Web based GUI for Data and Machine Learning Modeling

AI Dynamics is building an end to end data platform for managing datasets used in machine learning workloads. The end goal is to provide a web based interface where users can import datasets with a variety of different data types (audio, image, video, DICOM, DNA, RNA, amino acid, text, numbers) and add annotations in preparation for machine learning model generation. The platform further provides an analysis of data by displaying data summarization and visualization and uses heuristics to suggest various data cleaning possibilities. When a user accepts these suggestions, the platform prepares the data for machine learning. The platform then recommends initial deep learning architectures and provides a visual interface for reconfiguration including advice on creating the most appropriate architecture for the dataset on hand. The platform then helps train the models and analyzes the results. The user has an opportunity to make changes to the data cleaning and deep learning (DL) architecture and study its impact on the results. While the end-to-end platform is now in place, there are many additional data cleaning, data preparation, and reconfiguration of deep learning options that need to be explored and tested on different types of datasets.

Faculty Supervisor:

Pawan Lingras

Student:

Partner:

AI Dynamics

Discipline:

Computer science

Sector:

Information and cultural industries; Professional, scientific and technical services

University:

Saint Mary's University

Program: