In talk with Mayank!

In talk with Mayank

Exploring the coding journey!

Learning from a great teacher can never be erased!

I began my coding journey in the first semester by taking foundation course (C++ with Data Structures which was taught by Ankush Singla.

His amazing experience and what he loved!

The quality of course content and problems on the super awesome “CodeZen” helped to ace the coding rounds in various interviews.Teaching Assistant support was amazing as they were available most of the time and we used to get our doubts resolved within few hours. In addition the faculty was very supportive and I could come at the center anytime for doubts or any other help.


I further took the Web Development course which is based on Ruby on Rails because of which I grabbed an internship in Bizongo. The interview rounds covered a lot of questions on Ruby on Rails, Data Structures , Arrays , Link list concepts which I could clear only because of my knowledge I got through my mentors in Coding Ninjas.

Advice to Current Students!

I got a lot of question on Arrays, Link list, Trees. Advice all the students to practice more and more concepts plus code problems on the website itself rather than any other software.

How will you introduce Coding Ninjas to your friends?

An awesome play to start for your coding journey to become a NINJA coder

Thank you for constant support and guidance !

How Open Source Changed The Face Of Businesses And Programming


Everyone in this world is using open source platform in one way or the other, irrespective of whether they realize it or not. Ever since its inception, open source has been a huge hit among software developers as well as businesses. February 2018 marked the 20th Anniversary of the Open Source Initiative (OSI), a movement that was started way back in 1998 by Eric Raymond and Bruce Perens. The aim was to make the source code accessible to all so that developers around the world can tweak and make useful additions to a software to enrich it even further.

Essentially, the open source community is comprised of both professional and amateur developers who enhance, modify, and work on publicly available source code. And businesses can utilize the code through tools such as Github, to build their own products and services. Today, more and more businesses around the world are joining the Open source bandwagon as it helps them save both time and money.

Over the years, the open source platform has made a considerable impact on programming and businesses. In an annual survey conducted Black Duck, it was found that of the 1,300 respondents, nearly 78% agreed to run their business operations (at least partly) on open source software. Here’s how open source is helping enrich programming and businesses:

  1. Viable And Accessible

During the initial years of the OSI, Linux largely remained beyond the understanding of man. It was not very user-friendly, as in, it was pretty difficult to install and work on. Today, however, the situation has changed a lot. After decades of modifications and enhancements, Linux is now much more convenient and easy to use. In fact, the popularity of Linux has escalated so high that the top 500 supercomputers of the world run on Linux.

  1. The ‘Source’ Behind Android’s Power!

Yes, open source is what makes Android so easy to work with. It is based on Linux. Many companies in the market today are leveraging open source software such as Linux or Tizen to build their Android platform. However, while most Android phones are based on open source OS, the end product is a closed source one, that is, it cannot be tweaked or modified.

  1. Run The Cloud!

The popularity of open source has caught on so fast that more and more people are now inclining towards services/products that can be accessed through the Web instead of using local apps for the same. Basically, the websites are loaded on someone’s computer somewhere in the world and open source platform allows users to access such websites. FreeBSD and Linux are a hit for cloud servers that allow others to access web services remotely.

  1. The Web – Made Easy

Today, anyone can have their own website, all thanks to open source. Take WordPress, for instance. WordPress is the backend for many websites – while obtaining a domain name is going to cost you, building a website is essentially free. It is this open source feature that has made WordPress a renowned and trusted name all over the world.

  1. The Baton Continues To Progress Beyond The ‘Code’

Open source has gained such as strong foothold that companies are no longer just using it but are also helping progress it by launching their own open source codes. Reputed names such as Walmart and ExxonMobil are following suit. While Walmart launched an open source cloud management system, ExxonMobil released an open source developer toolkit. Furthermore, magnates of the finance world such as JP Morgan and London Stock Exchange Group have adopted Hyperledger, an open source software that holds the potential to completely transform the share market scenario.

