
Innovation Analyst, Text Miner, Inventor, Instructor, Duke University
Raleigh-Durham, North Carolina Area

Innovation Analyst, Text Miner, Inventor, Instructor, Duke University
Raleigh-Durham, North Carolina Area
In August 2007 I became the Research Analyst and Technologist for the Kimberley Jenkins Chair in New Technologies and Society at Duke University. As research analyst I assist the Jenkins Chair, Tim Lenoir, with research largely focused around patent landscape analysis and virtual collaboration tools. In my role at Duke I contribute original research and initiate multidisciplinary collaborations. Our major collaborations include allpatents.org, SparkIP.com, 3D visualization of RDF patent data, and the Virtual Peace initiative.
My main intellectual interest is in the automated acceleration of discovery, an area which draws heavily from a wide range of disciplines including text mining, information retrieval, information science, innovation studies, statistics, linguistics, computer science, cognitive science, economics, and philosophy. At Duke I also serve as an instructor for several courses run by the Jenkins Chair in new media studies.
In December 2006 I completed my MS in Information Science from UNC along with a minor in Computer Science and a Certificate of Specialization in Bioinformatics. My research in the program emphasized text mining, particularly for different types of highly heterogeneous health information corpora. My master's paper is forthcoming in print from VDM Verlag Dr. Mueller.
While in graduate school I maintained a consulting business supporting clients on web development and design projects. I also worked at ibiblio as a java developer and sys admin while in school. I also interned at SAS where I created a novel enterprise architecture classification framework based on automated analysis of user behavior and content.
Before school I worked as a software developer for a rapidly growing startup called TManage. Previous to TManage I was the Web Operations Manager for ICCVAM and NICEATM, two interagency toxicological testing groups originating out of the NIEHS.
text mining, innovation, text analytics, data mining, machine learning, knowledge discovery, natural language processing, enterprise search, web search, information retrieval, data visualization, information architecture, user experience (UX) design, graphic design, software development, database design & development, db administration, java, oracle, MySQL, SQL, clustering, classification, support vector machines (SVM), drug discovery, bioinformatics, project management, systems analysis
(Educational Institution; 10,001 or more employees; Research industry)
August 2007 — Present (1 year 1 month)
As part of the Kimberly J. Jenkins Chair of New Technologies and Society at Duke University, I develop novel tools and methods for innovation analysis using large document collections and economic data as inputs. The aim of our research is not only to identify the process by which products successfully form from innovations but also to automate that identification. I am also collaborating with SparkIP (http://sparkip.com) who is utilizing my novel cluster naming method that automatically labels an ever-evolving cluster set without relying on an ontology. I am working on the Virtual Conflict Resolution virtual gaming project, a radically innovative gaming solution for education in crisis resolution. Drawing from my various backgrounds in literature, biotechnology, text mining, linguistics, philosophy, computer science, information science, and statistics, I also teach courses in the ISIS Program at Duke.
(Sole Proprietorship; Myself Only; Information Technology and Services industry)
January 2007 — August 2007 (8 months)
With Hypothia I have been architecting a science discovery generation engine that will initially be purposed for the analysis of the entire Medline collection. I have also been working on developing advanced techniques for automatic ontology construction and for forecasting and decision modeling in areas as diverse as finance, professional football, consumer health, and drug discovery. You can read more about my vision for Hypothia as well as related areas of interest at http://hypothia.wordpress.com/. Many of the R&D elements of Hypothia will soon begin to emerge in an organic way through my position at Duke University.
