February 2021 BARUG Meetup

Agenda:
4:30 Networking
5:00 Eric Suess - Introducing the COVID-19 Data Hub
5:35 Yu-Shih Chen - TDAsweep: A Novel Dimension Reduction Method for Image Classification Tasks
6:05 Dave Hurst: Practical pins: Bizops with Salesforce data

############
Eric A. Suess
Professor, Statistics and Biostatistics, Engineering, CSU East Bay
machineVantage, Chief Statistician

Abstract:

In this talk we will introduce the COVID-19 Data Hub and the associated R R package. The COVID-19 Data Hub is an R Consortium supported project created by Emanuele Guidotti and David Ardia. The COVID-19 package provides access to a curated up-to-date dataset containing counts of various measures of the pandemic from nearly all countries around the world at 3 levels of aggregation, when available: Country, State, and City.

We will give examples of the use of the main function provided by the package and its few options, and some suggested usages of the data. We will examine the inner workings of the project, including the website, where user input can be submitted, the Github site, where code submissions can be made, and finally we will look at the process for adding data.

A project plan for inviting users to join the effort for further improvement and maintenance of the project will be presented.

#############
Yu-Shih is a fourth year undergraduate studying Computer Science at the University of California, Davis, and his research and interest lies mainly in the field Artificial Intelligence.

TDAsweep: A Novel Dimension Reduction Method for Image Classification Tasks

Abstract:

One of the most celebrated achievements of modern machine learning technology is the automatic classification of images. However, success is typically achieved only with major computational costs. Here we introduce TDAsweep, a machine learning tool aimed at improving the efficiency of automatic classification of images.

#############
Dave Hurst
Dave is RStudio's Business Development Director

Abstract:
As an orchestration language, R enables BizOps by providing a robust set of tools to interrogate, transform, model and visualize data from various business systems in order to better analyze and inform business decisions.

This talk will encompass a brief overview of the pins package, and techniques for accessing Salesforce data. Then we will put these two concepts together to illustrate how the combination enables a transformation in the way your teams can collaborate to store, analyze and report on data. We’ll discuss use cases in the wild as well as some examples you can try out yourself using the Salesforce Trailhead training environment.