Machine Learning (ML) is basically that field of computer science with the help of which computer systems can provide sense to data in much the same way as human beings do. In simple words, ML is a type of artificial intelligence that extract patterns out of raw data by using an algorithm or method. The key focus of ML is to allow computer systems to learn from experience without being explicitly programmed or human intervention.

Audience

This tutorial will be useful for graduates, postgraduates, and research students who either have an interest in this subject or have this subject as a part of their curriculum. The reader can be a beginner or an advanced learner. This tutorial has been prepared for the students as well as professionals to ramp up quickly. This tutorial is a stepping stone to your Machine Learning journey.

Prerequisites

The reader must have basic knowledge of artificial intelligence. He/she should also be aware of Python, NumPy, Scikit-learn, Scipy, Matplotlib. If you are new to any of these concepts, we recommend you to take up tutorials concerning these topics, before you dig further into this tutorial.

e are living in the ‘age of data’ that is enriched with better computational power and more storage resources,. This data or information is increasing day by day, but the real challenge is to make sense of all the data. Businesses & organizations are trying to deal with it by building intelligent systems using the concepts and methodologies from Data science, Data Mining and Machine learning. Among them, machine learning is the most exciting field of computer science. It would not be wrong if we call machine learning the application and science of algorithms that provides sense to the data.

What is Machine Learning?

Machine Learning (ML) is that field of computer science with the help of which computer systems can provide sense to data in much the same way as human beings do.

In simple words, ML is a type of artificial intelligence that extract patterns out of raw data by using an algorithm or method. The main focus of ML is to allow computer systems learn from experience without being explicitly programmed or human intervention.

Need for Machine Learning

Human beings, at this moment, are the most intelligent and advanced species on earth because they can think, evaluate and solve complex problems. On the other side, AI is still in its initial stage and haven’t surpassed human intelligence in many aspects. Then the question is that what is the need to make machine learn? The most suitable reason for doing this is, “to make decisions, based on data, with efficiency and scale”.

Lately, organizations are investing heavily in newer technologies like Artificial Intelligence, Machine Learning and Deep Learning to get the key information from data to perform several real-world tasks and solve problems. We can call it data-driven decisions taken by machines, particularly to automate the process. These data-driven decisions can be used, instead of using programing logic, in the problems that cannot be programmed inherently. The fact is that we can’t do without human intelligence, but other aspect is that we all need to solve real-world problems with efficiency at a huge scale. That is why the need for machine learning arises.

Why & When to Make Machines Learn?

We have already discussed the need for machine learning, but another question arises that in what scenarios we must make the machine learn? There can be several circumstances where we need machines to take data-driven decisions with efficiency and at a huge scale. The followings are some of such circumstances where making machines learn would be more effective −

Lack of human expertise

The very first scenario in which we want a machine to learn and take data-driven decisions, can be the domain where there is a lack of human expertise. The examples can be navigations in unknown territories or spatial planets.

Dynamic scenarios

There are some scenarios which are dynamic in nature i.e. they keep changing over time. In case of these scenarios and behaviors, we want a machine to learn and take data-driven decisions. Some of the examples can be network connectivity and availability of infrastructure in an organization.

Difficulty in translating expertise into computational tasks

There can be various domains in which humans have their expertise,; however, they are unable to translate this expertise into computational tasks. In such circumstances we want machine learning. The examples can be the domains of speech recognition, cognitive tasks etc.

Machine Learning Model

Before discussing the machine learning model, we must need to understand the following formal definition of ML given by professor Mitchell −

“A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.”

The above definition is basically focusing on three parameters, also the main components of any learning algorithm, namely Task(T), Performance(P) and experience (E). In this context, we can simplify this definition as −

ML is a field of AI consisting of learning algorithms that −

• Improve their performance (P)

• At executing some task (T)

• Over time with experience (E)

Based on the above, the following diagram represents a Machine Learning Model −

Let us discuss them more in detail now −

Task(T)

From the perspective of problem, we may define the task T as the real-world problem to be solved. The problem can be anything like finding best house price in a specific location or to find best marketing strategy etc. On the other hand, if we talk about machine learning, the definition of task is different because it is difficult to solve ML based tasks by conventional programming approach.

A task T is said to be a ML based task when it is based on the process and the system must follow for operating on data points. The examples of ML based tasks are Classification, Regression, Structured annotation, Clustering, Transcription etc.

