Getting started

• Interactively create flow diagrams for data reconciliation and data quality analysis problems with filtering and matching functions
• Easily plug and play the functional blocks and tune their parameters to analyze the processing done on data
• Rapidly generate customized pre-processing and matching functions for different domains of data
• Save and retrieve the flow diagrams in XML

• Take a subset of the dataset to be analyzed
• Build a flow diagram using the following guidelines to analyze the sample data
• Access the data from one or more files or databases
• Apply preprocessing or filtering functions to bring the data to a canonical format. The parameters to these functions can be tuned to obtain the desired results. Results can also be viewed at this intermediate stage
• Select matching functions to find records with matching columns, or a function to compute a distance between selected columns, such as edit distance. Again, the properties of functions can be tuned and the results can be viewed.
• Add a classifier, if needed, to separate the resulting records based on values of a column.
• Add more functional blocks until the desired results are obtained.
• The data results can be saved by adding a destination block to each output.
• The whole flow diagram can be run by using the Run item in the File menu.  This saves the results in all the destination blocks.
• Save the flow diagram in XML
• Use the flow diagram tested on the sample data, on the whole dataset.