Data defines the model by dint of genetic programming, producing the best decile table.

A Genetic Model to Identify Titanic Survivors
Bruce Ratner, Ph.D.

Since the fatal night of April 15, 1912, when the White Star liner Titanic hit an iceberg and went down in the mid-Atlantic, fascination with the disaster has never abated. In recent years, interest in the Titanic has increased dramatically as a result of the discovery of the wreck site by Dr. Robert Ballard. The ninety-four year old tragedy has become a national obsession. Any fresh morsel of information about the sinking is savored. The purpose of this article is to satisfy the appetite of the Titanic aficionado by building a 'Titanic' model to identify survivors genetically - not with their DNA, but with the  machine-learning alternative model to the logistic regression model – the GenIQ Model© – so when Titanic II sails we will know beforehand who will be most likely to survive an iceberg-hitting outcome odds of 2.0408e-12 to 1. This analysis shows that GenIQ is both an excellent data ming tool as it uncovers the patterns within the Titantic dataset, and an efficient model building method as it automatically determines the assumption-free, nonparametric Titanic model form.


Titanic Dataset
There were 2,201 passengers and crew aboard the Titanic. Only 711 persons survived, resulting in a 32.2% survival rate. See Table 1. For all persons we know their:
  1. GENDER - two categories - female or male
  2. CLASS - four categories - first, second, third or crew
  3. AGE - two categories - adult or child


The Titanic GenIQ Model Identification of Survivors - You'll be surprised who survived!

For more information about this article, call Bruce Ratner at 516.791.3544 or 1 800 DM STAT-1; or e-mail at
Sign-up for a free GenIQ webcast: Click here.