Open source is massive. Practically every software that is a big hit with us today is rooted in open source. Jim Whitehurst, CEO of Red Hat, maintains:

“The future of open source is bright. We are on the cusp of a new wave of innovation that will come about because information is being separated from physical objects thanks to the Internet of Things. Over the next decade, we will see entire industries based on open-source concepts, like the sharing of information and joint innovation, become mainstream.”

So, folks, this is only the beginning for there’s a lot more to come!

Pros and Cons of open source programming languages


We’ve all heard about open source software and open source programming languages at some point in our lives. However, do you know what the term actually means?

“Open Source” basically refers to a platform or product that can be accessed by anyone and everyone. Open-source allows you to access, analyze, and tweak the source code for the purpose of enhancing it. Similarly, an open-source programming language is one that falls within the open-source protocol, that is, it can be accessed and modified by the public. Today, programmers have an array of options when it comes to open-source programming languages. For a programming language to be considered as open-source, it has to fulfill the following criteria:

  • The source code must be accessible by all.
  • The integrity of the source code must be maintained.
  • It should allow free distribution and redistribution.
  • It should allow free distribution of license.
  • Its license should be technology-neutral and should not restrict other software.
  • It should not discriminate between persons/groups and also against any field of endeavor.

Now that we know what an open-source programming language is, let us look at some of the most widely used open-source languages as well as the advantages and disadvantages each brings.

  1. Java

Java is undoubtedly one of the most popular open-source programming languages among programmers. It is a highly class-based and featured object-oriented language. Today, Java forms the core of various web and mobile applications across a wide range of operating systems and devices.


  • Easy to write, compile, and debug.
  • Excellent for developing modular programs and reusable code.
  • It is platform independent, both at the binary and source levels.


  • It is not speed-intensive and hence, can be very time-consuming.
  • It is an interpreted language.
  • Lacks support for low-level programming languages.
  1. Python

Python is a versatile programming language that anyone can master with a little dedication and effort. It was developed by  Guido van Rossum back in the 1980s and is controlled and managed by the Python Software Foundation. It is used by developers around the world as a scripting language allowing them to generate readable and functional code in a short span of time.


  • It comes with readable and simple syntax.
  • It can be interpreted in many operating systems owing to its portability, such as Unix, Windows, and Mac OS, to name a few.
  • It supports both object-oriented and procedure-oriented programming.


  • Lacks pre-packaged solutions.
  • Lacks adequate GUI tools.
  • Not ideal for memory intensive functions.
  1. Ruby

Developed by  Yukihiro Matsumoto in the 90’s, Ruby is an object-oriented programming language that has gained huge traction over the past few years. The best thing about Ruby is its neat, clear, and simple syntax that makes it the ideal programming language for beginners. Today, Ruby along with the Rails framework is being used by reputed names such as Airbnb, GitHub, Basecamp, and Bloomberg.


  • It facilitates easy inheritance of features from standard classes.
  • Offers a clean object hierarchy thereby making metaprogramming very convenient.
  • Comes with a provision of open classes.
  • It allows you to write codes from any line or column.


  • Backed by a relatively small community.
  • Processing can often be quite slow.
  • Lacks regular updates.
  1. PHP

Over the years, PHP has emerged as one of the most developer-friendly open-source programming languages. It is much faster than many other programming languages, and that’s primarily why PHP-based solutions are today being used by developers, entrepreneurs, and SMEs around the globe.


  • Offers cross-platform compatibility. Can run on platforms such as Linux, Windows, Mac OS, and Unix.
  • Comes with pre-defined error reporting constants.
  • Allows the generation of dynamic page content.
  • Is a loosely typed language so you don’t need to explicitly declare the data type of variables.


  • Lacks named parameters.
  • Not ideal for very large projects since it is less modular.
  • The language semantics can only be altered by global configuration parameters that can pose complications during deployment.
  1. Perl

Perl is a mature, stable, and highly feature-rich programming language. It is an excellent choice for scripting and processing large projects. Also, it is backed by a wide community of developers. It runs successfully over a hundred platforms!


  • Excellent file manipulation.
  • Supports cross-platform usage. Thus, codes written for a specific platform can also be run on other platforms with minimal or no modifications.
  • Comes with a library of modules.


