David Talby's Home Page

I’m a CTO at John Snow Labs, helping healthcare & life science companies put AI to good use. My interests include natural language processing, applied artificial intelligence in healthcare, and responsible AI.

Talks

NLP in Healthcare

Healthcare-Specific Large Language Models: Capabilities & Benchmarks. Keynote at Predictive Analytics World for Healthcare, June 2023
2023 Trends in Applied NLP in Healthcare: Large Language Models, No-Code, and Responsible AI. AMIA Natural Language Working Group, April 2023
New Frontiers in Applied Natural Language Processing in Healthcare. Keynote at the Predictive Analytics World for Healthcare Conference, June 2022.
Clinical Question Answering on Real-World Patient Data. Keynote at the Healthcare NLP Summit 2022.
State of the Art Healthcare NLP in One Line of Code. Ai4 Healthcare Summit, DSS Miami, and Turkish AI Week. February 2022.
Automated Medical Question Answering in Practice. O’Reilly AI Superstream: NLP in Production, 2022.
New Lessons Learned Building Natural Language Processing Systems in Healthcare. AI DevWorld 2022.
Turning Medical Expert Knowledge into Responsible Language Models. K1st World, October 2022.
Natural Language Processing for Pediatric Mental Health. Children’s Hospital of Orange County Research Day 2022.
Advancing the State of the Art in Applied Natural Language Processing. Keynote at the Healthcare NLP Summit 2021.
Automating Clinical Data Abstraction From Unstructured Documents Using Spark NLP. Ai4 Healthcare Summit, May 2021.
State of the art Clinical and Biomedical Natural Language Processing. Nvidia GTC 2021.
Lessons Learned Applying Natural Language Processing for Healthcare AI. Global Big Data Conference, Sep 2020.
Automated and Explainable Deep Learning for Clinical Language Understanding at Roche. Spark+AI Summit, June 2020
Applying state-of-the-art natural language processing for personalized healthcare. ODSC East 2020 and Ai4Healthcare 2020.
Start Of The Art Natural Language Understanding In Healthcare. Global Big Data Conference, August 2018.
Natural Language Understanding in Healthcare: State of the Art NLP, Machine Learning & Deep Learning with Open Source Software.
Prediction Analytics World Healthcare, in Las Vegas, Nevada, June 2018.

Natural Language Processsing

Question Answering on Text & Tables in Corporate Responsibility Reports. Data Science Salon NYC, June 2023
State-of-the-Art Natural Language Processing with Spark NLP 4, Data+AI Summit 2022 and PyData Miami 2022.
Advanced Natural Language Processing with Free Low-Code and No-Code Tools., Global Artificial Intelligence Conference, March 2022.
State of the Art NLP with Free and Open-Source Software, keynote at the 2nd NLP Summit, October 2021.
The 2020 Trends for Applied Natural Language Processing, keynote at the 1st NLP Summit, October 2020.
State of the Art Natural Language Processing at Scale, keynote at AI DevWorld, October 2020.
Building Complete State-of-the-art Natural Language Processing Projects with Free Software. Scale by the Bay, October 2021.
Advanced Natural Language Processing with Spark NLP Spark+AI Summit 2020 and Data+AI Summit 2021.
Natural language understanding at Scale with Spark NLP, Strata Data NYC 2019 and ODSC East 2019.
Lessons Learned Building AI Systems That Understand Natural Language - Global AI Conference, April 2019
Spark NLP: Extending Spark ML to Deliver Fast, Scalable, and Unified Natural Language Processing, Spark+AI Summit 2019.
State of the Art Natural Language Processing at Scale Spark+AI Summit 2018 and ODSC West 2018.
Natural Language Understanding at Scale with Spark NLP.
Half-Day Tutorial with Claudiu Branzan and Alex Thomas. Strata New York, September 2018
State of the Art Natural Language Processing at Scale.
With Alex Thomas. Spark+AI Summit, San Francisco, June 2018.
Intelligent, high-accuracy fact extraction from long financial documents.
With Saif Addin Ellafi and Paul Parau, at Strata Data London, May 2018.
Extending Spark NLP: Training your own deep-learned natural language understanding models.
O'Reilly AI, Beijing, April 2018.
Natural Language Understanding at Scale on Apache Spark.
Open Data Science Conference, in Boston, May 2018 and San Francisco, November 2018
Natural language understanding at scale with spaCy, Spark ML, and TensorFlow.
Half-Day Tutorial with Claudiu Branzan and Alex Thomas. Strata New York, September 2017; Strata California, March 2018; and Strata London, May 2018.
Comparison of Java-based, Python-based and Spark-based Natural Language Processing Libraries.
Open Data Science Conference, Santa Clara, November 2017
Semantic Natural Language Understanding at Scale using Spark, Machine Learned Annotators & Deep-Learned Ontologies.
With Claudiu Branzan, at Strata Hadoop San Jose, CA, March 2017 & at Spark Summit East, Boston, MA, February 2017.
Reading between the lines: Using NLP, machine learning & deep learning algorithms to extract meaning from text.
Global AI Conference, Santa Clara, CA, January 2017.
Natural language understanding pipelines: From keywords & grammar to inference & prediction.
US Data Science Summit, in San Francisco, CA, July 2016.
Semantic natural language understanding with Spark Streaming, UIMA & machine learned ontologies.
With Claudiu Branzan, at Strata + Hadoop World, in London, May 2016 & in New York, September 2016.

