Simplifying the Structural Recursion of the Data Funnel Interface

Abstract

The paper describes recent improvements made to the structural recursion of the Data Funnel Interface. This end user interface organizes the raw data in a database table into a series of nested data lists that guide users to information. The new developments outlined in the article streamline the recursive algorithm and the data files that it produces. The program logic constructs a new SELECT statement on each recursive pass to retrieve a unique subset of attribute data from the table. Next, it transforms the subset’s structure into a structure of nested data lists that users can navigate in the interface. The branching data model (BDM) provides the framework for combining these two objectives into a single, recursive algorithm. The data BDM features a detailed description of the data mapping between two table attributes as one-to-many data relationships. The input data for the recursion is a list of table attributes that models the data funnel network. Each new subset of data in the series folds into the former one until it reaches the final subset, a primary key value. The prior algorithm draws on the depth-first tree growing methodology of the OHDS structure to manage these subsets. The paper introduces a new numbering scheme to organize the subsets of data. This improvement assigns a unique list number to each new subset when adding it to the database table. This capability replaces the need for the OHDS in the table. It also simplifies the data funnel software and makes it easier to maintain.