Experience (E)

As name suggests, it is the knowledge gained from data points provided to the algorithm or model. Once provided with the dataset, the model will run iteratively and will learn some inherent pattern. The learning thus acquired is called experience(E). Making an analogy with human learning, we can think of this situation as in which a human being is learning or gaining some experience from various attributes like situation, relationships etc. Supervised, unsupervised and reinforcement learning are some ways to learn or gain experience. The experience gained by out ML model or algorithm will be used to solve the task T.

Performance (P)

An ML algorithm is supposed to perform task and gain experience with the passage of time. The measure which tells whether ML algorithm is performing as per expectation or not is its performance (P). P is basically a quantitative metric that tells how a model is performing the task, T, using its experience, E. There are many metrics that help to understand the ML performance, such as accuracy score, F1 score, confusion matrix, precision, recall, sensitivity etc.

Challenges in Machines Learning

While Machine Learning is rapidly evolving, making significant strides with cybersecurity and autonomous cars, this segment of AI as whole still has a long way to go. The reason behind is that ML has not been able to overcome number of challenges. The challenges that ML is facing currently are −

Quality of data − Having good-quality data for ML algorithms is one of the biggest challenges. Use of low-quality data leads to the problems related to data preprocessing and feature extraction.

Time-Consuming task − Another challenge faced by ML models is the consumption of time especially for data acquisition, feature extraction and retrieval.

Lack of specialist persons − As ML technology is still in its infancy stage, availability of expert resources is a tough job.

No clear objective for formulating business problems − Having no clear objective and well-defined goal for business problems is another key challenge for ML because this technology is not that mature yet.

Issue of overfitting & underfitting − If the model is overfitting or underfitting, it cannot be represented well for the problem.

Curse of dimensionality − Another challenge ML model faces is too many features of data points. This can be a real hindrance.

Difficulty in deployment − Complexity of the ML model makes it quite difficult to be deployed in real life.

Applications of Machines Learning

Machine Learning is the most rapidly growing technology and according to researchers we are in the golden year of AI and ML. It is used to solve many real-world complex problems which cannot be solved with traditional approach. Following are some real-world applications of ML −

• Emotion analysis

• Sentiment analysis

• Error detection and prevention

• Weather forecasting and prediction

• Stock market analysis and forecasting

• Speech synthesis

• Speech recognition

• Customer segmentation

• Object recognition

• Fraud detection

• Fraud prevention

• Recommendation of products to customer in online shopping.

An Introduction to Python

Python is a popular object-oriented programing language having the capabilities of high-level programming language. Its easy to learn syntax and portability capability makes it popular these days. The followings facts gives us the introduction to Python −

• Python was developed by Guido van Rossum at Stichting Mathematisch Centrum in the Netherlands.

• It was written as the successor of programming language named ‘ABC’.

• It’s first version was released in 1991.

• The name Python was picked by Guido van Rossum from a TV show named Monty Python’s Flying Circus.

• It is an open source programming language which means that we can freely download it and use it to develop programs. It can be downloaded from www.python.org.

• Python programming language is having the features of Java and C both. It is having the elegant ‘C’ code and on the other hand, it is having classes and objects like Java for object-oriented programming.

• It is an interpreted language, which means the source code of Python program would be first converted into bytecode and then executed by Python virtual machine.

Strengths and Weaknesses of Python

Every programming language has some strengths as well as weaknesses, so does Python too.

Strengths

According to studies and surveys, Python is the fifth most important language as well as the most popular language for machine learning and data science. It is because of the following strengths that Python has −

Easy to learn and understand − The syntax of Python is simpler; hence it is relatively easy, even for beginners also, to learn and understand the language.

Multi-purpose language − Python is a multi-purpose programming language because it supports structured programming, object-oriented programming as well as functional programming.

Huge number of modules − Python has huge number of modules for covering every aspect of programming. These modules are easily available for use hence making Python an extensible language.

Support of open source community − As being open source programming language, Python is supported by a very large developer community. Due to this, the bugs are easily fixed by the Python community. This characteristic makes Python very robust and adaptive.

Scalability − Python is a scalable programming language because it provides an improved structure for supporting large programs than shell-scripts.

Weakness

Although Python is a popular and powerful programming language, it has its own weakness of slow execution speed.

The execution speed of Python is slow as compared to compiled languages because Python is an interpreted language. This can be the major area of improvement for Python community.

Installing Python

For working in Python, we must first have to install it. You can perform the installation of Python in any of the following two ways −

• Installing Python individually

• Using Pre-packaged Python distribution − Anaconda

Let us discuss these each in detail.

