Report on Enterprise Business Intelligence using SAS Enterprise Miner

This report provides an overview of enterprise business intelligence using SAS Enterprise Miner (SAS EM), a data analytics software designed for rapid predictive and descriptive modeling. It outlines the five steps of the SAS SEMMA approach: sampling, exploration, modification, modeling, and assessment, highlighting the software's graphical interface and support for various analytical techniques, including time series analysis, decision trees, and regression. The report also details data handling processes within SAS EM, emphasizing its user-friendly interface and automated data scoring capabilities. Furthermore, it discusses the advantages of SAS EM, such as its rapid predictive modeler add-on for MS Excel, support for R language integration, and the factory miner tool for automated model development. The document concludes by referencing relevant research and applications of SAS Enterprise Miner in data mining.

Running head: ENTERPRISE BUSINESS INTELLIGENCE
ENTERPRISE BUSINESS INTELLIGENCE
Name of the Student:
Name of the University:
Author Note:

1ENTERPRISE BUSINESS INTELLIGENCE
Introduction
The SAS enterprise miner or SAS EM is an advanced level data analytics software
that is designed to help users of the software develop predictive and descriptive models
quickly in a short period of time using a more streamlined process of data mining. The
enterprise miner’s interface is more graphical which allows the users to move through the
five step of SAS SEMMA approach more logically. These five steps are: exploitation,
sampling, modelling, modification and assessment. The users are allowed to select their
process flow by selecting the suitable tab from the software’s toolbar and then use the drag
and drop method to bring the needed nodes into the pallet. The SAS enterprise miner
software supports many different types of techniques and algorithms which includes time
series, market basket analysis, decision trees, link analysis, linear and logistic regression,
neural networks and web path analysis. The SAS enterprise miner’s architecture of the client
server helps data analysts and business users to join together, collaborate and share their work
and other models. The SAS EM client software is designed to run on windows 8, 7 and 10.
There are multiple options that can be used as hosts for this software which includes IBM
AIX R64, Solaris on Sparc version 10, HP-UX on Itanium 11i version 3 and IBM z/OS
V1R12 or higher.
Data mining process for SAS EM
The main steps which are followed in the SAS enterprise miner’s data mining process are:
1. Sample of the data is created by diving in to multiple datasets. The sample should be of a
size big enough to have all the needed information but compact enough to process. The data
preparation tools are included in this step. These tools are used for append, data import, filter
and merge.
2. The data set is then explored and searched for trends, relationships and anomalies to gain
better understanding of the dataset. This step uses tools that are required for graphical
exploration, variable selection methods and statistical reporting.
3. The data can then be modified by selecting, transforming and creating the variables to put
focus on the process of selecting the model. The step uses tools that are needed for value
recording, interactive binning and missing value handling.

2ENTERPRISE BUSINESS INTELLIGENCE
4. The data can then be modelled using analytical tools that are used to train the machine
learning model for predicting desired outcome. This part uses techniques like decision trees,
partial least squares, neural networks, nearest neighbour and logistic and linear regression.
5. The data then can be assessed and its usability and usefulness can be evaluated from the
findings of the data mining process. This step uses tools required for cut-off analysis,
comparing models, score code management, decision support and computing of new fit
statistics.
Data handling in SAS enterprise miner
The data handling can be done by the SAS enterprise miner has multiple steps and
processes. Each process implemented depends upon the type of data being handled. The data
mining process used in SAS EM is mainly run by a process flow diagram which can be
created using drag and drop technique and is structured using the SEMMA categories. The
graphical user interface of the software is designed in a simple manner and can be used even
by a person with no statistical expertise. The SAS EM processes and handles the data in a
fully automated process and provides complete scoring of the data process in C, SAS, and
Java and PMML languages.
Advantages of using the SAS enterprise miner
The three major advantages of the SAS enterprise miner are:
1. The SAS rapid predictive modeller is a part of the SAS EM software that can be run as an
add-on for MS Excel thus allowing business users to conduct predictive modelling from
within their excel spreadsheets. The models developed by using the rapid predictive modeller
can also be later customised by data analysts using the EM.
2. The developers and an analysts who code in the R language can use the enterprise miner
process flow to transform and integrate the models they write about. The SAS enterprise
miner has support for the Hadoop environment and can be used with the help of the SAS
scoring accelerator. These scoring accelerators are available for a different types of platforms
like Pivotal, Hadoop, IBM Netezza, Teradata, Oracle and DB2.
3. The SAS software also comes with a tool called the factory miner which comes as an add-
on and help users develop models quickly with automated, web based frameworks. This tool

3ENTERPRISE BUSINESS INTELLIGENCE
helps the users to construct, run and retain more than one predictive models over a wide
variety of customer parts and businesses.
References
Abraham, C. and Poston, M., 2015. Assessing Real World Applications of Data Mining With
SAS Enterprise Miner (EM): A Technical Report for Teaching the Big Data
Generation. Journal of Emerging Trends in Computing and Information Sciences, 6(9).
Hassan Zadeh, A., Schiller, S. and Duffy, K., 2016. Teaching Analytics: A Demonstration of
Association Discovery with SAS Enterprise Miner.
Matignon, R., 2007. Data mining using SAS enterprise miner(Vol. 638). John Wiley & Sons.
Pham, M., Tanjil, M. and Ruppert-Stroescu, M., 2016. Application of gradient boosting
through SAS Enterprise Miner 12.3 to classify human activities. In SAS Global Forum.
Sarma, K.S., 2013. Predictive modeling with SAS Enterprise Miner: Practical solutions for
business applications. SAS Institute.
Wang, R., Lee, N. and Wei, Y., 2015. A case study: Improve classification of rare events
with sas® enterprise miner™. Retrieved on June, 30(2016), pp.3282-2015.