Never before has there been so much information available for human consumption.
As this flow of information increases businesses must migrate from overcoming information scarcity to coping with information overload.
Lucky Sort grew out of the team's collective experience in academia, health care and finance, all of which pointed to a grand opportunity to leverage the statistics of language to empower knowledge workers.
Before Lucky Sort, Noah was Director of Research and Development for Qmedtrix where he oversaw construction of machine learning and visualization platforms to detect fraud and abuse in medical reimbursements.
The TopicWatch concept evolved from Noah's earlier work at the Reed Artificial Life Lab in the Center for Advanced Computation, focusing on the evolutionary dynamics of technology by applying language processing and network analysis techniques to US patents.
In addition to his role at Lucky Sort, Noah serves as a Collaborative Researcher for the Advanced Computation Group at Apple, Inc. and advises several other companies on data analysis problems.
Dr. Norman Packard Ph.D is former CEO and co-founder of Prediction Company, which was acquired by Union Bank of Switzerland (UBS) in 2005 for its data management and trading system.
His role at Lucky Sort follows over two decades of experience in chaos theory, learning algorithms, predictive modeling of complex systems, statistical analysis of evolution, artificial life, and complex adaptive systems.
Prior to his move from academia to business in 1991, Dr. Packard was associate professor of Physics at the University of Illinois, Urbana-Champaign.
Homer oversees development of Lucky Sort's data pipeline and database. This frequently entails integrating open-source projects into custom solutions - Homer has contributed to numerous Clojure, Redis, JavaScript, and R projects, and founded the Portland R User’s Group.
Before founding Lucky Sort, Homer worked with Noah, leading development of a distributed medical billing analytics system that applied statistical and machine learning methods to bill data.
Devin leads development of Lucky Sort's TopicWatch interfaces. He's been an active iOS developer since day-one of the App Store, delivering iOS applications for clients ranging from Whole Foods to Air New Zealand, the Democratic Party and most recently, public radios.
Devin brings 10 years of professional web and native development experience, with an emphasis on Cocoa/Objective-C, Ruby on Rails/Sinatra, HTML and JavaScript - always with a bent to the effective and beautiful presentation of dense information.
Devin is a member of the Reed Artificial Life Lab, where he worked with Noah on machine learning algorithms, data visualization tools, and the high-performance, multi-platform manipulation of very large datasets.
TopicWatch is a set of tools that enable data driven decision making through insight from social media, government filings, news and commentary.
TopicWatch allows you to shift the atom of information consumption from individual documents to aggregate patterns that give context making large volumes of text useful and navigable.
With TopicWatch you can import your own text content to analyze alongside material from Lucky Sort or one of our partners.
TopicWatch subscribers can access the platform from the web, iPad or API interfaces, giving you flexibility to move between modes of analysis seamlessly, either on your own or with a team.
Whether you're a macro finance analyst or a director of business development, you have hunches about trends in the market. Use the TopicWatch platform to easily construct a set of phrases that act as proxy variables for trends of interest.
In addition to using sets of phrases to measure a trend these sets can be used to surface content. Individual documents can be identified and linked to from TopicWatch, allowing any analysis of high level patterns to quickly track back to underlaying content.
The iPad touch interface allows analysts and decision makers to rapidly sift through information: at the office, at home, or on the road.
Rapid interactive exploration of information makes you an instant expert.
The TopicWatch web client provides a widely accessible interface for analyzing the big text data sets of your choice. This is the solution for users who wish to use TopicWatch from a desktop GUI environment.
Embed and share restricted web views that provide interactive exploration of events or topics.
The TopicWatch R client is a thin layer between the TopicWatch API and the R programming language. R allows users to conduct sophisticated custom analysis.
Coming in May 2012, the TopicWatch applications for iOS and Web are designed for use by anyone who needs to be rapidly informed. TopicWatch is designed to be intuitive for non-programmers.
TopicWatch can be used directly from the API. The API can be used with custom wrappers, or with the TopicWatch R package: TopicWatchr
The TopicWatch API is designed to be used by software developers and quantitative analysts alike.
For some applications security and privacy are top concerns. Lucky Sort provides solutions to enterprise users that wish to leverage TopicWatch visual or programmatic tools in a private environment with an appliance that can run inside any infrastructure.
TopicWatch is currently in closed beta, but you can fill out a beta signup to request early access. TopicWatch will be available in both metered and monthly plans.
We're adding beta users as we develop and enhance our platform. We look forward to sharing TopicWatch with you!