Avatar

Daren Chao

Ph.D. Candidate

University of Toronto

Toronto, Canada

drchao@cs.toronto.edu


Biography

Daren Chao is a Ph.D. student in Data Science Group, Department of Computer Science, University of Toronto. His advisor is Prof. Nick Koudas. He received his B.S. degree from Shandong University in 2019, under supervision of Prof. Jianhua Yin. His primary research interests fall in the field of Database, Data Mining, and Machine Learning, with an emphasis on video stream queries.



Ongoing Work

December, 2019

Query actions on a Video Stream. NOT on general scenarios! It is querying more complex query predicates, such as playing badminton with a pet dog.


February, 2020

Are general metrics enough to show the performance of a specific deep model? How do we evaluate the robustness of an unexplained model? Can we even measure how confident we are that deep nets is robust enough?


Education

University of Toronto
2019 - 2024 (Expected)

PhD Student - Computer Science


Shandong University
2015 - 2019

Bachelor Degree - Computer Science and Technology
He graduated from Taishan College, a pilot college originated from the Basic Research Training Program for Top-notch Students of China. It selects Top 15 from 300 students in the computer science department.


Publications

August, 2018

We proposed a model-based short text stream clustering algorithm, MStream, which can deal with the concept drift problem and sparsity problem naturally, and furthermore, an improved algorithm of MStream with forgetting rules, MStreamF, which can efficiently delete outdated documents. [Paper] [Code]


February, 2020

SVQ++ is a system for declarative querying on real-time video streams involving objects and their interactions. It can efficiently identify frames in a streaming video in which an object is interacting with another in a specific way, increasing the frame processing rate dramatically and speed up query processing by at least two orders of magnitude depending on the query. [Paper] [Demo]


This poster paper documents how people use and interact with a self-diagnosis chatbots in real world scenarios. [Paper]


There is little research documenting how health consumers (e.g., patients and their caregivers) use chatbots for self-diagnosis in real world scenarios. This research aims to understand how health chatbots are used in real context, what issues and barriers exist in the usage, as well as how to improve the user experience of this novel technology [Paper]