Data Mining Process: Case Study on Employee Attrition at Jet Airlines

This report presents a comprehensive analysis of the data mining process, using Jet Airlines as a case study to understand employee attrition. It details the three key stages: pre-modeling, modeling, and post-modeling. The pre-modeling phase involves data preparation, cleansing, and formatting by domain experts to create data models, including the creation of derived attributes. The modeling phase focuses on selecting and applying various mining activities, often requiring iterative exchanges with domain experts. The post-modeling phase evaluates the model's ability to meet business objectives and considers all business issues. The report highlights challenges such as managing large datasets and breaking heterogeneous populations into homogeneous groups. It also discusses the importance of selecting appropriate modeling techniques, designing tests, building models, and assessing them to achieve optimal results. The report concludes by emphasizing the importance of understanding model parameters and their application within data mining projects.

Running head: DATA MINING PROCESS
Data Mining Process
(Jen Airline)
Name of the student:
Name of the university:
Author Note

1DATA MINING PROCESS
Introduction:
The Jet Airlines established themselves as the biggest airline company at Singapore. Their
HR manager engaged the department of analytics for understanding the profile of the employees
who left the organization.
The following study demonstrates the stages of data modeling. This includes pre-modelling,
modelling and the post-modelling.
Pre-modelling stage:
The pre-modelling stage of data mining includes the data preparation. The domain experts
create the data models for modeling processes. The data are collected, cleansed and formatted by
them. This is because some of the functions of mining accept the data in a particular format. They
further build latest derived attributes like the average value (Wu et al., 2014). The potential business
problems arising in the pre-modelling phase is that the data get tweaked various times with no
estimated order. The preparation of data is done by the modeling tools through selecting the
attributes, records and selecting tables. However the meaning of the data does not get changed.
One of the queries faced by Jet Airline has been how to manage the large quantity of data
available into sensible structures. They are in the dilemma to break the huge heterogeneous
population into lesser homogeneous groups (Gupta, 2014).
Modelling phase:
The experts have been selecting and applying numerous mining activities. This is because the
various functions could be used for the same kind of data mining issues. Few mining activities need

2DATA MINING PROCESS
particular data types (Peña-Ayala, 2014). In this phase there is the often exchanging with the domain
experts from the pre-modelling phase is needed. This phase with the post-modelling phase is
coupled. They have been repeated various times for changing parameters till the optimal values gets
reached. As the ultimate modeling phase is finished, the high quality model is built. As the data is
already in a perfect shape, Jet Airlines might search for the helpful patterns in their data. The
modeling phase comprise of four jobs. They are the selection of modeling techniques, designing the
tests, the building of models and the assessing of the models.
Post modeling phase:
This includes the evaluation of the model. This is done by the data mining experts. As they
model gets unable to satisfy the demand, they return to the previous phase. Then they again create
the model by altering the parameters. This is done toll the optimal values are gained. As they get
satisfied with the model ultimately, Jet Airlines could retrieve the explanations of the business. Next
they need to evaluate some queries. The first one is whether the model has been achieving the
business objective. The next one is have Jet Airlines considered all the issues in their business
(Hofmann & Klinkenberg, 2013). As this phase ends, the experts take decision regarding how to
utilize the outcomes of data mining.
Conclusion:
As Jet Airlines gets comfortable with the data-mining career, they would clearly understand
how to search for the model parameters. They must be further aware of how to use it. These options
would vary broadly with the kind of model and the particular tool they have been using.

3DATA MINING PROCESS
References:
Gupta, G. K. (2014). Introduction to data mining with case studies. PHI Learning Pvt. Ltd..
Hofmann, M., & Klinkenberg, R. (Eds.). (2013). RapidMiner: Data mining use cases and business
analytics applications. CRC Press.
Peña-Ayala, A. (2014). Educational data mining: A survey and a data mining-based analysis of
recent works. Expert systems with applications, 41(4), 1432-1462.
Wu, X., Zhu, X., Wu, G. Q., & Ding, W. (2014). Data mining with big data. IEEE transactions on
knowledge and data engineering, 26(1), 97-107.