  • Can be slower for specific tasks such as scripting.
  • Poor argument handling.
  • Poor usability factor.

To conclude, each language has its distinct benefits and disadvantages. Depending on the kind of project you will be working on, you can choose the open-source programming language that best suits your project requirements.

Happy coding!

How To Prepare For ACM-ICPC?


Considered as the “Olympics of Programming Competitions,” the Association for Computing Machinery – International Collegiate Programming Contest or ACM-ICPC is one of the oldest and most esteemed programming competitions in the world. Every year, more than 2,000 universities spread across eighty countries participate in this contest to win the grand prize of $12,000 and a gold medal awarded by the ACM-ICPC. Apart from this, three runner-up teams receiving the Gold medal are also rewarded with $6,000 and teams that bag the Silver and Bronze medal receive $3,000 and $1,500 respectively.

The ACM-ICPC contest has two rounds, the regional round that is organized and conducted at the local universities of various regions across the world. In India, the Asia Regionals are held at Amritapuri, Chennai, and Kolkata. The regional contest location sites are allotted a ‘slot’ which is the formal invitation to enter the ACM-ICPC World Finals.

Here’s how you should approach the ACM-ICPC preparation!

ACM-ICPC Preparation

The first step in preparing for the ACM-ICPC is to start practicing and competing in Online Judges and contests similar to ICPC. OJs such as CodeChef, TopCoder, Codeforces, Hackerrank, Hackerearth, USACO are excellent online resources to practice and hone your programming skills. Begin by tracking down problems with the most number of submissions, see how people have approached it, and see if you can get to the solution through a different and better perspective.

The next step would be to make a detailed list of the things you have to study. Having a computer science background is highly preferred as you’ll have to deal with a lot of complex data structures and algorithms in the contest.

Programming Language

There are many programming languages to choose from including C++, Java, Ruby, Python, Perl, and so on. Pick one programming language that you think will be the best match for you and master it through and through. Once you master a programming language, you’ll find it much easier to learn other programming languages.

Data Structures

Mastering data structures is one of the prerequisites for participating in any competitive programming contest. Thus, you are expected to be well-versed with the fundamental data structures such as arrays, stacks, queues, strings, heap, and hash, to name a few. Once you’ve got a solid grasp on these basic data structures, it’s time to move on to more advanced data structures like Fenwick Tree, K-D Tree, Segment Tree, Interval Tree, and so on.

Sorting & Searching

After data structures, you should focus on learning sorting and searching functions like quick sort, merge sort, binary search, and order statistics. While learning the basic concepts of sorting and searching functions are fine, you must also familiarize yourself with as many library functions as possible.


Once you learn how to work with strings, you’ll find it extremely interesting. Strings are a must-learn as they are highly used in competitive programming contests. Try to learn how to manipulate string functions such as Z’s algorithm, KMP algorithm, Rabin Karp, and Aho Corasick string matching, to name a few. If you can learn library functions for strings, even better!


Just like data structures, algorithms are also very important for competitive programming. Algorithms can be classified under the following categories:

Dynamic Programming algorithms – Longest Common Subsequence, Longest Increasing Subsequence, Minimum Partition, Longest Path In Matrix, Subset Sum Problem, 0-1 Knapsack Problem, and Assembly Line Scheduling.

  • Graph algorithms – Breadth First Search (BFS), Depth First Search (DFS), Dijkstra, Floyd Warshall, Prim, Kruskal, Johnson’s algorithm, Topological Sort, and Bridges in a graph.
  • Greedy algorithms – Activity Selection Problem, Huffman Coding, Huffman Decoding, Egyptian Fraction, Job Sequencing Problem, Water Connection Problem, and Fitting Shelves Problem.
  • Geometric algorithms – Convex Hull, Graham Scan, Line Intersection, Matrix Exponentiation, Bentley Ottmann algorithm, Rotating Calipers Technique, Closest pair of points, and Voronoi Diagrams of n points using Fortune’s algorithm.
  • Network Flow algorithms – Maxflow Ford Fulkerson algorithm, Edmond Karp Implementation, Min-cut, Stable Marriage Problem, Dinic’s Algorithm, Cycle Cancelling algorithm, Stoer Wagner min-cut algorithm, and Hopcroft–Karp Algorithm.


