About Me

I completed my PhD under Professor James F. Allen in the Computer Science department at the University of Rochester (URCS). Currently I am working as a Sr. Research Scientist in the R&D of Nuance Communications.

My work primarily falls under Natural Language Processing/Understanding (NLP/NLU) and Big Data. I worked with Professor James F. Allen at URCS on problems related to temporal information processing (extraction, question answering, evaluation, etc.). I also collaborated with Professor Jeffrey P. Bigham on making information accessible. I had done multiple research internships in various domains on Big Data problems, such as, game prediction using social media (Yahoo! Research), medical NLP (Microsoft M3L), and car dialog systems & large scale data mining (Bosch RTC). Currently at Nuance Communications, I am working in the NLU Core Architecture group and the Data Mining group of Mobility Division.

I have way too many profiles scattered on the internet. Check out my website naushad.uzzaman.com to find about all of my activities in one place.


Area of Interest

Artificial Intelligence, Big Data Analysis, Data Mining, Natural Language Processing/Understanding, Social Media Text Analysis, Information Extraction, Temporal Information Processing, Dialog Assistant, and Medical NLP.

CV, Research and Teaching Statements

If you are interested, here are my: CV, Research Statement, and Teaching Statement.
full LinkedIn profile (with recommendations at the end) [LinkedIn].

Naushad UzZaman's Google Citation Naushad UzZaman's bibliography View Naushad UzZaman's profile on LinkedIn

Latest News

  • February, 2014. Co-Founded SocialTrendly. Become a fan of our Facebook page and keep updated about what's coming!
  • November 18, 2013. Participated in Boston Startup Weekend 2013. My idea on making a platform for companies to invest in social awareness formed as Dollar By Dollar. We got an honorable mention in the event.
  • May 18, 2013. Official commencement at University of Rochester [photos].
  • March 25, 2013. Finally done with the TempEval-3! It's been a ride. Thanks a lot to my coorganizers and the participants.
  • November 30, 2012. Published TwitterPaul paper in arXiv. Find out about it here.
  • November 19, 2012. Started a new blog - DecipherData. I am thinking of using my after-office hours for deciphering the public data. So I plan to use this blog to document my journey and also to hopefully report some interesting findings eventually.
  • July 30, 2012. Started at Nuance Communications as a NLP Research Engineer.
  • July 20, 2012. Successfully defended my PhD! Now I am Dr. UzZaman :-)
  • July 9, 2012. Our work on evaluation of temporal information understanding using temporal question-answering got accepted at ICSC 2012.
  • May 23, 2012. Our work on merging temporal annotations got accepted at TIME 2012.
  • March 14, 2012. Another reason for me to celebrate the Pi day (3/14). Accepted a job offer from Nuance Communications. I will be joining Nuance in August 2012.
  • December 8, 2011. Officially looking for job now. If you are interested, here is my Research Statement and CV.
  • December 7, 2011. Had a productive semester working with Hector Llorens. We prepared for organizing TempEval-3, and also worked on merging temporal annotations and temporal question answering.
  • September 1, 2011. Completed internship at Yahoo! Research, Barcelona. Worked on game prediction using social media.
  • May 13, 2011. Won the 3rd prize at UR's Center for Research Computing poster session for Multimodal Summarization for people with cognitive disabilities in reading, linguistic and verbal comprehension. [poster]
  • February 16, 2011. In an article about IBM's work on enabling the WWW with speech, our work on automatically illustrating complex sentences as multimodal summaries combining picture, simple sentence structure and summarized text is described with a few words, our [paper] [presentation].
  • [more news] [photography news]

Recent Publications

  • Naushad UzZaman, Hector Llorens, Leon Derczynski, James Allen, Marc Verhagen, James Pustejovsky. 2013. SemEval-2013 Task 1: TempEval-3: Evaluating Events, Time Expressions, and Temporal Relations. In Proceedings of the 7th International Workshop on Semantic Evaluation (SemEval 2013), Atlanta, USA, 2013. [paper] [presentation]

  • Leon Derczynski, Hector Llorens, Naushad UzZaman. 2013. TimeML-strict: clarifying temporal annotation. CoRR arXiv:1304.7289.

