Prediction of Mechanical Properties in Rotational Moulding …
5
Supervised Learning. In this type, the model learns from the given data set known
as training set, containing both input and output parameters. It contains mainly two
types, known as classification and regression. Classification is used to identify an
observation to which category it belongs, on the basis of the provided training set
to the model. Whereas regression is used when the output is continuous, i.e. there
exists an output variable for an input parameter, and to predict an output for a given
new input parameter from the model which has the training set. It can be a linear or
a polynomial regression on the basis of the variation of outputs with the inputs. In
linear regression model, hypothesis is done from one-degree polynomial, whereas
in polynomial regression model, hypothesis is done from a polynomial having more
than one degree depending upon the inputs and their corresponding outputs.
Unsupervised Learning. In this type, the model learns from the data sets that
contains only inputs, and from this training data set, the model analyses the structures
in the data set, like grouping and categorizing.
Semi-supervised Learning. This type is the combination of supervised and unsupervised learning in which model learns from the training set which has both labelled
(containing both input and output parameters) and unlabelled (containing only inputs)
data sets.
Reinforcement Learning. It has three major components: the agent, environment
and action. In this type, agent has the ability to interact with the environment and
analyses what would be the best outcome with the hit and trial method.
1.3 Application of Machine Learning in Different Fields
It has been evident that in the recent years, machine learning has shown greater
impact on our lives and onto various fields. Some of the major fields and works done
by researchers are as follows:
Scikit-learn: Machine Learning in Python. Scikit-learn is a Python module that as
the best machine learning algorithms for medium level of supervise and unsupervised
problems. This module mainly concentrates on the usage of machine learning by
non-specialists in this field [19].
Machine Learning in Medical Sector. It is important to note that large medical data
sets are present for us including patient reports, diagnostics, disease characteristics,
etc., for the training data sets for the machine learning models. And with the help
of these models, various problems have been solved. Like in the case of cancer, it
has predicted the survival outcome and also by the attracter metagene algorithm it
had found out a cluster of genes in the tumour cells which was similar in the cancer
patients [20].
Machine Learning in Financial Sector. Machine learning has also contributed to
the financial sector. It was found that machine learning model can learn to price the
5
Supervised Learning. In this type, the model learns from the given data set known
as training set, containing both input and output parameters. It contains mainly two
types, known as classification and regression. Classification is used to identify an
observation to which category it belongs, on the basis of the provided training set
to the model. Whereas regression is used when the output is continuous, i.e. there
exists an output variable for an input parameter, and to predict an output for a given
new input parameter from the model which has the training set. It can be a linear or
a polynomial regression on the basis of the variation of outputs with the inputs. In
linear regression model, hypothesis is done from one-degree polynomial, whereas
in polynomial regression model, hypothesis is done from a polynomial having more
than one degree depending upon the inputs and their corresponding outputs.
Unsupervised Learning. In this type, the model learns from the data sets that
contains only inputs, and from this training data set, the model analyses the structures
in the data set, like grouping and categorizing.
Semi-supervised Learning. This type is the combination of supervised and unsupervised learning in which model learns from the training set which has both labelled
(containing both input and output parameters) and unlabelled (containing only inputs)
data sets.
Reinforcement Learning. It has three major components: the agent, environment
and action. In this type, agent has the ability to interact with the environment and
analyses what would be the best outcome with the hit and trial method.
1.3 Application of Machine Learning in Different Fields
It has been evident that in the recent years, machine learning has shown greater
impact on our lives and onto various fields. Some of the major fields and works done
by researchers are as follows:
Scikit-learn: Machine Learning in Python. Scikit-learn is a Python module that as
the best machine learning algorithms for medium level of supervise and unsupervised
problems. This module mainly concentrates on the usage of machine learning by
non-specialists in this field [19].
Machine Learning in Medical Sector. It is important to note that large medical data
sets are present for us including patient reports, diagnostics, disease characteristics,
etc., for the training data sets for the machine learning models. And with the help
of these models, various problems have been solved. Like in the case of cancer, it
has predicted the survival outcome and also by the attracter metagene algorithm it
had found out a cluster of genes in the tumour cells which was similar in the cancer
patients [20].
Machine Learning in Financial Sector. Machine learning has also contributed to
the financial sector. It was found that machine learning model can learn to price the