Introduction (Definition of Predictive analytic):
Predictive analytics is a combination of many
statistical techniques from predictive, machine learning, modelling and data
mining; it uses historical and current data facts to make predictions and
decisions for future and unknown events. (Nyce, 2007).
Nowadays predictive analytics is used in diverse types
of industries like healthcare, finance, pharmaceuticals, automotive,
manufacturing and aerospace. (Savkovic & Subrahamanya, n.d.)
Predictive analytics can benefit executives, managers
and other decision-makers by providing tools to study the possible options and
take the best possible decision. (Martin, 2014)
One of the most common problems that faces the people
working on the model is: Are they using the right technologies? How to make
sure if we are going to get the wanted results? Because when using the wrong
technologies even if we had the right data, we will get the wrong results which
will affect the whole business, its strategy, and decisions.
Notes: Introduction is missing, check part 1
in the marking scheme
To create a predictive analytic model for your data
and based on your requirement and goal, you have to choose the right technology
to do your job so that you will not end up with wrong or unwanted results that
will end your project and leave your business with a high costly investment in
software and time and nothing to show for it.
Being part of the data analytic community, working
with data analysis, and creating models for our data is crucial; and it can be
risky and many times if we didn’t study our data right or if we choose the
wrong technology we usually end up with a messed up model and unwanted result
.so choosing the right tool for the right data for the job is what matters
The aim of this study is to state and clarify the
basics of predictive analytic model, how to create a data analytic model based
on our data and or target, and how to not end up with a messed up unwanted
model. The importance of this study is to stop getting the wrong model and to
start getting better results through our work. It’s expected that after
finishing this study we will get a very good knowledge and foundation about
data predictive analysis, how to avoid creating the wrong model what
technologies are available to do a predictive data analysis model, and how to
choose the right technologies and tools to create our data analytic model based
on our available data.
Many challenges, issues, and constraints I may get
through my study especially during searching and finding for real case scenario
and to find real and ready applications were as many as possible cases will be
available with their corresponding steps.
study we will try to find cases and examples from the real world problems and
applications where we can show different cases of successful and failure
solutions, study samples of both cases and how they started with their project,
what technologies they used, what was there starting data and desirable result,
and what was the real result that they got to compare between the results and
the technologies used.
In this study the research questions will be a series
of questions were the ones responsible of creating the predictive analytic
model have to take into consideration before and during their work, which are:
of predictive data analytic models and its applications.
are the different available technologies?
on my data and my target, which technologies are better?
• What are
the different types of predictive data analysis models?
and challenges when creating a predictive analysis model
• How to
choose the right technology?
data analytic have several applications in various applications, were companies
are implementing models to their industries. An example for were data analysis
is used is in travel industry were travel agencies are using data predictive
analysis to create recommender systems for travel and to find the perfect
combination of hotels and flights based on previous flights and hotel
reservation data. (Agost, 2016)
application were data predictive analysis is very used is in online marketing
and the ability to sell products and services with minimum exposure. So new algorithms
are being implemented that uses data analysis techniques to predict what type
of ad they have to show, were to place their ads and when to view their ad to
get the best attraction from the customer and increase their chance that the
viewer will be more interested in this product. (Agost, 2016)