Deep Data Analytics

Data, as “stuff,” are the starting point for all decisions. They are the basis for the information decision makers need. The information is buried inside the data and must be extracted. Extractions are easy for small data sets; means, standard errors, proportions, and simple charts suffice. Large data sets, so-called Big Data, present challenges best met with sophisticated tools and methodologies. This is Deep Data Analytics — the tools and methodologies used for analyzing Big Data.

Data Analytics Corp. uses a range of tools appropriate for crossing the Analytical Bridge to extract Rich Information from Big Data as well as Small Data.

This includes:

  • Data Visualization
  • Descriptive Statistics
  • Predictive Modeling
  • Machine Learning
  • Multi-Level Modeling
  • Data Reduction
  • Multivariate Analysis
  • Regression Analysis
  • Logistic Regression Analysis
  • Data Wrangling

Just as there is a continuum of information from Poor to Rich Information, so there is a corresponding continuum for analysis from Shallow to Deep Analysis. Shallow Analysis consists of simple means and proportions as well as pie charts, bar charts, and (in the marketing research space) cross-tabs. These are usually just reported in a Powerpoint presentation. Deep Analysis goes further and looks for relationships, trends, patterns, and anomalies. The goal with Deep Analysis is to look further into the data, to look beyond the obvious.

Information and Analysis Continuums

information and analysis continuums

Shallow Analytics

  • Means
  • Proportions
  • Pie Charts
  • Bar Charts
  • Tabs

Analytical Bridge

Deep Analytics

  • Regression Analysis
  • Predictive Modeling
  • Text Analysis
  • Multivariate Analysis
  • Multi-Level Modeling

Deep Data analysis is a complex process that is dynamic, not static. Dynamic means that you always:

  1. Link tables and graphs for drill-down.
  2. Filter data to subset them based on other variables.
  3. Drag & Drop variables onto and around a canvas, either tabular or graphic.
  4. Build models, either causal (i.e., using a member of the regression family) or probabilistic.
  5. Test hypotheses.
  6. Profile or simulate to study effects.
rich information
Close Menu