Home
Current Positions
Associate Professor at UTHealth (University of Texas Health Science Center at Houston)
Member at Tibyan Organization for Artificial Intelligence
Visiting Professor at University of Tripoli
My past employers include Addis Ababa University, OHSU, Trapit, Mayo Clinic, Starkey Laboratories, Honeywell Research, and Medtronic. For more details (and up-to-date info), check out my LinkedIn profile.
Objectives
I'm trying to empower people to improve society through the wise use of digital data. That means I want to develop a whole ecosystem for data science: business, education, and research. Right now, the thrust is in Libya, where despite an ongoing instability there is an educated and energetic middle-class population, alongside a dearth of businesses in advanced information technologies.
Research Interests
There are some foundational interests that guide the research directions I take on:
Computational semantics. I am interested in distributional/geometric models of semantics and in ontological semantics. Each type of representation may provide complementary benefits in NLP applications. I am also interested in compositionality and similarity in these semantic frameworks.
Lesser-resourced languages. During my time in Ethiopia (2017-2018), I began to work on deep learning NLP for Semitic languages. This interest and collaborative work has continued, and will likely expand over time, especially since I live somewhere where the majority language is not English.
Clinical information extraction/retrieval. Much of the important information in electronic medical records (EMR) is in free-text clinical notes rather than form fields. With clinical NLP, this information may be accessed and used to improve patient care. I am also interested in implementing the systems that make this information available to clinicians and researchers, hence my interest in information retrieval.
I secured and led an NLM-funded R01 (at OHSU) focused on information retrieval. I also had an NIAID-funded R21 (with Mayo Clinic) on aggregating patient information over time.
Teaching
I offer some introductory material on Natural Language Processing through Libya AI's Facebook Page and YouTube Channel.
In Winter Term 2016, I taught CS 562/662 Natural Language Processing at OHSU. I wrote a blog post about my experiences, advocating for a new teaching framework called techied.
In Spring 2018, I taught ITSC-5C22 Social Network Analysis at Addis Ababa University, and wrote another blog post about this very cross-cultural teaching experience.
I have also taught CS 555/655 Analyzing Sequences at OHSU.
Feel free to clone/use/fork my techied repository and build off of it!
Resources
cTAKES: Apache Clinical Text Analysis and Knowledge Extraction System. Originally developed at Mayo Clinic, this is a customizable UIMA-based pipeline that takes clinical text as input and gives lots of different clinical IE/NLP output.
NLTK: Natural Language Toolkit. Easy-to-use toolbox for natural language processing, written in Python. Includes great introductory material for those new to NLP. Good stuff, but I am not a developer.
keras: The deep learning toolkit that I typically use, now incorporated into TensorFlow.
Education
University of Minnesota, Minneapolis, MN
PhD, Computer Science (Natural Language Processing)
2006-2010
University of Minnesota, Minneapolis, MN
MS, Electrical Engineering (Digital and Wireless Communications)
2004-2006
Duke University, Durham, NC
BSE, Electrical Engineering and Biomedical Engineering
2000-2004
Other links
Computational Semantics in Clinical Text (CSCT), a workshop that I organized.
Some important wisdom.
A secondary passion.