When it comes to programming, one must have a solid foundation in Mathematics and Statistics since a lot of programming problems are rooted in these two. Thus, you must learn basic math concepts such as number theory, linear algebra, discrete mathematics and combinatorics, probability theory, game theory, graph theory, numerical analysis, and calculus. Also, be well-versed with principles like Induction, Pigeon Hole, Inclusion-Exclusion, and the like. Moving on to Statistics, make sure that you learn concepts like mean, median, variance, and so on.

Mastering the above-mentioned concepts should be enough for you to make it through the regional round of the ACM-ICPC, and once that’s done, there’s nothing to stop you from emerging as a winner in this prestigious competition. The key to success is dedication, commitment, and practice. During the learning process, try to connect with mentors and peers on the various OJ platforms. Usually, such communities are very active and can help you with helpful feedback if you are ever stuck somewhere. Be confident, and begin!

Till then, happy coding!

Getting Started With And Mastering The Basics Of Data Structures


Data Structures is one of the foundation courses in Computer Science. Essentially, it includes concepts such as stacks, queues, graphs, algorithm analysis, sorting algorithms, linked lists, and set implementations (binary search trees and hash tables). While it is true that data structures and algorithms are difficult to master, it is not an impossible feat. The key to learning these tricky concepts is to follow the right approach, and that’s what we’re going to show you in this post.

  1. Strengthen Your Fundamentals

The foremost requirement for mastering data structures and algorithms is a solid foundation of the fundamental concepts of Mathematics, Computer Architecture, and of course Data Structures.  

  • Maths – A concrete mathematical base is essential to learn algorithms. Try to cover Math concepts from different areas such as set theory, regular expressions, linear equations, finite-state machines, matrix multiplication, and basic combinatorial principles like pigeonhole, permutations, and combinations.
  • Primary Data Structures – Strengthen your knowledge about the primary data structure concepts like arrays, hash tables, graphs, stacks, queues, heaps, linked lists, and binary trees.
  • Computer Architecture – To be able to work with data structures and algorithms you need to have basic knowledge of computer arithmetic, boolean algebra, floating-point representation, cache design, and digital logic design. If you are well-versed in programming languages such as C, C++, Python, Java, and Ruby, you’ll have the edge over your competitors.
  1. Learn To Visualize Data Structures

To become a great coder, one must possess the ability to visualize different data structures. You need to picture how a data structure will look like, how you can implement it within your code, and also how it is organized in your computer ’s memory as well as in the abstract. To visualize data structures, you can pen down your thought pattern on paper. As you gradually master visualization, you’ll find it much easier to solve all kinds of problems, be it simple ones like stacks or be it self-balancing trees.

  1. Dive Into Algorithms

Once you’ve learned about the fundamental concepts of data structures, it’s time to leap into algorithms. An excellent way to start would be to learn Big-O notation as it would help you understand how to classify algorithms based on their running time and space requirements. Books like Introduction to Algorithms, and Algorithm Design Manual are great learning options.

As you learn more about algorithms, you should start implementing algorithms in your codes and learn about their running times in real-time. You could try implementing Euclid’s algorithm, Binary search, Binary tree traversals, Dijkstra’s shortest path, Min & max heaps, to name a few. Platforms such as HackerRank, LeetCode, CodeChef, and  Coderbyte are excellent for practicing coding and sharpen your coding skills.

  1. Learn Dynamic Programming

If you wish to master data structures and algorithms, dynamic programming is a must. It is a technique of solving complicated problems by breaking them down into smaller fragments of “subproblems.” These subproblems are then solved at once and their solutions are stored for future reference. Thus, in future if similar problems occur, you don’t need to solve it again from scratch; instead, you can refer to the previous solutions and save both time and effort.