Responsible AI

Data Driven Healthcare

Accelerating the adoption of precision oncology with artificial intelligence (AI). With Elise Berliner, Lulu Lee, and Vishakha Sharma, at the Real-World Evidence Industry Summit of Experts [R.I.S.E.], May 2023
Real-World Lessons from Applying Natural Language Processing to Personalized Healthcare. O’Reilly AI Superstream, March 2021
How SelectData uses AI to better understand home health patients.
With Alberto Andreotti, Stacy Ashworth and Tawny Nichols. Strata Data Conference, New York, September 2018.
Improving patient flow forecasting at Kaiser Permanente.
With Santosh Kulkarni. Strata Data Conference, in San Jose, California, March 2018.
Clinical natural language understanding at scale.
EU Data Science Summit, in Tel Aviv, Israel, June 2016.
Moving Beyond Templates and Coercion to Improve Physician Documentation.
With Jill Wolf, at the 23rd Annual WEDI National Conference, in Hollywood, CA, May 2014.
Data driven approach to revenue capture process improvement.
With Gene Boerger, at the AHIMA 85th Convention, in Atlanta, GA, October 2013.
Data driven models to minimize hospital readmissions.
With Miriam Paramore, at the 2013 Strata Rx Conference, in Boston, MA, September 2013.

Data Science in Production

Model Governance: A Checklist for Getting AI Safely to Production ODSC Europe 2020 and ODSC Europe 2021.
What to Expect When You Are Putting AI in Production? - ODSC East 2019
Executive Briefing: Why machine-learned models crash and burn in production and what to do about it – at Strata Data NYC 2019, ODSC East 2019, and Strata Data London 2018.
When models go rogue: Hard earned lessons about using machine learning in production, 3rd Data Science Summit Europe, May 2017
Executive Briefing: Why machine-learned models crash and burn in production and what to do about it.
Strata Data Conference, in London, May 2018 and New York, September 2018.
Executive Briefing: What you must know to build AI systems that understand natural language.
O'Reilly AI, in San Francisco, September 2018.
When models go rogue: Hard earned lessons on using machine learning in production.
Strata Data Conference, in London, May 2017; New York, September 2017; and San Jose, March 2018.
Machine learning in production: Lessons from the field.
At the Big Data, Analytics & Machine Learning Israeli Innovation Conference, in Ramat Gan, Israel, May 2015.
Concept Drift, A/B Testing, Online Measurement, Adversaries & Other Risks when Operating Data Science Models.
Global Data Science Conference, Santa Clara, CA, March 2017.
Engineering Management for Data Science Systems & Teams: Battle-Tested Best Practices.
Enterprise Data World, in San Diego, CA, April 2016.
Lessons learned productizing data science: Software Development vs. Model Development Best Practices.
Big Data for Executives, in Santa Clara, CA, March 2016.

