psychology, statistics, etc., who may not have a deeper knowledge of programming. Learning to build machine learning models without writing or debugging the programming code would democratize machine learning and accelerate the data analytics process.
Benefits for attendees
Participants will receive a thorough introduction to the supervised machine-learning approach for creating predictive models and will participate in guided hands-on exercises using Orange, a visual programming environment designed to facilitate the development and evaluation of machine-learning models.
Course Content
The course aims to introduce machine-learning concepts for data analytics and provide hands-on experience in developing machine-learning models without writing computer programs.
Agenda
- Introduction to Machine Learning
- kNN Algorithm
- Hands-on Exercise on kNN Model Building
- Model Evaluation
- Hands-on Exercise on Model Evaluation
- Decision Tree Algorithm + Hands-on Exercise on Model Building
Bio Sketch of Course instructor
Dr. Dvijesh Shastri is a Professor of Computer Science and an Assistant Chair of the Department of Computer Science and Engineering Technology at the University of Houston – Downtown. He specializes in affective computing and human-centered computing. He has developed advanced computer vision and image-processing algorithms capable of extracting physiological signals from thermal imagery and linking them to human psychological states, providing a novel methodological framework for human behavior analysis. He has developed a novel non-contact way of measuring emotional perspiration responses from the face, which is an innovative alternative to the traditional contact-based clinical standard that measures perspiration at the fingers. Dr. Shastri has co-authored more than thirty peer-reviewed publications and has served as principal investigator or co-principal investigator on multiple funded research projects. He received a B.E. in Electrical Engineering from Sardar Patel University in 1997, a M.S. in Computer Science from Wright State University in 2001, and a Ph.D. in Computer Science from the University of Houston in 2007.