  • Naushad UzZaman, Roi Blanco and Michael Matthews. 2012. TwitterPaul: Extracting and Aggregating Twitter Predictions. CoRR arXiv:1211.6496.

  • Naushad UzZaman. Interpreting the Temporal Aspects of Language. PhD Thesis, Department of Computer Science, University of Rochester, July 2012. [paper] [presentation]

  • Naushad UzZaman, Hector Llorens and James Allen. Evaluating Temporal Information Understanding with Temporal Question Answering. Proceedings of IEEE International Conference on Semantic Computing, Italy, September 2012. [paper]

  • Hector Llorens, Naushad UzZaman and James F. Allen. Merging Temporal Annotations. Proceedings of 19th International Symposium on Temporal Representation and Reasoning, Leicester, UK, September 2012. [paper]

  • Naushad UzZaman, Hector Llorens, James F. Allen, Leon Derczynski, Marc Verhagen, James Pustejovsky. 2012. TempEval-3: Evaluating Events, Time Expressions, and Temporal Relations. CoRR arXiv:1206.5333.

  • Naushad UzZaman and James F. Allen. Temporal Evaluation. Proc. of The 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies (Short Paper), Portland, Oregon, USA, June 2011. [paper] [poster] [toolkit]

  • Naushad UzZaman and James F. Allen. Event and Temporal Expression extraction from raw text: first step towards a temporally aware system. International Journal of Semantic Computing, 2011. [paper]

  • Naushad UzZaman, Jeffrey P. Bigham and James F. Allen. Multimodal Summarization of Complex Sentence. Proc. of International Conference on Intelligent User Interfaces (IUI), Palo Alto, California, 2011. [paper] [presentation]

  • Naushad UzZaman, Jeffrey P. Bigham and James F. Allen. Multimodal Summarization for people with cognitive disabilities in reading, linguistic and verbal comprehension. 2010 Coleman Institute Conference, Denver, CO, 2010. [abstract] [poster]

  • Naushad UzZaman and James F. Allen. Extracting Events and Temporal Expressions from Text. Fourth IEEE International Conference on Semantic Computing (IEEE ICSC2010), Pittsburgh, USA, September 2010. [paper] [presentation]

  • Naushad UzZaman and James F. Allen. TRIPS and TRIOS System for TempEval-2: Extracting Temporal Information from Text. International Workshop on Semantic Evaluations (SemEval-2010), Association for Computational Linguistics (ACL), Sweden, July 2010. [paper] [presentation]

  • Naushad UzZaman and James F. Allen. TRIOS-TimeBank Corpus: Extended TimeBank corpus with help of Deep Understanding of Text. Proceedings of The seventh international conference on Language Resources and Evaluation (LREC), Malta, May 2010. [paper] [presentation]

  • Naushad UzZaman, Jeffrey P. Bigham and James F. Allen. Pictorial Temporal Structure of Documents to Help People who have Trouble Reading or Understanding. International Workshop on Design to Read, ACM Conference on Human Factors in Computing Systems (CHI), Atlanta, GA, April 2010. [paper] [presentation]

  • Recent papers with previews and additional comments can be found in my publication page and Bangla language processing related papers can be found in my previous webpage. Citation records are available in Google Scholar page.


Research Jobs

Nuance Communications
Senior Research Scientist (July 2012 - Present)
Department: R&D - NLP Research
Group: NLU Core Architecture and Data Mining group
  • Designed and prototyped different modules of new architecture dialog assistant.
  • Automatically extracted named entities from just a few seed examples and lots of tweets or logs.
  • Automatically clustered similar queries and similar named entities from query logs.
  • Designed the ontology for named entities.

PhD program

  • Temporal Information Extraction: We implemented a hybrid system with linguistically motivated solutions and machine learning classifiers for extracting temporal information from raw text. We extract events, temporal expressions and classify temporal relations. Our system had a state-of-the art performance in TempEval-2.