Installing Python Individually

If you want to install Python on your computer, then then you need to download only the binary code applicable for your platform. Python distribution is available for Windows, Linux and Mac platforms.

The following is a quick overview of installing Python on the above-mentioned platforms −

On Unix and Linux platform

With the help of following steps, we can install Python on Unix and Linux platform −

• First, go to www.python.org/downloads/.

• Next, click on the link to download zipped source code available for Unix/Linux.

• Now, Download and extract files.

• Next, we can edit the Modules/Setup file if we want to customize some options.

  • Next, write the command run ./configure script

  • make

  • make install

On Windows platform

With the help of following steps, we can install Python on Windows platform −

• First, go to www.python.org/downloads/.

• Next, click on the link for Windows installer python-XYZ.msi file. Here XYZ is the version we wish to install.

• Now, we must run the file that is downloaded. It will take us to the Python install wizard, which is easy to use. Now, accept the default settings and wait until the install is finished.

On Macintosh platform

For Mac OS X, Homebrew, a great and easy to use package installer is recommended to install Python 3. In case if you don't have Homebrew, you can install it with the help of following command −

$ ruby -e "$(curl -fsSLhttps://raw.githubusercontent.com/Homebrew/install/master/install)"

It can be updated with the command below −

$ brew update

Now, to install Python3 on your system, we need to run the following command −

$ brew install python3

Using Pre-packaged Python Distribution: Anaconda

Anaconda is a packaged compilation of Python which have all the libraries widely used in Data science. We can follow the following steps to setup Python environment using Anaconda −

• Step 1 − First, we need to download the required installation package from Anaconda distribution. The link for the same is www.anaconda.com/distribution/. You can choose from Windows, Mac and Linux OS as per your requirement.

• Step 2 − Next, select the Python version you want to install on your machine. The latest Python version is 3.7. There you will get the options for 64-bit and 32-bit Graphical installer both.

• Step 3 − After selecting the OS and Python version, it will download the Anaconda installer on your computer. Now, double click the file and the installer will install Anaconda package.

• Step 4 − For checking whether it is installed or not, open a command prompt and type Python as follows −

You can also check this in detailed video lecture at www.tutorialspoint.com/python_essentials_online_training/getting_started_with_anaconda.asp.

Why Python for Data Science?

Python is the fifth most important language as well as most popular language for Machine learning and data science. The following are the features of Python that makes it the preferred choice of language for data science −

Extensive set of packages

Python has an extensive and powerful set of packages which are ready to be used in various domains. It also has packages like numpy, scipy, pandas, scikit-learn etc. which are required for machine learning and data science.

Easy prototyping

Another important feature of Python that makes it the choice of language for data science is the easy and fast prototyping. This feature is useful for developing new algorithm.

Collaboration feature

The field of data science basically needs good collaboration and Python provides many useful tools that make this extremely.

One language for many domains

A typical data science project includes various domains like data extraction, data manipulation, data analysis, feature extraction, modelling, evaluation, deployment and updating the solution. As Python is a multi-purpose language, it allows the data scientist to address all these domains from a common platform.

Components of Python ML Ecosystem

In this section, let us discuss some core Data Science libraries that form the components of Python Machine learning ecosystem. These useful components make Python an important language for Data Science. Though there are many such components, let us discuss some of the importance components of Python ecosystem here −

Jupyter Notebook

Jupyter notebooks basically provides an interactive computational environment for developing Python based Data Science applications. They are formerly known as ipython notebooks. The following are some of the features of Jupyter notebooks that makes it one of the best components of Python ML ecosystem −

• Jupyter notebooks can illustrate the analysis process step by step by arranging the stuff like code, images, text, output etc. in a step by step manner.

• It helps a data scientist to document the thought process while developing the analysis process.

• One can also capture the result as the part of the notebook.

• With the help of jupyter notebooks, we can share our work with a peer also.

Installation and Execution

If you are using Anaconda distribution, then you need not install jupyter notebook separately as it is already installed with it. You just need to go to Anaconda Prompt and type the following command −

C:\>jupyter notebook

After pressing enter, it will start a notebook server at localhost:8888 of your computer. It is shown in the following screen shot −

Now, after clicking the New tab, you will get a list of options. Select Python 3 and it will take you to the new notebook for start working in it. You will get a glimpse of it in the following screenshots −

On the other hand, if you are using standard Python distribution then jupyter notebook can be installed using popular python package installer, pip.

pip install jupyter

Types of Cells in Jupyter Notebook

The following are the three types of cells in a jupyter notebook −

Code cells − As the name suggests, we can use these cells to write code. After writing the code/content, it will send it to the kernel that is associated with the notebook.

