Browse Prior Art Database

Transformation of Demographic Data for Enhanced Analytics

IP.com Disclosure Number: IPCOM000225699D
Publication Date: 2013-Feb-27
Document File: 8 page(s) / 95K

Publishing Venue

The IP.com Prior Art Database

Abstract

The concept relates to adding a psychometric dimension to the demographic data using transformation units those use machine learning between multiple psychometric instruments, and then modify the psychometric profiles with social networking analysis if available. For adding psychometric dimension, appropriate psychometric instrument is identified based on target analysis and if needed an equivalence is derived from transformation from other psychometric instruments by an optimal chaining.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 30% of the total text.

Page 01 of 8

Transformation of Demographic Data for Enhanced Analytics

Data mining and On-Line Analytical Processing (OLAP) products both target the business intelligence community. Both these analytics methods have "demographic data" as an element or in OLAP parlance- dimension. Granular data as well as summarized data have patterns. Summarized data along different dimensions may have different patterns and there might be a relation between them.

Organizational data needs typically have multiple dimensions. Multidimensional analysis is well suited to answering the What questions such as what areas of my business are doing well? What products are failing to meet forecast? Mining is better suited to the Why questions. For example, multidimensional analysis could quickly identify that car loans exceeded profit goals in the first quarter, but mining would be more effective in identifying that the reason was due to increased activity from women aged 18-25 who are single head of households. Let's look at another example. Multidimensional analysis quickly identifies that sales of outdoor leisure-wear in the Northeast territory was down by 5%. But mining may be the best way to find that it was due to the combined effect of lower revenues from vests, pants, and specialty items during different time periods. What these examples show is that the combination of multidimensional and data mining methods yield more than using either technology alone.

While OLAP is by its nature retrospective, data mining is prospective. OLAP is driven by experts and is deductive in nature. Data mining is driven by the data itself and is inductive in nature. In either data mining or OLAP, better the data collection, better is the "actionable" result.

Demographic Element or Dimension:
Above and beyond just the OLAP and data mining integration, there is another dimension to this analysis.

Taking example of Audio CDs, if one were to conclude that new movies sale was dependent only on high population density, but did not account for the fact that this large density was those of GenY, and NOT of senior citizen, one would reach a wrong conclusion.

Similarly, just as average age is important, in understanding customer behavior, there are many demographic parameters those are taken into account- gender, income levels, educational levels, etc. Slicing and dicing of data along these various sub-dimensions are possible and are done.

The concept relates to adding a psychometric dimension to the demographic data using transformation units those use machine learning between multiple psychometric instruments, and then modify the psychometric

1


Page 02 of 8

profiles with social networking analysis if available. For adding psychometric dimension, appropriate

psychometric instrument is identified based on target analysis and if needed an equivalence is derived from transformation from other psychometric instruments by an optimal chaining.

Feature 1:

1. What we are proposing is adding an eleme...