AI Platform Architecture

How to build an open source data science platform.
Half-day tutorial at TDWI Anaheim, California, August 2018.
Building a new predictive model & API in 30 minutes.
With Claudiu Barbura, at PAPIs.io — The Predictive APIs and Apps Conference, in Barcelona, Spain, November 2014.
Building an intelligent big data app in 30 minutes.
With Claudiu Barbura, at the Strata Barcelona Conference, in Barcelona, Spain, November 2014.
Lessons Learned from Embedding Cassandra in an enterprise-grade big data platform.
With Claudiu Barbura, at Cassandra Day Seattle, in Bellevue, WA, July 2014.
Leveraging a big data infrastructure to accelerate the data science workflow.
At the 5th Timisoara Big Data Meetup, in Timisoara, Romania, June 2014.

Machine Learning for Fraud Detection

Hunting Criminals with Hybrid Analytics.
At Data by the Bay, San Francisco, May 2017; Global Data Science Conference 2016, in Santa Clara, CA, March 2016; and IBM Datapalooza, in Seattle, WA, February 2016.
Architecting a predictive, petabyte-scale, self-learning fraud detection system.
Global Predictive Analytics Conference, Santa Clara, CA, March 2017.
Online Predictive Modeling of Fraud Schemes from Multiple Live Streams.
With Claudiu Branzan, at Spark Summit East, in New York, NY, February 2016.
Online fraud detection: A reference architecture for adversarial learning.
At MLConf Atlanta, in Atlanta, GA, September 2015.
Hunting Criminals with Hybrid Analytics, Semi-supervised Learning & Agent Feedback.
With Claudiu Branzan, At the Smart Data Conference, in San Jose, CA, August 2015 & at at Strata + Hadoop World, in London, UK, May 2015.
Active learning from streams of graph, language & time series signals.
With Claudiu Branzan, At the Data Science Summit & Dato Conference, in San Francisco, CA, July 2015.

Research

Natural Language Processing in Healthcare

Arshaan Nazir, Thadaka Kalyan Chakravarthy, David Amore Cecchini, Rakshit Khajuria, Prikshit Sharma, Ali Tarik Mirik, Veysel Kocaman, and David Talby. LangTest: A comprehensive evaluation library for custom LLM and NLP models. Software Impacts vol. 2, Feb. 2024.
Veysel Kocaman, David Talby, and Hasham Ul Hak. Beyond Accuracy: Automated De-Identification of Large Real-World Clinical Text Datasets. Value in Health, volume 26 issue 12, Dec. 2023.
Veysel Kocaman, Youssef Mellah, Hasham Haq, and David Talby. Automated De-Identification of Arabic Medical Records. Proceedings of ArabicNLP 2023, Association for Computational Linguistics, Dec. 2023.
Karargyris et. al., Federated benchmarking of medical artificial intelligence with MedPerf. Nature Machine Intelligence, vol. 5 issue 7, pp. 799-810, July 2023.
Vishakha Sharma, Alex Thomas, Vishnu Vettrivel, David Talby, and Antoaneta Vladimirova. Developing emerging AI technologies to empower precision medicine: Advanced queries, search and biomedical predictions leveraging disparate oncology unstructured data compendia. Annals of Oncology, vol. 33, October 2022.
Vikas Kumar, Lawrence Rasouliyan, Veysel Kocaman, and David Talby. Detecting Adverse Drug Events in Dermatology Through Natural Language Processing of Physician Notes. ICPE 2022, the 38th International Conference on Pharmacoepidemiology and Therapeutic Risk Management (ICPE), Copenhagen, Denmark, August 2022.
Hasham Ul Haw, Veysel Kocaman and David Talby, “Deeper Clinical Document Understanding Using Relation Extraction”. SDU (Scientific Document Understanding) workshop at AAAI 2022.
Hasham Ul Haw, Veysel Kocaman and David Talby, “Mining Adverse Drug Reactions from Unstructured Mediums at Scale”. Proceedings of the 6th International Workshop on Health Intelligence at AAAI 2022.
Veysel Kocaman and David Talby, “Accurate Clinical and Biomedical Named Entity Recognition at Scale”, Software Impacts, July 2022.
Hasham Ul Haw, Veysel Kocaman and David Talby, “Connecting the dots in clinical document understanding with Relation Extraction at Scale”, Software Impacts, April 2022.
Veysel Kocaman and David Talby, “Spark NLP: Natural Language Understanding at Scale”. Software Impacts, January 2021.
Veysel Kocaman and David Talby, “Biomedical Named Entity Recognition at Scale”. CADL 2020 (International Workshop on Computational Aspects of Deep Learning) at ICPR 2020.
Veysel Kocaman and David Talby, “Improving Clinical Document Understanding on COVID-19 Research with Spark NLP”. SDU (Scientific Document Understanding) workshop at AAAI 2021.
Veysel Kocaman, Gursev Pirge, Bunyamin Polat and David Talby, “Biomedical Named Entity Recognition in Eight Languages with Zero Code Changes”. SDU (Scientific Document Understanding) workshop at AAAI 2021.
Veysel Kocaman, Cabir Celik, Damla Gurbaz, Gursev Pirge, Bunyamin Polat, Halil Saglamlar, Meryem Vildan Sarikaya, Gokhan Turer, and David Talby, “Social Media Mining for Health (#SMM4H) with Spark NLP”. Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task, pages 44–47, Gyeongju, Republic of Korea, 2022.
Ali Emre Varol, Veysel Kocaman, Hasham Ul Haq, David Talby, “Understanding COVID-19 News Coverage using Medical NLP”. Proceedings of the Text2Story 2022 Workshop, Stavanger, Norway.