Markdown cells − We can use these cells for notating the computation process. They can contain the stuff like text, images, Latex equations, HTML tags etc.

Raw cells − The text written in them is displayed as it is. These cells are basically used to add the text that we do not wish to be converted by the automatic conversion mechanism of jupyter notebook.

For more detailed study of jupyter notebook, you can go to the link www.tutorialspoint.com/jupyter/index.htm.

NumPy

It is another useful component that makes Python as one of the favorite languages for Data Science. It basically stands for Numerical Python and consists of multidimensional array objects. By using NumPy, we can perform the following important operations −

• Mathematical and logical operations on arrays.

• Fourier transformation

• Operations associated with linear algebra.

We can also see NumPy as the replacement of MatLab because NumPy is mostly used along with Scipy (Scientific Python) and Mat-plotlib (plotting library).

Installation and Execution

If you are using Anaconda distribution, then no need to install NumPy separately as it is already installed with it. You just need to import the package into your Python script with the help of following −

import numpy as np

On the other hand, if you are using standard Python distribution then NumPy can be installed using popular python package installer, pip.

pip install NumPy

For more detailed study of NumPy, you can go to the link www.tutorialspoint.com/numpy/index.htm.

Pandas

It is another useful Python library that makes Python one of the favorite languages for Data Science. Pandas is basically used for data manipulation, wrangling and analysis. It was developed by Wes McKinney in 2008. With the help of Pandas, in data processing we can accomplish the following five steps −

• Load
• Prepare
• Manipulate
• Model
• Analyze

Data representation in Pandas

The entire representation of data in Pandas is done with the help of following three data structures −

Series − It is basically a one-dimensional ndarray with an axis label which means it is like a simple array with homogeneous data. For example, the following series is a collection of integers 1,5,10,15,24,25...

151015242528364089

Data frame − It is the most useful data structure and used for almost all kind of data representation and manipulation in pandas. It is basically a two-dimensional data structure which can contain heterogeneous data. Generally, tabular data is represented by using data frames. For example, the following table shows the data of students having their names and roll numbers, age and gender −

NameRoll numberAgeGenderAarav115MaleHarshit214MaleKanika316FemaleMayank415Male

Panel − It is a 3-dimensional data structure containing heterogeneous data. It is very difficult to represent the panel in graphical representation, but it can be illustrated as a container of DataFrame.

The following table gives us the dimension and description about above mentioned data structures used in Pandas −

Data StructureDimensionDescriptionSeries1-DSize immutable, 1-D homogeneous dataDataFrames2-DSize Mutable, Heterogeneous data in tabular formPanel3-DSize-mutable array, container of DataFrame.

We can understand these data structures as the higher dimensional data structure is the container of lower dimensional data structure.

Installation and Execution

If you are using Anaconda distribution, then no need to install Pandas separately as it is already installed with it. You just need to import the package into your Python script with the help of following −

import pandas as pd

On the other hand, if you are using standard Python distribution then Pandas can be installed using popular python package installer, pip.

pip install Pandas

After installing Pandas, you can import it into your Python script as did above.

Example

The following is an example of creating a series from ndarray by using Pandas −

In [1]: import pandas as pdIn [2]: import numpy as npIn [3]: data = np.array(['g','a','u','r','a','v'])In [4]: s = pd.Series(data)In [5]: print (s)0 g1 a2 u3 r4 a5 vdtype: object

For more detailed study of Pandas you can go to the link www.tutorialspoint.com/python_pandas/index.htm.

Scikit-learn

Another useful and most important python library for Data Science and machine learning in Python is Scikit-learn. The following are some features of Scikit-learn that makes it so useful −

• It is built on NumPy, SciPy, and Matplotlib.

• It is an open source and can be reused under BSD license.

• It is accessible to everybody and can be reused in various contexts.

• Wide range of machine learning algorithms covering major areas of ML like classification, clustering, regression, dimensionality reduction, model selection etc. can be implemented with the help of it.

Installation and Execution

If you are using Anaconda distribution, then no need to install Scikit-learn separately as it is already installed with it. You just need to use the package into your Python script. For example, with following line of script we are importing dataset of breast cancer patients from Scikit-learn −

from sklearn.datasets import load_breast_cancer

On the other hand, if you are using standard Python distribution and having NumPy and SciPy then Scikit-learn can be installed using popular python package installer, pip.

pip install -U scikit-learn

After installing Scikit-learn, you can use it into your Python script as you have done above.