Machine Learning and the Bigdata
Paradigm
Ninan Sajeeth Philip
Abstract The use of machines to replicate human intelligence so that they may be
used to consistently predict scenarios without human intervention is the subject of
machine learning. The article describes two of the machine learning models that
were developed by the author during his Ph.D. under the supervision of Prof. K
Babu Joseph. The methods have high relevance in the present era that witnesses data
explosion in all branches of knowledge, nicknamed Bigdata.
Keywords Machine learning · Neural networks · Bigdata
1 Introduction
The human brain works with the help of chemical computers called Neurons. Each
neuron has a set of output called axons and inputs called dendrites. [See Fig. 1].
The output from the axons are input to the next set of dendrites and the strength
of the connection between them is controlled by the concentration of the chemicals
exchanged between them. This model of neurons lead to the design of computerbased neurons that also can accept input and pass on outputs to subsequent compute
nodes. A network of such neurons is called an Artificial Neural Network or ANN. The
experiential learning that takes place in the human brain is still not fully understood.
But the process of learning through examples in the initial childhood years is well
understood [1]. During this time, several chemical changes happen in the brain and
the neurons learn to fire in accordance with every stimulus such that the same amount
of chemicals are fired when a similar event occurs. The ANN mimics the chemical
transportation strength in the biological neuron by assigning a connection weight
between the output node and the subsequent input nodes. A set of examples are then
used to adjust these connection weights so that the output of the ANN matches the
N. S. Philip (B)
Artificial Intelligence Research and Intelligent Systems, Thelliyoor, Kerala, India
e-mail: ninansajeethphilip@airis4d.com
Department of Physics, St. Thomas College, Kozhencherry, Kerala, India
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
K. S. Sreelatha and V. Jacob (eds.), Modern Perspectives in Theoretical Physics,
https://doi.org/10.1007/978-981-15-9313-0_11
153
Paradigm
Ninan Sajeeth Philip
Abstract The use of machines to replicate human intelligence so that they may be
used to consistently predict scenarios without human intervention is the subject of
machine learning. The article describes two of the machine learning models that
were developed by the author during his Ph.D. under the supervision of Prof. K
Babu Joseph. The methods have high relevance in the present era that witnesses data
explosion in all branches of knowledge, nicknamed Bigdata.
Keywords Machine learning · Neural networks · Bigdata
1 Introduction
The human brain works with the help of chemical computers called Neurons. Each
neuron has a set of output called axons and inputs called dendrites. [See Fig. 1].
The output from the axons are input to the next set of dendrites and the strength
of the connection between them is controlled by the concentration of the chemicals
exchanged between them. This model of neurons lead to the design of computerbased neurons that also can accept input and pass on outputs to subsequent compute
nodes. A network of such neurons is called an Artificial Neural Network or ANN. The
experiential learning that takes place in the human brain is still not fully understood.
But the process of learning through examples in the initial childhood years is well
understood [1]. During this time, several chemical changes happen in the brain and
the neurons learn to fire in accordance with every stimulus such that the same amount
of chemicals are fired when a similar event occurs. The ANN mimics the chemical
transportation strength in the biological neuron by assigning a connection weight
between the output node and the subsequent input nodes. A set of examples are then
used to adjust these connection weights so that the output of the ANN matches the
N. S. Philip (B)
Artificial Intelligence Research and Intelligent Systems, Thelliyoor, Kerala, India
e-mail: ninansajeethphilip@airis4d.com
Department of Physics, St. Thomas College, Kozhencherry, Kerala, India
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
K. S. Sreelatha and V. Jacob (eds.), Modern Perspectives in Theoretical Physics,
https://doi.org/10.1007/978-981-15-9313-0_11
153