Analytics
buzzwords are in the ascendency. Buzzwords can reflect froth and
pretension (google "buzzword bingo"), but they can also reflect real
phenomena - companies are making billions off the analytic value in their
data. Statistics.com offers over a dozen courses related to analytics and
data mining; a good place to start is "Introduction to Predictive
Modeling".
"Introduction
to Predictive Modeling" focuses on five predictive analytics techniques:
k-nearest neighbors, classification and regression trees (CART), neural nets,
logistic regression and multiple linear regression. It will also cover
the use of partitioning to divide the data into training data (to build a
model), validation data (to assess and fine tune models) and test data (to
predict the performance of the final model). For more details please visit at http://www.statistics.com/predictive_modeling/.
Who Should Take
This Course:
Analysts of
business data, consultants, MBAs seeking to update their knowledge of
quantitative techniques, managers who want to see what predictive modeling can
do, and anyone who wants a practical hands-on grounding in basic predictive
modeling techniques.
Course Program:
Course outline: The course
is structured as follows
SESSION 1: Introduction
- Core ideas in data mining
- Supervised and unsupervised
learning
- The steps in data mining
- SEMMA
- Preliminary steps
- Sampling
from a database
- Pre-processing
and cleaning the data
- Partitioning
the data
- Building a model
- An example
with linear regression
- K-nearest neighbor
SESSION 2: Classification
- Judging the performance of
classification algorithms
- Classification trees
- Logistic regression
- Lift
SESSION 3: Neural nets
- Neural nets
- Comparing different models
SESSION 4: Prediction
- Multiple linear regression
- Regression trees
The instructor,
Anthony Babinec, is the president of AB Analytics and served previously as
Director of Advanced Products Marketing at SPSS; he worked on the marketing of
Clementine and introduced CHAID, neural nets and other advanced technologies to
SPSS.
Participants can ask questions and exchange
comments with Dr. Phadke via a private discussion board throughout the period.
The course takes place online at statistics.com in a series of 4 weekly lessons
and assignments, and requires about 15 hours/week. Participate at your own
convenience; there are no set times when you are required to be online.
For Indian participants statistics.com accepts registration for its courses
at special prices in Indian Rupees through its partner, the Center for
eLearning and Training (C-eLT), Pune.
Call: 020 66009116
Websites:
