Beginners in Data Analytics with Machine Learning ..? You have know these things..?!

A wonderful term Machine Learning is a very wonderful concept in real. Experience of one thing keeps our task easy, effective and fast. Same thing apply in machines that is called Machine Learning. Machine Learning work in a similar way to human learning.

What is Machine Learning?

“Machine Learning is the study of computer algorithms that improve automatically through experience.”

Technical definition of ML is -“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.”

“The only source of knowledge is Experience.” -Albert Einstein

Machine learning is the subset of Artificial Intelligence(AI).

Deep Learning is the subset of Machine Learning.

What is Artificial Intelligence (AI)?

Artificial intelligence (AI) is the ability of a computer program or a machine to think and learn. It is also a field of study which tries to make computers “smart”. They work on their own without being encoded with commands.

AI is a very big term and it is used in most of the area of computer science. Above figure shows that Machine Learning is an important part of AI.

Machine Learning in Data Analytics :

Machine learning is a method of data analysis that automates analytical model building. Analytics involves studying historical data to research potential trends , to analyze the effects of certain decision or event or to evaluate the performance of a given tools or scenario. The goal of analytics improves the processes by gaining knowledge which can be used to make improvement or changes.

There are mainly three types of analytics :

1. Descriptive Analytics :

It is the primary form of analytics that aggregates big data and provides useful insights into the past.

2. Predictive Analytics :

Next step in data reduction;

It uses various statistical modeling and machine learning techniques to analyze past data and predict the future outcomes.

3. Prescriptive Analytics :

New form of analytics that uses a combination of business rules, machine learning, and computational modeling to recommend the best course of action for any pre-specified outcome.

How ML use in Predictive Analytics ?

ML is the core of Predictive Analytics (PA).

Predictive analytics means technology that learns from experience (data) to predict the future behaviour of individuals in order to drive better decisions.

Have you accessed websites such as Amazon, Netflix, YouTube etc. for purchasing different products, books, watching movie and videos ? Every time you visit these sites, the recommendations of products, books, movies or television shows based on your linking, show up on your screen. How does this happen?

These websites use ML to churn your past purchase and browsing data in order to predict what you like and thus recommend the items as per these prediction.

Use cases of Machine Learning based on Predictive Analysis :

1. Sales and Marketing
2. E-commerce
3. Finance
4. Customer Service
5. Medical Diagnosis
6. Cyber Security

Let’s see an example,

Advantages to Sales and Marketing sector :

In this sector Massive data consumption from unlimited sources

ML virtually consumes unlimited amount of comprehensive data. These data used to constantly review and modify sales and marketing strategies based on the customer behavioural patterns.

Once the model is trained, it identifies highly relevant variables and possible to get focused data feeds by foregoing long and complicated integrations.

What is the path of Evolution of Machine Learning?

Evolution of ML up to year 2010

2010 — The Microsoft Kinect can track 20 human features at a rate of 30 times per second, allowing people to interact with the computer via movements and gestures.

2011 — IBM’s Watson beats its human competitors at Jeopardy

• Google Brain is developed, and its deep neural network can learn to discover and categorize objects much the way a cat does.

2012 — Google’s X Lab develops a machine learning algorithm that is able to autonomously browse YouTube videos to identify the videos that contain cats.

2014 — Facebook develops Deep Face, a software algorithm that is able to recognize or verify individuals on photos to the same level as humans can.

2015 — Amazon launches its own machine learning platform.

• Microsoft creates the Distributed Machine Learning Toolkit, which enables the efficient distribution of machine learning problems across multiple computers.
• Over 3,000 AI and Robotics researchers, endorsed by Stephen Hawking, Elon Musk and Steve Wozniak (among many others), sign an open letter warning of the danger of autonomous weapons which select and engage targets without human intervention.

2016 — Google’s artificial intelligence algorithm beats a professional player at the Chinese board game Go, which is considered the world’s most complex board game and is many times harder than chess. The AlphaGo algorithm developed by Google DeepMind managed to win five games out of five in the Go competition.

Now Machine Learning and Deep learning applications are vast.

Traditional v/s New ML platforms

Now, let’s dive into types of ML Algorithm!

There are mainly three approaches of Machine Learning Algorithms :

1. Supervised Approach
2. Unsupervised Approach
3. Reinforcement Approach

1. Supervised Learning Approach :

Supervised approach is similar to human learning under the supervision of teacher. The teacher provides examples to the students and the student then derive general rules from these specific examples.

Supervised Learning Approach of ML

• Supervised leaning is task driven.
• It occurs when an algorithm learns from example data and associated target responses, in order to predict the correct response when posed new examples.
• The target response may consist of numeric values or labels such as classes and tags.
• The main types of Supervised learning problems :

1) Classification

2) Regression

• List of Common Algorithms to solve classification problems:

Techniques to solve classification problem

• List of Common Algorithms to solve regression problems:

(1) Linear Regression

(2) Ordinary Least Square Regression

(3) Logistic Regression

• Ensemble Methods :

Ensemble methods are the meta-algorithms that combine several machine learning algorithms and techniques into one predictive model in order to decrease the bagging(variance), boosting(bias) or improve the predictions (stacking).

2. Unsupervised Learning Approach :

Unsupervised approach occurs when an algorithm learns from examples provided without any associated response. The algorithm determines the data patterns on it own.

Unsupervised Learning Approach of ML

• This type of algorithm tends to restructure the data into something else, such as new features that may represent a class or a new series of uncorrelated values.
• They are quite useful in providing humans with insights into the meaning of data and also new useful inputs to supervised machine learning algorithms.
• So, in this type the training data does not include targets , we do not need to tell the system where to go and the system has to understand itself from the data provided.
• The main types of Unsupervised learning problems :

1) Association

2) Anomaly Detection

3) Clustering

• List of Common Algorithms to solve unsupervised learning problems:

(1) K-means clustering

(2) Apriori algorithm

(3) Principle component analysis(PCA)

(4) Singular Value Decomposition

(5) Independent Component Analysis

3. Reinforcement Learning Approach :

In Reinforcement Learning Approach, the algorithm is presented with examples without Labels, as Unsupervised Learning Approach. The algorithm determines the solution on its own. On the basis of solutions, a positive or negative feedback is provided. This helps to algorithm to learn the correct responses.

Reinforcement Learning Approach of ML

• It is just like learning by trial and error method in the living world. Errors helps to learn because they have a penalty associated with them in terms of cost, loss of time, regret, pain, etc.
• It is used in applications wherein the algorithm is required to take decisions which bear consequences.

Which Model/Algorithm to choose ?

Mainly there are three aspects of selecting an algorithm :

1. Accuracy
2. Training time
3. Ease of use

• Many users give priority to accuracy, while beginners have a tendency to go with algorithms they know best.
• But, true method is when we are presented with dataset, the first thing to focus on is how to obtain results, no matter what those results might look like.
• Beginners tend to choose algorithms that are easy to implement so that the results can be obtained quickly.
• Once you obtain some results and become familiar with the data, you may use more sophisticated algorithm to strengthen your understanding of data and further improve the results.

So, Machine Learning part very interesting and effective path in data analytics!!

Is it interesting ?

For the further knowledge about types of machine learning follow this link : ML Algorithms in detail