  • Temporal Evaluation: We proposed a new metric for evaluation of temporal annotation. Our metric uses temporal closure to reward relations that are equivalent but distinct. It also measures the overall performance of systems with a single score, making comparison between different systems straightforward. Our intuitive and computationally inexpensive metric is used to evaluate participants in the premier temporal annotation shared task TempEval-3, and also temporal information processing shared task on clinical data - i2b2.

  • Temporal Question Answering: We proposed a temporal QA system that performed temporal reasoning and showed how it can be used to evaluate automated temporal information understanding. With temporal reasoning, our QA system can answer list, factoid and yes/no questions. TempEval-3 organizers are discussing now to use this evaluation methodology for temporal annotation shared task TempEval-4.

  • Creating 500K word corpus with temporal annotation and organizing TempEval-3: We teamed up with other participants of TempEval-2, Hector Llorensand Leon Derczynski, and the organizers of TempEval-2, Marc Verhagen and James Pustejovsky, to create a 500K word corpus with temporal annotation and to organize TempEval-3. We automatically annotated a 500K word corpus from news domain by merging our system outputs and human-reviewing a portion of it. We organized TempEval-3 with that corpus.

  • Merging Temporal Annotations: We proposed algorithms for merging multiple temporal annotations. With such a merging algorithm, we can merge multiple systems, multiple human annotations, or a combination of both. We evaluated our algorithm over TimeBank and AQUAINT corpora considering three state-of-the-art systems. The results of merging improve over individual systems and baseline merging algorithms. We will be using this merging tool to merge automated system outputs for TempEval-3 to create the 1-million word corpus.

  • Multimodal Summarization: We worked on illustrating complex sentences as multimodal summaries combining pictures, simple sentence structure and summarized text. MMS could be used to help people with cognitive disabilities, children, older people, or people whose first language is not English.

Research Internship

  • Game Prediction with Social Media at Yahoo! Research, Barcelona (2011): I explored the use of social media for game prediction. Our system reliably extracted predictions from tweets, identified the predictive power of individual users, ranked predictions and aggregated them to find the most likely outcome of upcoming games. We evaluated our results on the 2010 FIFA World Cup tournament using a corpus of over 1.8 billion tweets containing over hundred-fifty thousand predictions. A key benefit of our system is that it does not rely on domain dependent knowledge for prediction, rather extracts people's predictions from social media. As a result, the framework can be used to predict the outcome of other sport events, elections, product release dates and any other future events discussed in the social media. Find more here: TwitterPaul.

  • Information Extraction on Demand at Microsoft Medical Media Lab (2009): I implemented an Information Extraction tool that can learn different patterns from user's examples and extract the pattern instances from natural language texts in medical documents, e.g. release notes, radiology reports, etc. For pattern matching, I used the medical ontology (UMLS) to understand medical patterns like diseases, medicines, etc. and other general language features. My implemented tool was shared with the clinical staff at Washington Hospital Center and they found it useful.

  • In-Car Dialog System at Bosch RTC (2008): I developed a syntactic and semantic grammar for a commercial in-car dialog system. It was very similar to the famous iPhone's dialog assistant Siri. My work was in navigation, local business and MP3 domain.
  • Data Mining at Bosch RTC (2008): I worked on a large scale data mining problem for KDD cup 2008: Breast Cancer Detection. I experimented with different machine learning techniques, e.g. Support Vector Machine (SVM), Decision Tree, KNN, Bayes Net, Neural Network, etc. Finally built the system using SVM and also implemented the Feature Selection technique to improve the performance and reduce the computation time. Additionally, I worked on a recommendation system for Netflix contest.

Contact Information

Naushad UzZaman, PhD
Computer Science Department [Why URCS?]
University of Rochester

Web: http://www.naushad.uzzaman.com
Research Profile: www.cs.rochester.edu/~naushad
Photo blog: http://www.naushadzaman.com

naushadzaman - View my most interesting photos on Flickriver
Locations of visitors to this page