Parallel Computer Scheduling & Workload Modeling

David Talby, User Modeling of Parallel Workloads: Predicting and Optimizing the Use of Parallel Computers, VDM Verlag Publishing, 2009.
David Talby, Dror G. Feitelson and Adi Raveh, "A Co-Plot Analysis of Logs and Models of Parallel Workloads". ACM Transactions on Modeling and Computer Simulation, vol. 12 no. 3 art. 12, July 2007. ©Copyright 2007 by ACM, Inc. Definitive version available from the ACM Digital Library.
David Talby, Dan Tsafrir, Zviki Goldberg and Dror G. Feitelson, "Session-Based, Estimation-less, and Information-less Runtime Prediction Algorithms for Parallel and Grid Job Scheduling". Technical Report 2006-77, School of Computer Science and Engineering, The Hebrew University of Jerusalem, August 2006.
Julia Zilber, Ofer Amit and David Talby, "What is Worth Learning from Parallel Workloads? A User and Session Based Analysis". International Conference on Supercomputing 2005 (ICS '05), June 2005.
David Talby and Dror G. Feitelson, "Improving and Stabilizing Parallel Computer Performance Using Adaptive Backfilling". 19th Intl. Parallel & Distributed Processing Symposium (IPDPS '05).
David Talby and Dror G. Feitelson, "Supporting Priorities and Improving Utilization of the IBM SP2 Scheduler Using Slack-Based Backfillingquot;. In 13th Intl. Parallel Processing Symp., pp. 513-517, April 1999. ©Copyright 1999 by IEEE. Definitive version available from the IEEE Computer Society Digital Library. Also available as a more detailed technical report.
David Talby, Dror G. Feitelson, and Adi Raveh, "Comparing Logs and Models of Parallel Workloads Using the Coplot Method". In Job Scheduling Strategies for Parallel Processing, pp. 43-66, Springer-Verlag, 1999. Lecture Notes in Computer Science Vol. 1659. ©Copyright 1999 by Springer-Verlag. Definitive version available from Springer LINK.
S. J. Chapin, W. Cirne, D. G. Feitelson, J. P. Jones, S. T. Leutenegger, U. Schwiegelshohn, W. Smith, and D. Talby, "Benchmarks and Standards for the Evaluation of Parallel Job Schedulers". In Job Scheduling Strategies for Parallel Processing, pp. 67-90, Springer-Verlag, 1999. Lecture Notes in Computer Science Vol. 1659. ©Copyright 1999 by Springer-Verlag. Definitive version available from Springer LINK.