So, there, you have it – your stepwise guide to approaching and mastering data structures and algorithms. If you follow these four steps diligently and with dedicated passion, you can surely master data structures in a matter of few months. And, if you feel stuck anywhere along your journey, don’t forget to check us out at,

Happy learning!

The OOPS Concepts You Must Know


OOPS, or Object Oriented Programming System is one of the most popular programming architectures even today. It is used by software developers around the world owing to the many, many advantages that it offers. However, to be able to use OOPS to its full potential you need to have a fair understanding of its fundamental concepts.

So, we’ve prepared this informative guide to introduce you to some of the basic concepts of OOPS.

  1. Class

A class denotes a blueprint containing elements having certain common traits, that is, they perform the same functions, or they possess the same features. In other words, a class represents data with variables known as ‘fields’ and behaviors with functions known as ‘methods.’

In a class, the members (functions and variables) may either have productive behaviors or may be intended to perform specific actions along with member attributes. The members of a class can only be accessed by the corresponding class or its objects.  

  1. Object

While class denotes that model or prototype, an object is the actual player of a class, that is, an object is an instance of a class. Precisely, objects are variables of a user-defined data type class. The variables of a class contain the data and the functions perform specific operations based on the objects contained in a class.

  1. Inheritance

Often, solving complex programming problems requires you to build classes that possess certain common characteristics or behaviors. Thus, to avoid repetition you need to gather all the common characteristics within a single class known as the ‘parent class.’ Now, all the classes having similar features can ‘inherit’ the common traits from the parent class. This is called ‘Inheritance.’ Apart from inherited features from the parent class, the new classes can have their distinct fields and methods.

There also exists the concept of ‘multiple inheritances’ wherein some object-oreinted programming languages support the inheritance of specific features from multiple classes.

  1. Abstraction

Abstraction refers to the process of eliminating complexity by hiding the unnecessary details of the implementation from the user. This allows the user to apply a more complex logic on top of the given abstraction without even being aware of the underlying or hidden complexity.

  1. Encapsulation

Encapsulation refers to the process of bundling methods and data that function on the data contained within a single unit such as a class. There are two aspects of encapsulation – one is the restrict the access to particular fields and methods of a class and second is the merge similar data and methods.

It is often used to hide the internal representation of an object from the user which is known as ‘information hiding.’

  1. Interface

An interface is a programming protocol for two unrelated classes to communicate with one another. An alternative to inheritance, an interface defines the actions of objects. An interface is placed within its distinct file bearing the same name as the interface (in Capital) and ends with the familiar language extension, for instance, “.as”. The primary purpose of interfaces is to enable the computer to know what kind of properties a particular object should have and to enforce those properties on the object.

  1. Association

An association denotes any type of relationship between two objects. Precisely, it refers to the multiple associations between objects such as one-to-one, one-to-many, many-to-one, and many-to-many. The objects may be instances of unrelated classes and they collaborate to accomplish their unique goals without manipulating the lifecycle of the other.

  1. Aggregation

Aggregation is nothing but a special type of association. It is a relationship where one object belongs to another object but despite that, they are both independent objects. It is also known as a “has-a” relationship.

  1. Composition

A composition is a relationship between two objects where one object is under complete control of the other, and hence, has no independent lifecycle. While it is similar to aggregation as in one object contains another object, unlike aggregation, the object contained cannot exist without the container object.  

  1. Polymorphism

In general terms ‘polymorphism’ refers to the ability to exist in multiple forms. However, in OOPS, the term polymorphism refers to the ability of a programming language to process objects differently in accordance with their data type or class. Thus, polymorphism is essentially a language’s ability to ‘redefine’ the methods for different derived classes.

While this is just a nudge to steer you in the right direction, there are many other concepts that you must learn to master OOPS. For now, starting with the basics would be the wise way to go. If guidance is what you seek, drop by at Coding Ninjas and we’ll get you covered with our wide array of courses in and around the most-used programming languages.

Competitive programming helps in bagging a better job. Here’s how!