(Privately Held; 1-10 employees; Information Technology and Services industry)
January 2002 — December 2006 (5 years)
• Provided information architecture, web development, & web design services
• Consulted with clients on web strategies and information architecture standards and policies
• Development in Python, PHP, XHTML, JavaScript
• Project management, budgeting and operations, contracts & negotiations, and risk analysis
(Privately Held; 10,001 or more employees; Information Technology and Services industry)
June 2006 — August 2006 (3 months)
• Innovated advanced solutions for information architecture & search engine optimization (SEO) tasks for entire support.sas.com site
• Designed dynamic and static classification & navigation system prototypes for 100,000+ web support documents
• Consulted internal business clients and presented intelligent content management and metadata solutions
• Constructed a content analysis, search term analysis & reporting application suite prototype in Java & MySQL as a solo effort – a groundbreaking information architecture and metadata management tool that provides intelligent automated consumer-responsive analysis, content organization, and site management of very large and heterogeneous enterprise web content collections
(Educational Institution; 1-10 employees; Information Technology and Services industry)
August 2003 — August 2005 (2 years 1 month)
• Developed XML transformation utilities in Java and XSLT for Charles Dickens 3-D book reader application
• Reverse-engineered, restored, and reversioned archaeology educational software package written in Java
• Consulted on SQL and web analytics development
• Linux systems administration for hosting services with extensive bash and sed scripting
(Privately Held; 201-500 employees; Information Technology and Services industry)
January 2000 — December 2001 (2 years)
• Developed, supported, and extended enterprise B2B data integration system in Java, XML, & Oracle for Fortune 100 clients
• XSL technical lead for company’s leading revenue-generating product and continuing project management of B2B extensibility
• Database administration & development in Oracle and MS SQL Server
• Developed web-based sales forecasting tool prototype tied to CRM system
• Web interface design consulting for internal and external clients
(Public Company; 51-200 employees; Biotechnology industry)
May 1998 — January 2000 (1 year 9 months)
• Organized new Federal health regulatory policy agencies (ICCVAM and NICEATM) which focused on toxicological testing methods regulation as contractor
• Organizations set toxicology animal testing standards across 17 Federal agencies and harmonized standards with over 50 nations
• Designed, developed, & supported the original ICCVAM and NICEATM website in XHTML, JavaScript
• Provided technical documentation and translation services for toxicology test standards research
(Public Company; Information Technology and Services industry)
1997 — 1998 (1 year)
• Re-engineered enterprise-wide Mac-to-PC migration process
• Re-engineering led to early project completion & $500k savings
(Information Technology and Services industry)
1996 — 1997 (1 year)
MS Information Science, Information Science, Bioinformatics, Computer Science, Text Mining, 2002 — 2006
• Minor in Computer Science with a Certificate of Specialization in Bioinformatics
• Master’s Thesis on the application of text mining to pharmacogenomics-based drug discovery (http://hdl.handle.net/1901/341)
• Accumulated 70 credit hours of study with 27 credits dedicated exclusively to data mining and text mining of health-related data
• Designed mining applications & executed text mining & data mining experiments
• Wrote text and data mining software in Java, Oracle, Weka and sed along with thousands of SQL queries and baseline statistics in R and Excel
• Research focus on semantics-based feature representation of heterogeneous web text collections for automatic clustering and classification
• CS coursework in basic programming, algorithms, and web development in Java, C++, JSP, DOM, and JDOM
• High Pass (H) average (>95%); 4.0 average in computer science minor coursework
• Margaret Ellen Kalp Fellowship (declined)
Premed, Genetics, Biochemistry 1995 — 1997
• 1997 University Scholar in Vienna, Austria
• 3.8 GPA
• 60 credits across core competencies in biology, chemistry, microbiology, biochemistry, statistics and genetics, with graduate-level study in Genetics
BA, Philosophy, 1989 — 1993
• Honors Program Scholar (class size average of 8; all coursework led by university’s top scholars)
• Focused on cognitive science with extensive study in linguistics and mathematical logic
• 3.4 GPA with 3.7 in major
HS Diploma, 1985 — 1989
text mining, linguistics, programming, poetry, visual design, boating, snorkeling
HIMSS, AMIA, ACM, Subsubpoetica Americana
2003 Margaret Ellen Kalp Fellowship, UNC School of Information and Library Science (declined); 2005 NC Triangle Independent Arts Award; 1997 University Scholar, NC State; 1993 Honors AB, UNC; Undergraduate Honors Program Scholar, UNC