Agile Software Development

David Talby and Yael Dubinsky, "Governance of an Agile Software Project". 2nd ICSE Workshop on Software Development Governance (SDG 2009), May 2009.
Yael Dubinsky, Orit Hazzan, David Talby and Arie Keren. "Transition to Agile Software Development in a Large-Scale Project: A Systems Analysis and Design Perspective". In the Advances in Management Information Systems (AMIS) Monograph Series, 2008.
David Talby, Lessons Learned from Scaling XP. Software Practice Advancement 2007, Cambridge, UK, March 2007.
David Talby, Orit Hazzan, Yael Dubinsky and Arie Keren, "Agile Software Testing in a Large-Scale Project". IEEE Software, Special issue on Software Testing - Jul/Aug 2006. ©Copyright 2006 by IEEE. Definitive version available from the IEEE Computer Society Digital Library.
David Talby, Orit Hazzan, Yael Dubinsky and Arie Keren, "Reflections on Reflection in Agile Software Development". Agile 2006 Conference, July 2006.
Yael Dubinsky, Orit Hazzan, David Talby and Arie Keren, "System Analysis And Design in a Large-Scale Software Project: The Case of Transition to Agile Development". 8th International Conference on Enterprise Information Systems (ICEIS 2006), May 2006.
Yael Dubinsky, David Talby, Orit Hazzan and Arie Keren, "Agile Metrics at the Israeli Air Force". Agile 2005 Conference, Denver, Colorado, July 2005.

Model Driven Software Engineering

David Talby, "The Perceived Value of Authoring and Automating Acceptance Tests Using a Model Driven Development Toolset". Fourth International Workshop on the Automation of Software Test (AST’09), May 2009.
David Talby, Enterprise-Scale Model Driven Development: Challenges after the initial adoption. Code Generation 2007, Cambridge, UK, May 2007.
David Talby, Model Driven Testing: Growing your own solution. Code Generation 2007, Cambridge, UK, May 2007.
David Talby, Ori Nakar, Noam Shmueli, Eli Margolin and Arie Keren, "A Process-Complete Automatic Acceptance Testing Framework". In Proc. of 2005 IEEE Intl. Conference on Software - Science, Technology and Engineering (SwSTE '05), February 2005. ©Copyright 2005 by IEEE. Definitive version available from the IEEE Computer Society Digital Library.
David Talby, Dotan Adler, Yair Kedem, Ori Nakar, Noa Danon and Arie Keren, "The Design and Implementation of a Metadata Repository". In Proc. of Intl. Council on Systems Engineering Israeli chapter conference, March 2002.

Patents

Healthcare

Jose Pablo Andreotti, Saif Addin Ellafi, and David Talby. Using unsupervised machine learning for automatic entity resolution of natural language records. US Patent 11783130, issued 2023.
Rodney Kinney, Michael Sandoval, Jonathan Cross, Bryan Tinsley, Manjula Iyer and David Talby, "Automatic generation of evaluation and management medical codes". Patent publication number US 20150039344 A1, filed in 2014.
Vishnu Vettrival, Penny Yee, Michael Sandoval and David Talby, "Automated system for handling files containing protected health information".
Patent publication number US 20160063187 A1, filed in 2015.
Rodney Kinney, Michael Sandoval, David Talby, Robert Payne, Bryan Tinsley and Alex Thomas, "Methods and systems for medical auto-coding using multiple agents with automatic adjustment". Patent publication number WO 2014046707 A1, filed in 2013.
Rodney Kinney, Michael Sandoval, David Talby, Gunjan Gupta and Manjula Iyer, "Methods and automated systems that assign medical codes to electronic medical records". Patent publication number WO 2014047051 A1, filed in 2013.

Data Science

Jose Alberto Pablo Andreotti and David Talby. Preprocessing images for OCR using character pixel height estimation and cycle generative adversarial networks for better character recognition. US Patent 11836969, issued 2023.
Jose Alberto Pablo Andreotti and David Talby, “Preprocessing images for OCR using character pixel height estimation and cycle generative adversarial networks for better character recognition”, patent publication number US 11176410 B2, granted 2021.
Wolf Kohn, Michael Sandoval, Vishnu Vettrivel, Jonathan Cross, Jason Knox, David Talby and Mike Lazarus, "Cooperative distributed control of target systems".
Patent publication number WO 2016025080 A1, filed in 2015.
Wolf Kohn, Michael Sandoval, Vishnu Vettrivel, Jonathan Cross, Jason Knox, David Talby and Mike Lazarus, "Determining control actions of decision modules".
Patent publication number WO 2016025080 A1, filed in 2015.
Gunjan Gupta, Aman Thakral, John Morris, Robert Payne, Michael Sandoval and David Talby, "Automated experimentation platform". Patent publication number WO 2015095411 A1, filed in 2014.
Gunjan Gupta, Wolf Kohn, Robert Payne, Aman Thakral, Michael Sandoval and David Talby, "Method and system for estimating values derived from large data sets based on values calculated from smaller data sets". Patent publication number WO 2015095405 A4, filed in 2014.
Radu Bisca, Michael Sandoval, Claudiu Barbura, Wolf Kohn and David Talby, "Method and system for searching and analyzing large numbers of electronic documents". Patent publication number US 20140250377 A1, filed in 2014.
Zhimin Chen, Eduardo Laureano, Renfei Luo, Tsheko Mutungu, Vivek Narasayya and David Talby, "Dictionary for hierarchical attributes from catalog items". Patent publication number US 8606788 B2, filed in 2011 and granted in 2013.