FireShot Capture 095 - Competitive programming helps in bagg_ -

Competitive programming has become a worldwide trend among software developers and coders. It is both thrilling and challenging at the same time – the pressure of solving complex coding problems within a limited time and the thrill of being able to perform well! Today, competitive programming sites like TopCoder, Coderbyte, HackerRank, CodeChef, and so on are becoming hugely popular as more and more people are joining in to compete in their coding competitions.


While participating in coding competitions is not an absolute necessity, you should definitely give it a try. Why, you ask? Competitive programming has certain advantages that can help prepare you for coding interviews or better yet, to bag that prestigious job position in the company you’ve been dying to work for!

Let’s look at how competitive programming can help you become a better coder and also bag better jobs in the industry.

Makes You Faster And More Focused

The very essence of competitive programming is to be able to solve coding problems within an allotted time period. The more you participate in such competitions, the better and faster you get at solving problems quickly. Appearing in these contests regularly will enhance your coding skills and speed in solving problems. You will learn how to stay focused on the task because if you lose focus, you lose time, and ultimately, you lose the contest.

Makes You A Team Player

In competitive programming contests, you often have to work with other people on your team and come up with efficient solutions for specific tasks. As you start working with other people, you understand your core strengths and weaknesses as well as those of your team members too. You learn how to stay organized and well-coordinated even during stressful situations.

These qualities are what most companies look for in a potential candidate. However, one of the most compelling reasons to pursue competitive programming is that it provides you exposure.

You Are Being Watched!

Yes, when you enter the world of competitive programming, you are being observed by potential employers. Competitive programming is nothing short of a hunting ground for IT and Tech companies who are always on the lookout for talented, experienced, and well-learned software professionals. In fact, some of these contests are closely followed by industry magnates such as Apple, Google, and Facebook. So, if you are a dedicated and aspiring coder, participating in competitive programming challenges might someday turn out to be a big win for you. If you’re excellent at it, you might catch the attention of a reputed software company or an MNC.

Coding competitions are great for building the right foundation for cracking coding interviews. This is primarily because:

  • Both, coding contests and interviews will require you to have a good knowledge of computer science and data science concepts.
  • During both, you’ll have to write codes and find solutions to problems while adhering to the standard coding norms.

In coding interviews, however, you’ll not only be judged by the deftness of your problem-solving skills but also various other soft skills such as communication skills, team skills, cultural fit, and so on. Thus while, coding competitions are excellent for preparing yourself for the rigorous interview rounds, it alone does not guarantee that you’ll bag a job at reputed firms or companies. The greatest advantage of having a competitive coding background is that it’ll not only help you break the ice during coding interviews but it will also provide you an edge over your competitors.

Many success stories of competitive coders have come to the fore of how they bagged jobs at prestigious companies while leaving some of the top candidates with higher ranking CGPA behind them. For instance, Anudeep Nekkanti from Visakhapatnam bagged a job at Google. Passionate about coding, Anudeep states:


“My failure at IOPC 2012 made me start it. I thought I will do well in IOPC 2013 and stop programming. That is how I started it. Very soon I started to like it…My heart beat raises whenever I submit a solution. I get goosebumps. It was that fun that kept me going. Don’t do it, Play it. Enjoy it…Right now I am preparing for world finals. I am doing problems from various online judges like Topcoder, Codechef, Codeforces.”


So, that’s what we wanted to get across to you – start competitive coding and who knows you too might end up bagging a great job at an MNC just like Anudeep!

Most Commonly Asked ML Interview Questions


While Data is the new currency of the business and the industrial world, Data Science is the pathway to the next Industrial Revolution. The rising importance of data is creating a demand for skilled professionals who are well-versed in data science technologies such as Machine Learning (ML) and Artificial Intelligence (AI). However, bagging a job in the field isn’t a walk in the park. You need to be prepared to face the challenging interview process during which your mastery over a variety of data skills will be assessed such as your basic knowledge about data science and ML concepts; your ability to analyze and visualize data; your technical and programming skills, and so on.

We know that interviews can be tricky and overwhelming, and hence, we’ve prepared a list of ten most commonly asked machine learning interviews questions.

