Meghanath Macha

Adobe, Ph.D., Carnegie Mellon University


About Me

I build Machine Learning services for Adobe's Experience Cloud. Two areas I am actively developing are real-time fraud detection and scene text extraction.

Prior to joining Adobe, I did my Ph.D. at Carnegie Mellon University. I was fortunate to be advised by Prof. Beibei Li. During my Ph.D., I developed machine learning models to study the interplay between the value proposition and the trade-offs that come from understanding consumer behavior from location data. I studied trade-offs in data sharing between location data collectors and advertisers, explanations of anomalous consumer behavior, and also identified social determinants of health from location data.

Before joining CMU, I spent two years with Adobe Research, where I developed methods for marketing attribution, click-stream purchase prediction. During my Ph.D., I spent two summers at Adobe working on click fraud detection, identifying critical events in the consumer purchase funnel, and summer at Amazon developing active learning methods for record linkage.

I graduated top of my class in 2014 with a Bachelors and Masters in Mathematics and Computing at I.I.T. Kharagpur.

Education

Carnegie Mellon University

Ph.D. in Information Systems Aug. 2016 - March 2021

Indian Institute of Technology Kharagpur

M.Sc. and B.Sc. in Mathematics and Computing2009 - 2014

Work Experience

Adobe, Experience Cloud Intelligence

Machine Learning Engineer Mar. 2021 - Present

Carnegie Mellon University

Research Assistant Aug. 2016 - Present

Adobe, Data Science Team

Research Intern May. - Aug. 2019 & 2017

Amazon, AWS AI Labs

Research Intern May. 2018 - Aug. 2018

Adobe, Big Data Intelligence Labs

Data Scientist June. 2014 - July. 2016

Adobe, Big Data Intelligence Labs

Research Intern May. 2013 - July. 2013

Teaching

Geographic Information Systems

Teaching Assistant Summer 2020

Data Structures and Algorithms

Teaching Assistant Fall 2018, Fall & Spring 2019

Decision Making Under Uncertainty

Teaching Assistant Fall 2018, Fall and Spring 2019

Awards

Best Paper Award

WITS 2019, Munich, Germany December 2019

Suresh Konda Award

Best First Paper at Heinz, CMU May 2018

CMU - PwC Presidential Fellowship

Outstanding academic excellence 2017 - 2018

Institute Silver Medal

Highest GPA in undergraduate July 2014

Prof. K.L. Chopra Award

Best Masters thesis in Science July 2014

Working Papers

[W4]
Consumer Mobility Data and Research Opportunities on Location-Smart Retailing

Natasha Foutz Zhang, Baohong Sun, Meghanath Macha
Presented at Marketing Science 2020 (Paper under preparation.)

[W3]
Learning Individual Social Determinants of Health from Location Big Data

Meghanath Macha, Beibei Li and Natasha Foutz Zhang.
Shorter version accepted at WITS 2021
(Paper under preparation.) [ working paper ]

[W2]
Trading Privacy for the Greater Social Good: How Did America React During COVID-19?

Anindya Ghose, Beibei Li, Meghanath Macha, Chenshuo Sun, Natasha Zhang Foutz and Josh Anton (Major Revision at Management Science) [ Shorter version accepted at ICIS 2020 ] [ working paper ]

[W1]
Perils of Location Tracking? Personalized and Interpretable Privacy Preservation in Consumer Trajectories

Meghanath Macha, Beibei Li, Natasha Zhang Foutz and Anindya Ghose (Major Revision at ISR)
WITS 2019 Best Paper Award
Shorter version accepted at ICIS 2019 conference proceedings.
Presented at NYU/ABA NextGen Antitrust Conference, 2020
[ working paper | video]

Publications

[ Google Scholar | DBLP ]

[R11]
Multiple Attribute Fairness: Application to Fraud Detection

Meghanath Macha, Sriram Ravindran, Deepak Pai, Anish Narang and Vijay Srivastava
Ethical Artificial Intelligence: Methods and Applications, KDD 2022 [ paper ]

[R10]
Revisiting How to Focus: Triplet Attention for Joint Entity and Relation Extraction

Debraj Basu Meghanath Macha, and Deepak Pai
Document Intelligence Workshop, KDD 2022 [ paper ]

[R9]
Value and Tradeoffs in Learning from Consumer Location Data

Meghanath Macha
Thesis [ link ]

[R8]
Trading Privacy for the Greater Social Good: How Did America React During COVID-19?

Anindya Ghose, Beibei Li, Meghanath Macha, Chenshuo Sun, Natasha Zhang Foutz and Josh Anton
International Conference on Information Systems (ICIS) 2020 [ paper ]

[R7]
CrEOS: Identifying Critical Events in Online Sessions

Meghanath Macha, Shankar Venkitachalam and Deepak Pai.
Temporal Web Analytics Workshop , WebConf (WWW) 2020 [ paper ]

[R6]
Geo-Targeting, Privacy, and the Rise of Consumer Location Trajectories

Meghanath Macha, Beibei Li and Natasha Zhang Foutz
International Conference on Information Systems (ICIS) 2019 [ paper ]

[R5]
ConOut:Contextual Outlier Detection with Multiple Contexts

Meghanath Macha, Deepak Pai and Leman Akoglu
ECML PKDD 2018 [ paper | code | website ]

[R4]
X-PACS: eXPlaining Anomalies by Characterizing Subspaces

Meghanath Macha, Leman Akoglu
DMKD Special Issue , ECML PKDD 2018 Journal [ paper | arxiv | code | website ]

[R3]
Anti-Ad Blocking Strategy: Measuring Its True Impact

Atanu R. Sinha, Meghanath Macha, Pranav Maneriker, Sopan Khosla, Avani Samdariya, and Navjot Singh
Advertisement workshop, KDD 2017 [ paper ]

[R2]
A non-parametric approach to the multi-channel attribution problem

Meghanath Macha, Shiv Kumar Saini and Ritwik Sinha
Web Information System Engineering ( WISE ) 2015 [ paper ]

[R1]
Modelling visit similarity using click-stream data: A supervised approach

Deepak Pai, Abhijit Sharang, Meghanath Macha, and Shradha Agrawal
Web Information System Engineering ( WISE ) 2014 [ paper ]

Patents

[ Justia Patents ]

[P9]
A practical approach to ensure fairness in machine learning

Meghanath Macha, Sriram Ravindran, Anish Narang, Vijay Srivastava, Deepak Pai [ To be filed ]

[P8]
A method to identify stages in the path to online e-commerce

Meghanath Macha, Shankar Venkitachalam and Deepak Pai [ To be filed ]

[P7]
User Segment Generation and Summarization

Meghanath Macha and Deepak Pai [ Filed ]

[P6]
Techniques to quantify efectiveness of site wide actions

Atanu Sinha, Meghanath Macha, Pranav Maneriker, Sopan Khosla, Avani Samdariya and Navjot Singh [ link ]

[P5]
Generating and utilizing a conversational index for marketing campaigns

Meghanath Macha, Moumita Sinha, Kokil Jaidka and Niyati Chhaya [ link ]

[P4]
Simulation-based evaluation of a marketing channel attribution model

Meghanath Macha, Ritwik Sinha and Shiv Kumar Saini [ link ]

[P3]
Buying Stage Determination in a Digital Medium Environment

Meghanath Macha, Ritesh Noothigattu, Shubham Garg, Abhishek Kandoi and Atanu R. Sinha [ link ]

[P2]
Value function-based estimation of multi-channel attributions

Meghanath Macha, Ritwik Sinha and Shiv Kumar Saini [ link ]

[P1]
Visitor session classification based on clickstreams

Deepak Pai, Abhijit Sharang, Meghanath Macha, and Shradha Agrawal [ link ]