Machine Learning CA1 Assignment - Dublin Business School Analysis

This document provides a comprehensive solution to a Machine Learning assignment, specifically CA1 from Dublin Business School. The assignment covers key concepts in artificial intelligence, including the definitions of AI, machine learning, and deep learning. It explores parametric and non-parametric models, supervised and unsupervised learning techniques, and identifies common error types in machine learning. The solution addresses the core questions, offering concise definitions and explanations. The assignment also involves a case study focusing on building a machine learning project plan. References to relevant academic literature are included to support the analysis. The solution is structured to provide a clear understanding of the subject matter, making it a valuable resource for students studying machine learning.

Running head: MACHINE LEARNING
MACHINE LEARNING
Name of the Student
Name of the University
Author Note:

1MACHINE LEARNING
Answer to Question 1:
Artificial Intelligence can be defined as the simulation of human intelligence in machine
which are programmed similar like human. Machine learning is all about use of artificial
intelligence that gives the concept of learning and enhancing overall experience (Alpaydin
2020). Deep learning is known to be a subset of machine learning in artificial intelligence which
comes up with the capability of learning from unstructured data.
Answer to Question 2:
Parametric Machine learning
This is a learning model which is all about summarization of data along with collection of
parameter needed for fixed size (Biamonte et al. 2017). Irrespective of about the amount of data
which is thrown in parametric model, it does not make changes in mind about how the parameter
is needed.
Non-Parametric Machine learning
Non-parametric method are found to be good when an individual has lot of data and no
earlier knowledge. An individual does not think much about choosing the right feature.
Answer to Question 3:
Supervised learning is defined as the machine learning job of learning a given function
where mapping is done in between input to output. It is just an example of the relation in
between input and output pair (Abadi et al. 2016). Supervised learning can be easily divided into
two categories of algorithm that are classification and regression. Classification issue is where

2MACHINE LEARNING
the output variable stand out to be category. Regression problem is where output variable comes
up with a real value.
Answer to Question 4:
Unsupervised learning deals with machine training by making use of information which
is neither labelled nor even classified. It merely allows algorithm to work like information
without any kind of idea (Papernot et al. 2017). Unsupervised learning can be classified into
categories that are clustering and association. In clustering problem, an individual wants to
discover and inherent group of data.
Answer to Question 5:
There are various kind of error which is associated with machine learning and predictive
analytics (Biamonte et al. 2017). These are mainly of two kinds that are in-sample and out of
sample errors. In case of sample errors, all the error are found in the training data. Out-of sample
error are stated as the error rates which are found in the new data set.

3MACHINE LEARNING
References
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving,
G., Isard, M. and Kudlur, M., 2016. Tensorflow: A system for large-scale machine learning.
In 12th {USENIX} Symposium on Operating Systems Design and Implementation ({OSDI}
16) (pp. 265-283).
Alpaydin, E., 2020. Introduction to machine learning. MIT press.
Biamonte, J., Wittek, P., Pancotti, N., Rebentrost, P., Wiebe, N. and Lloyd, S., 2017. Quantum
machine learning. Nature, 549(7671), pp.195-202.
Papernot, N., McDaniel, P., Goodfellow, I., Jha, S., Celik, Z.B. and Swami, A., 2017, April.
Practical black-box attacks against machine learning. In Proceedings of the 2017 ACM on Asia
conference on computer and communications security (pp. 506-519).