| .NET Time Series Foundation on Sho |
| Written by Janet Swift |
| Friday, 18 November 2011 |
|
A set of .NET components for time series analysis recently made available for public download, will be of interest to NET developers who are building Business Analytics, Business Intelligence and Data Mining applications. A three-man team of researchers in the Machine Learning Department of Microsoft Research, Alex Bocharov, Bo Thiesson and Chris Meek, has released Beta 2 of Time Series Foundation (TSF), a .NET toolset for exploring new algorithms in time series analysis and forecasting.
According to its download page on the Microsoft Research site: TSF is based on state space model methodology that includes all types of exponential smoothing, some autoregressive algorithms, and innovative algorithms for event detection and calendar event impact prediction. TSF implements an Excel interop layer and offers an Excel add-in that exposes a large subset of TSF functionality through the Excel ribbon UI. It is suggested that TSFcould be used with time series data for
In each of these application areas TSF can be used to understand and predict how the data evolves in time. New features in Beta 2 include forecasting the effect of calendar and recurring events, outlier/event detection methods and the TSF Excel Add-in.
TSF is based Sho, on an interactive environment for data analysis and scientific computing from Microsoft Research that lets you connect scripts (in IronPython) with compiled code (in .NET) to enable fast and flexible prototyping. The environment includes libraries for linear algebra as well as data visualization that can be used from any .NET language, as well as a feature-rich interactive shell for rapid development.
More information
To be informed about new articles on I Programmer, subscribe to the RSS feed, follow us on Google+, Twitter or Facebook or sign up for our weekly newsletter. 
|
| Last Updated ( Friday, 18 November 2011 ) |
