Big Data Ignite 2021
In 2021, Big Data Ignite hosted a series of webinars in place of an in-person conference.
Disproportional Impact of COVID-19 on Marginalized Communities
Speaker:
David J Corliss (Director, Peace-Work)
Abstract:
Disasters of all kinds often have a particularly severe impact on minorities and other marginalized communities, and COVID-19 has been no exception. This analysis examines the disproportionate impact of the COVID-19 pandemic among different groups. The New York Times COVID-19 database metrics provide the outcomes, using deaths per capita the pandemic. Counties with the highest percent population BIPOC, Indigenous and below the poverty line are compared to those with those with lowest levels. Log odds are calculated and compared with published levels for medical risk factors (e.g., chronic lung disease). The two factors most at risk, with a log odds of about 10, are persons of color and cardiac patients. A time series analysis of the changing impact by month is also given, including breakthrough infections of vaccinated people. The Implications of these results are discussed to advise potential mitigation strategies.
Reflections on a Decade in Data Science
Presenter:
Michael Bloem (Applied Data Scientist, Amazon)
Abstract:
Within the last decade or two, data science emerged as a discipline. In this talk, I will share observations from four data science jobs: one in government research, one with a manufacturing enterprise, one with a consultancy, and one with a large tech company. I will reflect on transitioning between jobs and working in different geographic regions.
Building a Data-Science Capability: Aspirations, Obstacles and Lessons-Learnt
.
Multi-Cloud Event Driven Analytics with Multiple Clouds
Presenters:
Patrick Druley, Solution Engineer, Confluent
Abstract:
We don’t always get to choose our cloud provider and sometimes we end up with more than one. Taking an event driven approach to Analytics and leveraging the flexibility of Confluent Cloud allows you to deliver low latency data regardless of where your data starts.
Sponsored by:
Confluent
AI Technologies to Support Smart City Infrastructure
Presenters:
Michael Farmer (Professor and Department Head of Computer Science, Kettering University)
Peter Stanchev (Professor of Computer Science, Kettering University)
Abstract:
There are many factors that contribute to the development of high-level concepts such as smart devices, smart cars, smart homes and even smart cities. In one hand it is inevitable that the advancements of handheld devices create the basis for realization of many smart technologies. This is due to the increase of computational power and minimization of electronics worldwide. On the other hand, the exponential increase of data according to different sources leads to a growing necessity for the development of adequate technologies that can clean, structure and analyze more and more data every day. This in turn leads to the creation of new smarter technologies, which have a profound impact on our everyday life.
In this talk we will focus on the advancements of smart technologies that form the smart city ecosystem.
The following topics using Smart Services will be discussed:
a) Image retrieval using high level semantic features based on extraction of low level color, shape and texture characteristics and their conversion into high level semantic features.
b) Machine learning, with regard to deep learning, helping to identify, classify, and quantify patterns in medical images.
c) Sensor fusion framework using human cognitive models combined with probabilistic models.
Auto-Encoders and Deep Neural Networks for Structural Classification
Speaker:
Saroja Kanchi (Professor of Computer Science, Kettering University)
Abstract:
Various learning tasks require dealing with graph data which contains rich relation between elements. Applications include chemistry and drug design, social networks, predicting protein interface, and classifying diseases demand etc. In this talk, a survey of current techniques of trends in structural classification will be presented. While there are powerful models available for learning with large graphs such as GNN, computation challenges must be overcome due to neighborhood explosion. GNN and its variations will be compared for performance and accuracy of the models. Future research topics will be presented.
The Home Depot EDW Journey to Google Cloud Platform
Presenter:
Lisa Barber (Sr Manager, Data & Analytics, The Home Depot)
Abstract:
The migration of The Home Depot (THD) Enterprise Data Warehouse to Google Big Query was a large, complex and challenging endeavor. This talk will cover the problem THD set out to solve, strategy & approach for solving it, as well as the practical lessons we learned along the way.
Sponsored by:
The Home Depot
Twitter-Patter: How Social Media Drives Foot Traffic to Retail Stores
Presenter:
Thomas J. Weinandy (Senior Data Scientist, BlueGranite)
Abstract: