Emaad
Manzoor

I am a fifth year PhD student in Information Systems at the Heinz College of Carnegie Mellon University. I am advised by Dokyun Lee and George Chen. My research is supported by a McKinsey & Company Ph.D. Fellowship.

Methodologically, I develop scalable machine-learning methods that operate on data with complex structure, such as networks and text. Substantively, I use such methods to examine causal questions pertaining to persuasion and its linguistic determinants.

---

Working Papers

• • On the Persuasive Power of Reputation in Deliberation Online
  • Emaad Manzoor, George Chen, Dokyun Lee, Michael D. Smith
  • Major revision (first round) at Management Science
  • Resources: Website, Preprint, Discussion slides by Laura K. Nelson
  • Impact: Best paper (e-business cluster) finalist at INFORMS 2019
  • Funding: CMU GuSH Grant / CMU GSA-Provost Grant
  • Talks (2020): Stanford Computational Sociology / MIT IC2S2 / SCECR / Marketing Science Conference / CIST
• • Interpretable Deep Learning for Text Exploration
  • Dokyun Lee, Emaad Manzoor, Zhaoqi Cheng (firstname α-β)
  • Under review (first round) at at Information Systems Research
  • Resources: Website, Preprint
• • Linguistic Determinants of Persuasive Message Elaboration
  • Emaad Manzoor, Dokyun Lee, George H. Chen, Alan L. Montgomery
  • Work in progress
  • Talks (2019): Marketing Science Conference / CIST / INFORMS Workshop on Data Science
• • Uncovering Latent Biases in Text
  • Emaad Manzoor, Nihar B. Shah
  • Under review at AAAI 2021
  • Resources: Preprint
  • Talks (2020): INFORMS Workshop on Data Science

---

Publications

• • Detecting Attackable Sentences in Arguments
  • Yohan Jo, Seojin Bang, Emaad Manzoor, Eduard Hovy, Chris Reed
  • Empirical Methods in Natural Language Processing (EMNLP), 2020
  • Resources: Paper, Code & Data
• • Expanding Taxonomies with Implicit Edge Semantics
  • Emaad Manzoor, Dhananjay Shrouty, Rui Li, Jure Leskovec
  • The Web Conference (WWW), 2020
  • Resources: Website, Paper, Code & Data, Slides, Video
  • Impact: External press, Production launch, Patent filed in Oct. ‘19
  • Funding: CMU GSA-Provost Grant
• • xStream: Outlier Detection in Feature-Evolving Data Streams
  • Emaad Manzoor, Hemank Lamba, Leman Akoglu
  • Knowledge Discovery and Data Mining (KDD), 2018
  • Resources: Website, Paper, Code, Data, Slides, Video, Poster
  • Impact: Assigned reading in York University’s Data Mining
  • Funding: CMU GSA-Provost Grant
• • RUSH! Targeted Time-limited Coupons via Purchase Forecasts
  • Emaad Manzoor, Leman Akoglu
  • Knowledge Discovery and Data Mining (KDD), 2017
  • Resources: Paper, Code, Poster
  • Funding: PNC Center for Financial Services Innovation
• • Fast Memory-Efficient Anomaly Detection in Streaming Heterogeneous Graphs
  • Emaad Manzoor, Sadegh M. Milajerdi, Leman Akoglu
  • Knowledge Discovery and Data Mining (KDD), 2016
  • Resources: Website, Paper, Code, Data, Slides, Video, Poster
  • Impact: Assigned reading in the Advanced Computer Security (UIUC, Spring ‘20) and Advanced Data Mining (University of Michigan, Winter ‘19) graduate courses
  • Funding: DARPA Transparent Computing

---

Manuscripts

• • Scheduling Broadcasts in a Network of Timelines
  • Emaad Manzoor, Haewoon Kwak, Panos Kalnis
  • Master’s Thesis (KAUST), 2015
  • Resources: Website, Paper, Slides, Video

---

Note on Publication Venues

Computer science research is disseminated primarily via conference proceedings. WWW and KDD are flagship conferences of the data mining subdomain (ranked A* by CORE). EMNLP is one of the top conferences in the natural language processing and computational linguistics subdomain (based on the Google Scholar h5-index).

Academic business research is disseminated primarily via journal publications. Management Science, Marketing Science, Information Systems Research and Management Information Systems Quarterly comprise the highest tier of journals in the information systems and quantitative marketing subdomain (ranked by the Financial Times).