David D. Lewis, PhD

My expert witness work is in two areas: software patent infringement cases and electronic discovery in the law. My work in patent cases draws on my 30 years of experience as a computer scientist in information retrieval, machine learning, computational linguistics, and applied statistics. Before taking up consulting, I held research positions at the University of Chicago, Bell Labs, and AT&T Labs. I am an active researcher, collaborating with university faculty and publishing peer-reviewed articles. I was made a Fellow of the American Association for the Advancement of Science in 2006. My work in e-discovery focuses on the design and statistical evaluation of technology-assisted review (TAR) processes. I have been a consulting or testifying expert in negotiations and disputes on predictive coding, including in the da Silva Moore, Kleen Products, Actos, and FHFA cases. In addition to expert witness work, I do due diligence and advisory work for companies and investors. I also design algorithms, software, processes, and inventions. I frequently teach and speak in my areas of interest. Applications I've worked on include search, categorization, clustering, filtering, routing, mining, extraction, authorship attribution and identification, parsing, and other text analytics tasks. Industries I've consulted to include web search, computational advertising, information security, spam filtering, customer relationship management, data mining, venture capital, biotech, defense, nonprofits, open source, government, and start-ups. Technologies I've worked with include both supervised learning (naive Bayes, logistic regression, support vector machines, decision trees, neural networks, nearest neighbor classification or k-NN, relevance feedback AKA continuous active learning or CAL,...) and unsupervised learning (clustering, LSI, LSA, LDA, factor analysis, dimensionality reduction,...), plus a variety of statistical techniques.