Data Science
In order to store the data which grows
every day, there was a wide need for data science. The main aim of data science
was to find new solutions to store the big data successfully. We can say that
data science is the future data processing by using artificial intelligence. In
this article we will discuss the meaning of data science and its importance in
everyday life.
Data Science Definition
Data science is the way of discovering hidden
patterns which exist in the raw data by using a combination of algorithms,
tools and machine learning techniques. By processing the history of data, data
scientist can explain what is going on and also predict what may occur in the
near future by using various algorithms. Let us know what data science can do
in the field of making decisions and predictions.
§ Predictive Casual Analytics: by
applying this model, you can predict the occurrence of an important event in
the future, such as predicting the time in which customers make credit payments
in the future, in this case building a model for predictive analytics is so
important for you.
§ Prescriptive Analytics: this is used when you need to build
a model for making decisions in addition to modifying these decisions upon use,
this will be done according to many dynamic parameters which can suggest
prescribed actions. We can take Google's self-driving car as an example.
§ Making predictions using machine
learning: it's the
best way for determining the future trend of finance company by building a
model which use machine learning algorithms.
§ Pattern Discovery Using Machine
Learning: it's about
finding the hidden patterns for finding the parameters which enable you to make
meaningful predictions. Clustering is the most common algorithm used for this
purpose.
Data Science
Data science life cycle include six
main phases, they will be discussed as follows:
§ Discovery_ This's done before beginning the
project in order to understand the main budget of the project, its requirements
and specifications. You have to be sure that the required resources for the
project are present also the project problem is defined.
§ Data Preparation_ It's the step of establishing a relationship
between the variables you have, this will be done by exploring and conditioning
data before modeling. We can say that you prepare data for exploratory
analytics.
§ Model Planning_ In this step you will create a
relationship between all variables by using different techniques. You also use
visualization tools for applying exploratory data analytics. The tools which
are used in model planning are:
-
R: It's a complete set of modeling capabilities.
-
SQL Analysis Services.
-
SAS/ACCESS.
§ Model Building_ Purposes are trained and tested in
this phase by developing new datasets. In order to build your model, you have
to analyze various learning techniques.
§ Operationalize_ In this phase you are asked to
provide final reports and technical documents for providing a clear picture on
the project performance.
§ Communicate results_ It's
the phase in which you know if you achieved the project goal or not.
ليست هناك تعليقات:
إرسال تعليق