DATA-DRIVEN IN UNCERTAIN TIMES

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenter: Nagdev Amruthnath

LinkedIn Profile

Slide Deck

Abstract:

The manufacturing industry is currently going through its fourth industrial revolution with advancements in quality, computer vision, safety, and maintenance. Among these advancements, predictive maintenance is seen as the magical unicorn for achieving a competitive edge and significantly improving efficiency. In a study by PWC, it is estimated that by implementing predictive maintenance, manufacturing industries could reduce cost by 12 percent, improve uptime by 9 percent, reduce various risks by 14 percent and extend the life of the machine by 20 percent. In this presentation, we introduce predictive maintenance, build predictive models using R, and deploy them in a production environment.

 Co-Sponsored by:

West Michigan R Users Group

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenter: Christopher Brooks

LinkedIn Profile

Slide Deck

Abstract:

This presentation will outline a program of reproducibility research in the domain of learning analytics with a specific focus on predictive models of student success.

Presenter: Siqian Shen, http://www-personal.umich.edu/~siqian/

Slide Deck

Abstract:

The outbreak of coronavirus disease 2019 (COVID-19) has created a global health crisis and the response to the COVID-19 pandemic is deeply influenced by local, national and global policies and decisions. In this talk, we present a few examples to demonstrate (i) how infection status dynamically affects mobility patterns and travel behavior, (ii) how to optimize business/state reopening and closedown strategies, (iii) how to distribute testing kits and locate testing facilities, and (iv) how to redesign public transit systems like city buses to reduce passengers¡¯ infection risk. In particular, we show the use of data analytics tools and optimization models for solving these problems, validated using real data of COVID-19 infection, business economy and local mobility. The talk is based on an ongoing collection of COVID-related system engineering problems and their solution methods at:  https://sites.google.com/umich.edu/decision-tools-for-covid19/

Sponsored By:
Jing

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenter: Mark Fei

LinkedIn Profile

Abstract:

In this session we’ll discuss the rapidly spreading approach organizations are embracing, moving to a real-time, event driven architecture for all their applications. We’ll answer questions like, “Why?”  “What are the advantages?” and we’ll examine some real life use cases that bring these answers into focus. Following that, we’ll delve into an overview of Apache Kafka, the de facto standard for achieving these transformations. During and after the presentation you will be able to ask questions of the speaker.

 Sponsored By:

Confluent

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenter: David Corliss

LinkedIn Profile

Slide Deck

Abstract:

With so much Big Data technology starting out in university and government research centers, learning to leverage these resources and getting them to work in the real world presents special challenges. This presentation maps out a step-by-step process to move emerging smart technology from lab, classroom, and journal all the way through to industrial applications.

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenters:
Terry Crist, LinkedIn Profile
Steve Krause, LinkedIn Profile
Tom Weinandy, LinkedIn Profile

Slide Deck

Abstract:

The peanut butter problem occurs when a customer orders a specific kind of peanut butter to be delivered in a grocery order, but the picker fulfilling the order sees that it’s not on the shelf. Which of the fifty other kinds of peanut butter should the picker select instead? This talk lays out a framework for substituting out of stock items from the perspective of a retail store or order fulfillment platform. We first define the problem from a business perspective and argue for how this solution helps a store’s bottom line. Then we propose a design for building a recommendation engine that uses available data to estimate a percent match between all products within a store. Finally, we propose an architecture for hosting the product data and recommender using a modern data warehouse. Our proof of concept shows that the recommendation engine successfully identifies meaningful product substitutions. The design is robust to work with limited data and flexible to allow for customization at both the store level and customer level.

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenter:

Jay Qi

LinkedIn Profile

Follow on Twitter

Slide Deck

Abstract:

There is a growing awareness of ethical concerns in data science like algorithmic bias. The next step is to apply ethics to everyday practice. In this talk, you will learn about an actionable approach that enables you as a data scientist to start identifying and addressing ethical concerns in your work today. We will discuss the idea of ethics checklists, and show you how you can use an open-source tool, deon (deon.drivendata.org), to integrate an ethics checklist into your data science projects. Deon and the checklist framework enables you to actively consider the ethical implications of your work, and to preemptively address issues that may otherwise get overlooked. Rather than providing simple answers, our checklist framework spurs an ongoing dialogue that helps surface trade-offs, nuances, and unintended consequences. Using the default checklist in deon, we will walk through the relevant ethical concerns that arise at different points in the data science process—ranging from data collection to storage to modeling and deployment. We will use concrete real-world examples to illustrate the common ethical issues that arise. Ethics in data science requires more than just good intentions. Come learn how to jumpstart the conversation that all data science teams should be having and translate those good intentions into ethical actions.

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presented by:

Raymond Joseph, LinkedIn Profile
Adi Eyal, LinkedIn Profile

Slide Deck

Abstract:

The presenters will recount the development and application of data-analysis tools that were pivotal in investigative reporting that exposed widespread corruption in the South African lottery. The presenters will also comment on other applications of data analytics to global investigations, such as the UN-guaranteed right to access to clean water.

Co-presented by:

The Center for Collaborative Investigative Journalism (CCIJ) and the Padnos/Sarosik Civil Discourse Program at GVSU

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presented by: Sray Agarwal

LinkedIn Profile

Abstract:

This presentation will be an overview of ML/AI tools & techniques to achieve fair and ethical results across the data-collection and data-modeling lifecycle.

Sponsored By:

Initech Global

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenters: Elliott Church, Omar Malik, Tuhin Mitra, Randall Prium

Abstract:

“SafeZones” is a mobile/web app designed to facilitate safe return to normal commercial activity in the COVID-19 era. Safezones incorporates two functionalities: It provides consumers with a ready source of accurate information about COVID-19 safety/risk factors about local businesses and also provides local businesses with information regarding availability of personal protective equipment. SafeZones is a collaborative project of the Research Division of Big Data Ignite, The Right Place, Grand Rapids Chamber of Commerce, Kent County, Calvin University, Grand Valley State University, and other organizations. The overarching goal in creating SafeZones is to provide a data-driven tool designed to dispel unwarranted fear, restore consumer confidence, and motivate appropriate caution where needed. SafeZones uses data contributed by businesses, community organizations, and the general public to analyze health-and-safety conditions at local businesses and public venues. In this session, a panel of key contributors will discuss the goals, development, deployment, and data analytics of the SafeZones app.

Sponsored By:

Initech Global

—————————
To receive a link to watch the recording of this talk now, register at https://bigdataignite.org/register/
—————————

Presenters:

Balu Vyamajala, Principal Software Engineer, Initech Global
Mahesh Boosi, Lead Data Engineer, Initech Global

Abstract:

This talk will go over how to build a scalable data lake in the cloud. The presenters will highlight the ability to consume streaming data from AWS Kinesis, build a Datalake in S3 and run SQL queries from Athena.

Abstract:

This talk will go over how to build a web application to manage images and videos with custom tagging and the ability to query in the cloud with image recognition in a matter of minutes.