Visual and interactive data analytics platform employing machine learning methods

• Main Author: Chen, Tianrui
• Other Authors: Mao Kezhi
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Computer science and engineering::Data; Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/140067
• Institution: Nanyang Technological University

With machine learning playing a gradually essential part in the data analysis field, machine learning skills are in great demand by entities. However, machine learning is not an easy subject and is hard to master by ordinary users due to its complex structures and huge data magnitude. Interactive Machine Learning and machine learning platforms applying the concept have been introduced, but many face limitations in usability and understandability. The project develops an interactive machine learning data analytics platform with interpretable visual explanations in every step of the model building process. It aims to explore the best methodology behind and create an extremely friendly user experience without requiring much learning time or prior knowledge from users. Therefore, users could be able to directly participate into the model building process, contributing their own expertise and improving model performance catering more towards real needs. The platform is also a suitable education tool for machine learning beginners much more powerful than literal lecturing.