What is Data Science? | Introduction to Data Science | Data Science for Beginners
Data Science is an increasingly important field, with an ever-increasing demand for
data scientists. It is used for a variety of tasks, from predictive analysis like
predicting delays in airlines or predicting demand for certain products, to
creating promotional offers and choosing the most efficient routes for certain
journeys. Mohan Mohan discussed the need for data science and definitions, as well
as the differences between business intelligence and data science. He also
discussed the prerequisites for learning data science. Lastly, he mentioned how
data science can be used in politics to create personalized messages tailored to
the voters.
The first step in data science is asking the right questions and exploring the
data. This helps to identify the problem that needs to be solved and serves as the
basis for the modelling process. After modelling, results need to be visualized and
communicated to those who need to know them. Business intelligence relies heavily
on structured data, while data science involves much more complexity, such as
machine learning and the extrapolation of future trends like sales. Data science
goes beyond just presenting what has happened in the past and seeks to understand
why certain behavior has occurred.
Python is becoming increasingly popular in data science for its ease of use and the
variety of libraries it supports for data science, machine learning, and powerful
visualization through matplotlib. SAS is a well-established tool, and R provides
excellent visualization during development. Spark is an excellent computing engine
for distributed data analysis or machine learning. Additionally, there are standard
tools such as Informatica Data Stage, Talend, and AWS Redshift that can be used for
on-the-cloud operations. Raw data is collected, processed and analyzed before being
fed into the analytic system to create output which is then formatted in a way that
is useful for stakeholders.
Decision tree is primarily used for classification and can also be used for
regression. It is a clustering mechanism which determines which objects belong to
which cluster based on their scores. One advantage of decision tree is that it's
very easy to understand why a certain object has been classified in a certain way.
Data scientists explore the data, looking at its structure and removing any columns
that don't add value from an analytical perspective. Data must be cleaned and
prepared in order for the system to work properly, although the way of doing this
can vary from project to project. If there are too many missing values in few
records of large data sets, it's ok to get rid of those entire rows.
Data preparation is an essential step before analyzing or applying data. Model
planning follows, and which model to use depends on the problem you're trying to
solve. For example, if it is a regression problem, 80% of the training data can be
used to train a machine learning model. The training process may have to be
iterative, and MATLAB is a popular tool for educational purposes. As an example,
data scientists might build a model based on diamond carats in order to predict the
price of a 1.35 carat diamond. This would involve passing the information through a
linear regression model or creating an appropriate model for the task.
The demand for data scientists is currently huge and the supply is very low,
creating a large gap. Gaming and healthcare are two industries that are
particularly reliant on data science, as it is used for consumer-facing activities
such as diagnosis, predicting, and lifecycle management. The global demand for data
scientists is also high, which further highlights the importance of these skills.
To conclude this session, it is clear that the demand for data scientists will
remain high and their skills will be highly sought after.
Data Science is an increasingly important field, with an ever-increasing demand for
data scientists. It is used for a variety of tasks, from predictive analysis like
predicting delays in airlines or predicting demand for certain products, to
creating promotional offers and choosing the most efficient routes for certain
journeys. Mohan Mohan discussed the need for data science and definitions, as well
as the differences between business intelligence and data science. He also
discussed the prerequisites for learning data science. Lastly, he mentioned how
data science can be used in politics to create personalized messages tailored to
the voters.
The first step in data science is asking the right questions and exploring the
data. This helps to identify the problem that needs to be solved and serves as the
basis for the modelling process. After modelling, results need to be visualized and
communicated to those who need to know them. Business intelligence relies heavily
on structured data, while data science involves much more complexity, such as
machine learning and the extrapolation of future trends like sales. Data science
goes beyond just presenting what has happened in the past and seeks to understand
why certain behavior has occurred.
Python is becoming increasingly popular in data science for its ease of use and the
variety of libraries it supports for data science, machine learning, and powerful
visualization through matplotlib. SAS is a well-established tool, and R provides
excellent visualization during development. Spark is an excellent computing engine
for distributed data analysis or machine learning. Additionally, there are standard
tools such as Informatica Data Stage, Talend, and AWS Redshift that can be used for
on-the-cloud operations. Raw data is collected, processed and analyzed before being
fed into the analytic system to create output which is then formatted in a way that
is useful for stakeholders.
Decision tree is primarily used for classification and can also be used for
regression. It is a clustering mechanism which determines which objects belong to
which cluster based on their scores. One advantage of decision tree is that it's
very easy to understand why a certain object has been classified in a certain way.
Data scientists explore the data, looking at its structure and removing any columns
that don't add value from an analytical perspective. Data must be cleaned and
prepared in order for the system to work properly, although the way of doing this
can vary from project to project. If there are too many missing values in few
records of large data sets, it's ok to get rid of those entire rows.
Data preparation is an essential step before analyzing or applying data. Model
planning follows, and which model to use depends on the problem you're trying to
solve. For example, if it is a regression problem, 80% of the training data can be
used to train a machine learning model. The training process may have to be
iterative, and MATLAB is a popular tool for educational purposes. As an example,
data scientists might build a model based on diamond carats in order to predict the
price of a 1.35 carat diamond. This would involve passing the information through a
linear regression model or creating an appropriate model for the task.
The demand for data scientists is currently huge and the supply is very low,
creating a large gap. Gaming and healthcare are two industries that are
particularly reliant on data science, as it is used for consumer-facing activities
such as diagnosis, predicting, and lifecycle management. The global demand for data
scientists is also high, which further highlights the importance of these skills.
To conclude this session, it is clear that the demand for data scientists will
remain high and their skills will be highly sought after.