Software Engineering

David Talby and Scott McMaster, "Dynamic model based software application development". Patent publication number US 8949772 B1, filed in 2009 and granted in 2015.
Dan Tsafrir, Yoav Etsman, David Talby and Dror G. Feitelson, "System and method for backfilling with system-generated predictions rather than user runtime estimates". Patent publication number US 8261283 B2, filed in 2006 and granted in 2012.

Software

Spark NLP

John Snow Labs' Spark NLP is an open source text processing library for Python, Java, and Scala. It provides production-grade, scalable, and trainable versions of the latest research in natural language processing.

Healthcare NLP

Spark NLP for Healthcare is a commercial extension of Spark NLP for clinical and biomedical text mining. It provides healthcare-specific annotators, pipelines, models, and embeddings for clinical entity recognition, clinical entity linking, entity normalization, assertion status detection, de-identification, relation extraction, and spell checking and correction. It also includes over 2000 pre-trained healthcare models, that can recognize the entities such as clinical, drugs, risk factors, anatomy, demographics, and sensitive data.

Visual NLP

Visual NLP is another commercial extension of Spark NLP for visual document understanding (VDU) from images, scanned documents, photos, and files (PDF, DOCX, DICOM). It can extract information at scale from text, tables, charts, and forms.

LangTest

The open-source langtest Python library enables data scientists to test custom LLMs for all aspects of Responsible AI - and deliver language models that are more robust, fair, unbiased, representative, and accurate. The library can generate and run automated tests a majority of popular LLM & NLP frameworks.

Generative AI Lab

The Generative AI Lab is a no-code tool that enables teams of domain experts - like doctors and lawyers - to train, tune, and share custom AI models. It's designed for high-compliance environments and supports team & project management, tasks, workflows, versioning, analytics, AI-assisted annotation, and a private hub for reusing models, rules, and prompts.

Visual Co-Plot

A statistical analysis tool, tailored for datasets with few observations and many variables which may be intercorrelated. Co-Plot enables visually analysing observations, variables and the correlations between them together.

Courses

Software Design

Computer Science

Software Engineering Management

Software Team Leaders Workshop
The Industrial Survey
Marketing

Groups

John Snow Labs

John Snow Labs, an AI and NLP for healthcare company, provides state-of-the-art software, models, and data to help healthcare and life science organizations build, deploy, and operate AI projects.

Pacific AI

Pacific AI provides consulting CTO services for high-growth software companies, specializing in applying AI, big data and data science. I conduct technology due diligence, board reviews and AI strategy & architecture workshops.

Coalition for Health AI

The Coalition for Health AI is a community of academic health systems, organizations, and AI expert practitioners. Its mission is to provide guidelines regarding an ever-evolving landscape of health AI tools to ensure high quality care, increase credibility amongst users, and meet health care needs. I lead the fairness, equity, and bias mitigation workgroup at CHAI.

Forbes Technology Council

I'm a member of the Forbes Technology Council, an invitation-only community for CIOs, CTOs and technology executives. It's a curated network of successful peers, provides access to a variety of benefits and resources, and includes the opportunity to submit thought leadership articles and short tips on industry-related topics for publishing on Forbes.com.

Data Science in Healthcare

I started and run the Linkedin group for data science in healthcare. Doing my best to keep an open, evidence based and vendor neutral discussion, focused on recent research results and field case studies.

Awards

Best CTO of the Year - USA, 2023

Chief Technology Officer of the Year - US, 2022 & 2023

David Talby