DSA Meetup: Predictive Analytics & Automated Intelligent Outlier Detection
University of Colorado Boulder - Tuesday November 12, 2013 @ 6:30pm MST
Predictive analytics is a hot topic in the data science and business communities. This presentation will cover the the latest state-of-the-art techniques for data science practitioners and real case studies for decision makers in business who need to understand how predictive analytics can help achieve durable competitive advantage.
Predictive analytics turns data into valuable, actionable information to determine the probable future outcome of an event or a likelihood of a situation occurring - and encompasses a variety of techniques from machine learning, algorithms, data mining, statistics, modeling and game theory that analyze current and historical facts to make predictions about future events.
Predictive analytics utilizes the power of data. Computers can learn from data how to predict the probable future behavior of individuals. While perfect prediction is not possible, determining probable behavior for certain profiles and demographics is useful. It can also be dangerous if predictive limitations are not understood or if negligently practiced. It can also backfire if folks get spooked and feel privacy has been violated.
In business, predictive models exploit patterns found in historical and transactional data to identify risks and opportunities. Models capture relationships among many factors to allow assessment of risk or potential associated with a particular set of conditions, guiding decision making for candidate transactions.
Michael Walker is a managing partner at Rose Business Technologies, a professional technology services and systems integration firm. He leads the Data Science Professional Practice at Rose. Mr. Walker received his undergraduate degree from the University of Colorado and earned a doctorate from Syracuse University. He speaks and writes frequently about data science and is writing a book on Data Science Strategy for Business. Learn more about the Rose Data Science Professional Practice at http://bit.ly/10TgVHG.
Automated Intelligent Outlier Detection
Bad data is common in the atmospheric sciences. Often instruments are operated in sever weather conditions and in remote locations with limited power and communication infrastructure. Furthermore, the devices themselves might be research grade instrumentation and need constant attention. Typically, data is collected during a field program and is analyzed after the end of the field campaign. The intelligent outlier detection algorithm was developed to quality control time-series data collected by anemometers located on the mountains near Juneau Alaska. The intent was to develop an algorithm that mimics the human ability to identify suspect data – regardless of a priori knowledge about a given time series. Essentially, the algorithm segments a time series using basic image processing techniques and auto-correlation. The details of the algorithm will be presented in the context of several example time series.
Andrew Weekley is currently an analyst in the Strategic Energy Analysis Center at the National Renewable Energy Lab (NREL). Currently; Mr. Weekley is working on synthetic solar time series data used in solar integration models. Prior to NREL, Mr. Weekley was a software engineer in the Research Applications Laboratory at the National Center for Atmospheric Research (NCAR) where he developed image-processing algorithms and participated in numerous research projects. Mr. Weekley received his BA in physics and mathematics from the CU Boulder.