04/04/2022
By Vaishali Mahipal
The Kennedy College of Science, Department of Computer Science, invites you to attend a master’s thesis defense by Vaishali Mahipal on ” DoodleIt – Introducing Middle School Students to Image Recognition.”
Candidate Name: Vaishali Mahipal
Degree: Master’s
Defense Date: Friday, April 15,2022
Time: 10 to 11 a.m. EST
Location: DAN 309, Dandeneau Hall, North Campus
Thesis/Dissertation Title: DoodleIt – Introducing Middle School Students to Image Recognition
Committee Members:
- Fred Martin (advisor), Computer Science Department, University of Massachusetts Lowell
- Ruizhe Ma, Computer Science Department, University of Massachusetts Lowell
- Joseph Gonzales, Department of Psychology, University of Massachusetts Lowell
Brief Abstract:
We designed and implemented DoodleIt, a web application to engage users in understanding basic concepts of convolutional neural networks. DoodleIt is an interactive software application where students make simple line drawings on a canvas area, and a previously-trained CNN identifies the object drawn. The app visualizes the different layers that are involved in the process of convolution neural networks, including a display of kernels, resulting feature maps and the percentage of the match at output neurons.
We analyzed prior work that is aimed toward building artificial intelligence tools and curricula for K-12 studies and their impact on children. This work found that teaching AI to children from an early age is beneficial and these tools were able to change the children’s perception of AI.
The objective of our work is to teach the structure and components of a convolution neural network involved in the process of image classification to middle school students.
In this research study, we have collected qualitative data from middle school students at Lowell Public School and we will be conducting single-case analyses and qualitative assessments.