Profile

Mossaab Bagdouri

Mossaab Bagdouri

Ph.D., Senior Data Scientist

Address: Walmart Labs,
860 W California Ave,
Sunnyvale, CA 94086

Expertise in: Information Retrieval, Machine Learning and Natural Language Processing.

Education

2004 - 2007

Ingénieur d'Etat en Informatique

ENSIAS (Rabat, Morocco)

Professional Experiences

2017 - Present

Data Scientist

@WalmartLabs
  • Deep query understanding for e-commerce.
2011 - 2017

Research Assistant

CLIP Lab, UMIACS, University of Maryland at College Park
  • Thesis topic: Cross-platform question answering in social networking services.
  • Use of CLIR to answer non-factoid questions from multilingual social media.
  • Optimal allocation of annotated data for training and testing text classifiers.
2016 (4 months)

Intern

IBM TJ Watson Research Center
  • Deep networks for ranking user-generated answers to non-factoid questions. Tools: Keras, Theano.
2007 - 2009

Project Engineer

Maroc Telecom (Morocco)
  • Integration of the Internet and Lineline Information Systems (BSCS).
  • Project management of offers and promotions for enterprise products.
  • BSCS interfacing with external applications.
2007 (5 months)

Intern

SIS Consultants (Morocco)

Design and implementation of an E-gov web site realizing the dematerialization of the invitations to tender procedure. Tools: JEE, JSF, MySQL.

Projects

2009 - present

A multilingual news aggregator for the Arab world

Tools: JSF, PrimeFaces, Lucene

Maghress, Djazairess, Turess, Sudaress, Sauress and Yemeress crawl hundreds of online newspapers to get the latest news, classify and rank them before presenting the most important ones in the main page.

Papers in Conference Proceedings

[Paper]
[Presentation]

What Questions Do Journalists Ask on Twitter?

Hasanain, Maram, Mossaab Bagdouri, Tamer Elsayed, and Douglas W. Oard (May 2016)

In: Proceedings of the First Social Media in the Newsroom workshop .
Cologne, Germany, pp.127–134.

Sequential Testing in Classifier Evaluation Yields Biased Estimates of Effectiveness

Webber, William, Mossaab Bagdouri, David D Lewis, and Douglas W Oard (July 2013)

In: Proceedings of the 36th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval.
Dublin, Ireland, pp.989-998.

Language identification for creating language-specific Twitter collections

Bergsma, Shane, Paul McNamee, Mossaab Bagdouri, Clayton Fink, and Theresa Wilson (June 2012)

In: Proceedings of the Second Workshop on Language in Social Media.
Montreal, Canada, pp.65-74.

[Paper]
[Presentation]

“Beacons of hope” in decentralized coordination: learning from on-the-ground medical twitterers during the 2010 Haiti earthquake

Sarcevic, Aleksandra, Leysia Palen, Joanne White, Kate Starbird, Mossaab Bagdouri, and Kenneth Anderson (Feb. 2012)

In: Proceedings of the ACM 2012 conference on Computer Supported Cooperative Work .
Seattle, WA, USA, pp.47-56.

Book Chapters

Promoting Structured Data in Citizen Communications During Disaster Response: An Account of Strategies for Diffusion of the ‘Tweak the Tweet Syntax’

Starbird, Kate, Leysia Palen, Sophia B Liu, Sarah Vieweg, Amanda Hughes, Aaron Schram, Kenneth M Anderson, Mossaab Bagdouri, Joanne White, Casey McTaggart, and Chris Schenk (2011)

Ed. by C. Hagar. Chandos information professional series. Chandos.

Theses

Cross-Platform Question Answering in Social Networking Services

Bagdouri, Mossaab (2017)

Topic Modeling as an Analysis Tool to Understand the Impact of the Iraq War on the Iraqi Blogosphere

Bagdouri, Mossaab (2011)