  1. What is the difference between supervised and unsupervised machine learning?

The primary difference between supervised and unsupervised learning is that while supervised learning focuses on training labeled data unsupervised learning does not require the data to be trained explicitly. For instance, for the classification function of supervised learning, one first needs to label the data that will be used to train the data model to classify the data into labeled subsets. This kind of specialized training is not required in unsupervised learning.

  1. What are parametric and non-parametric models?

Parametric models refer to those models that contain a finite number of parameters. In such a model, one only needs to know the parameters of the model to be able to predict new data. Linear regression, logistic regression, Naive Bayes, and Perceptron, are some examples of parametric models. Non-parametric models, on the other hand, contain an unlimited number of parameters and hence, are more flexible. In this model, apart from knowing the parameters of the model you also need to be aware of the state of the observed data. Decision trees, SVMs, and k-nearest neighbors are examples of non-parametric models.

  1. Explain the bias-variance tradeoff.

Predictive models usually have a tradeoff between bias and variance. While bias refers to the error occurring due to erroneous or overly simplistic assumptions in the learning algorithm being used, the variance is the error occurring due to excessively complicated assumptions in the learning algorithm in question. The purpose of the bias-variance is to  minimize the learning error of a specific algorithm by adding the bias and the variance along with some other irreducible errors due that originate from the noise in underlying datasets. For instance, you can reduce the bias by adding more variables to the model to make it complex, but in the process, you’ll add some variance to the model. Thus, to strike a perfect balance in the model, you need to have a tradeoff between bias and variance.

  1. What is the difference between stochastic gradient descent (SGD) and gradient descent (GD)?

Both SGD and GD algorithms are techniques of finding a set of parameters that can reduce the loss function of a model. The parameters are first evaluated against the data, and then adjustments are made accordingly. However, there lies a subtle difference in the approach of the two algorithms. While in GD one needs to evaluate all the training samples for each set of parameters, in SGD you need to evaluate only one training sample for the given set of parameters. Also, GD is ideal for small datasets while SGD is ideal for more massive datasets.

  1. What is the purpose of the Box-Cox transformation?

The Box-Cox transformation is a standard power transformation process of transforming datasets to facilitate normal distribution. In other words, it is used to stabilize the variance in datasets. Since most well-known statistical methods sync well with normally distributed data, it is wise to normalize the distribution using this method.

  1. Why is Naive Bayes ‘naive’?

Naive Bayes is considered to be ‘naive’ mainly because it makes such assumptions that are nearly impossible to observe in real-life data. This algorithm assumes that the presence or absence of a particular feature of a class is unrelated to the presence or absence of any other feature of the class variable in question. This entails the “absolute independence of features,” a condition that can never be fulfilled in reality.

  1. What is the difference between machine learning and deep learning?

Deep learning is a branch of machine learning exclusively concerned with neural networks. It focuses on the ways to leverage certain principles of neuroscience to model large sets of unstructured or semi-structured data with increased accuracy. To be precise, deep learning is much like an unsupervised learning algorithm that aims to ‘learn’ data representations by leveraging neural nets.

  1. How will you choose a classifier based on a training set?

In case the training set is a small, models with high bias/variance, for example, Naive Bayes, are the best fit since they are less likely to overfit. Whereas if the training set is too large, models with low bias/variance such as logistic regression are best as they can detect more complex relations in data models.

  1. What is Latent Dirichlet Allocation (LDA)?

Latent Dirichlet Allocation (LDA) is a generative model that represents documents as an amalgamation of topics, each of which has their distinct probability distribution of possible words. In other words, LDA is a technique of classifying topics or documents according to the subject matter.

  1. What is the ROC curve? What is AUROC?

The ROC (Receiving Operating Characteristic) curve is a graphical representation of the contrast between true positive rates and the false positive rate at varying thresholds. It’s mostly used to assess the sensitivity of the true positives against the false positives’ probability to trigger a false alarm.

AUROC (Area Under the Receiving Operating Characteristic) denotes a standard performance metric used to evaluate binary classification models.

Are you eager to master machine learning? Do you wish to begin a career in machine learning? Come, join our Machine Learning course at Coding Ninjas today!