iManageDataTM
         High Performance Data Pre-Processing
 		  

Feature Detail

Here's a description of the many data handling features of iManageData.  In many cases, these features are vital to successful data modeling and analysis. 

Data Sampling Schemes    
    Why Data sets can be large.  In those cases where it is not practical to model ALL the data you would like to take a representative sample.  iManageData offers these alternative methods to sample your data...
 
    What You Can Do Fixed Data Sources (files, database queries, workbooks)
  • Every One (load all data)
  • Every N'th (load every nth row of data from the source)
  • First N rows of data
  • Last N rows of data
  • Randomly Select N rows of data

Real-Time Data Sources (PI Data Archive, Control Systems, ...)

  • Last N records of a specified interval size (X seconds, minutes, hours, days...)
Binary Conversions
    Why Sometimes you would like to create binary flags indicating whether a condition exists or simplify your data into major categories.  For example, you can take a continuously valued variable and break it up into new, simpler indicators based on ranges.  "Income" can be mapped into new variables: HighIncome, MediumIncome and LowIncome.  Also, sometimes your data is text and you would like to create new variables that are numeric for purposes of modeling.
 
    What You Can Do
  • Convert continuous variables into new binary variables based on range
  • Convert single textual values into new numeric flag variables.  For example, if a column of data called "CarType" has a value of "Porsche" a new variable can be created called "Porsche" that has a "1" in the column where that occurs.
  • Convert a group of textual values into new numeric flag variables.  For example, if a column of data called "CarType" has a value of "Porsche" or "BMW" or "Corvette", a new variable can be created called "ExpensiveCar" that has a "1" in the column where any of those textual values occur.
Data Substitutions
    Why Sometimes your data has odd, yet consistent values that you would like to substitute other values for, or you need to convert text to numeric within the same column of data.
    What You Can Do
  • Substitute Numeric Values for Other Numeric Values
  • Convert Textual Values to Numeric Values
  • Convert a Range of Numeric Values to a New Column of Binary Flags
Filters
    Why Most data has "outliers", values that are unusually high or low, or you have periods of time in your data that you would like to focus upon or exclude.
    What You Can Do
  • Horizontal (filters range on all records) 
  • Vertical (filters indicated set of records) 
  • Free Form (filters ranges on specified records)

Support for including/excluding inside/outside filter and eight (8) other exclusionary methods

 
Transforms  
    Why Often you would like to perform mathematical functions on your data.  iManageData offers a suite of math functions you can use.
    What You Can Do
  • Change 
  • Percent Change 
  • Binary +1/-1 
  • Binary 0/1 
  • Sum
  • Difference 
  • Product 
  • Division 
  • Ratio 
  • Delay 
  • Simple Moving Average
  • Absolute Value
  • X power N 
  • X Power Y 
  • SQRT 
  • Exponent 
  • Minumum 
  • Maximum 
  • N-Log 
  • Natural Log 
  • N-Root 
  • SineTangent 
  • ArcSine 
  • ArcCos 
  • ArcTan 
  • SinH 
  • CosH 
  • TanH 
  • ArcSinH 
  • ArcCosH 
  • ArcTanHSecant 
  • CoSecant 
  • CoTangent 
  • ArcSecant 
  • ArcCoSecant 
  • ArcCoTangent 
  • SecantH 
  • CoSecantH 
  • CoTangentH 
  • ArcSecantH 
  • ArcCoSecantH 
  • ArcCoTangentH

Legal Policy | Privacy Policy | Contact Us
Copyright © 1995-2007 BioComp Systems, Inc.. All rights reserved.
IntelliDynamics is a registered trademark of BioComp Systems, Inc.
Last modified: Sunday February 13, 2005.
Serving the web since Feb 2, 1996