Jacob J Barkovitch

Lecce, Italy before the 2022 ICIAP conference.

Hello! My name is Jacob Barkovitch and I am a motivated Data Science and Machine Learning Researcher with 5+ years experience in multiple research streams and 2+ in industry roles. From developing a novel dataset of 50,000+ high-stake lying facial images, to leveraging databases of over 20 million samples, I have striven to broaden my research skills whenever I can.

Having grown up on a sustainable agricultural farm, I have always been passionate about environmentalism. I believe that the work we do in the field of computer science has the potential to lead to a healthier planet. Through my research, I ultimately want to leave a positive impact. The high-stake lying dataset I helped create and publish could help the wrongfully convicted. The research I do in materials science, on discovering new metal-organic frameworks, could be used in gas separation, air purification, and carbon capture.

My experience as a researcher allowed me to serve in an internship as an Artificial Intelligence Research Associate at a startup company directly involved in lowering harmful emissions from power plants. This rewarding position lets me leave a positive impact on the planet. Through analyzing furnace sensors and camera image data, we can optimize the combustion of fuel in real-time for improved efficiency and functionality.

My goal is to continue to improve society and the planet in one way or another, and follow this trend throughout my career. I currently work as a Consultant on the Strategic Sourcing Optimization team for The Resource Group, subsidiary of Ascension Healthcare. In this role I build out analytical tools, dashboards and reports for business leaders and departments to help guide their decision making. The work I do helps save costs for both the company and the patients seeking affordable care.

selected publications

Poker Bluff Detection Dataset Based on Facial Analysis

Jacob Feinland, Jacob Barkovitch, Dokyu Lee, and 2 more authors

In Image Analysis and Processing – ICIAP 2022 2022

Abs HTML

Unstaged data with people acting naturally in real-world scenarios is essential for high-stakes deception detection (HSDD) research. Unfortunately, multiple HSDD studies involve staged scenarios in controlled settings with subjects who were instructed to lie. Using in-the-wild footage of subjects and analyzing facial expressions instead of invasive tracking of biological processes enables the collection of real-world data. Poker is a high-stakes game involving a deceptive strategy called bluffing and is an ideal research subject for improving HSDD techniques. Videos of professional poker tournaments online provide a convenient data source. Because proficiency in HSDD generalizes well for different high-stakes situations, findings from poker bluff detection research have the potential to transfer well to other more practical HSDD applications like interrogations and customs inspections. In the hopes of encouraging additional research on real-world HSDD, we present a novel in-the-wild dataset for poker bluff detection. To verify the quality of our dataset, we test its regression accuracy and achieve a Mean Square Error of 0.0288 with an InceptionV3 model.