Machine Learning Interview Questions | Machine Learning Interview Preparation | Intellipaat

Summary

This video dives into essential machine learning interview questions, explaining key concepts such as the differences between machine learning, artificial intelligence, and deep learning. It covers topics like bias and variance, clustering, linear regression, decision trees, and overfitting. The script also explores hypothesis testing, supervised vs. unsupervised learning, PCA, SVM, cross-validation, entropy, epochs, and the variance inflation factor. It discusses metrics like confusion matrices, type 1 and type 2 errors, and the use of logistic regression. Additionally, it provides insights on handling missing data in datasets, offering a comprehensive guide for those preparing for a career in data science.

Takeaways

• 🤖 Machine Learning, Artificial Intelligence (AI), and Deep Learning are distinct yet interrelated fields, with Deep Learning being a subset of Machine Learning, and Machine Learning being a subset of AI.
• 🔍 Bias in machine learning refers to the difference between a model's average prediction and the correct value, while Variance measures the fluctuation in the model's output, with lower values being preferable for both.
• 👥 Clustering is an unsupervised learning technique that groups similar data points together based on features and properties, with algorithms like K-Means and Mean Shift Clustering being commonly used.
• 📊 Linear Regression is a supervised learning algorithm that models the linear relationship between dependent and independent variables for predictive analysis.
• 🌳 Decision Trees are a hierarchical model used to map out decisions and actions, helping to predict outcomes based on a sequence of choices.
• 🔧 Overfitting occurs when a model learns the training data too well, including its noise and outliers, which can be mitigated by techniques like cross-validation.
• ✂️ Hypothesis Testing in machine learning involves using a dataset to approximate an unknown target function that maps inputs to outputs effectively.
• 🏷️ Supervised Learning uses labeled data to train models that can predict outcomes, while Unsupervised Learning works with unlabeled data to discover underlying structures and patterns.
• 📚 The Bayes' Theorem is fundamental in machine learning, particularly for Bayesian Belief Networks and Naive Bayes classifiers, providing a way to calculate conditional probabilities.
• 📉 Principal Component Analysis (PCA) is a technique used to reduce the dimensions of multi-dimensional data by keeping only the most relevant dimensions, helping with data visualization and analysis.
• 🛡️ Support Vector Machines (SVM) are used for classification tasks and work by finding the hyperplane that best separates data into different classes.
• 🔄 Cross-Validation is a technique to ensure that a machine learning model generalizes well to an independent dataset, involving methods like hold-out, k-fold, and leave-one-out.
• 🗂️ Entropy measures the randomness or unpredictability in data, with higher entropy indicating more difficulty in drawing conclusions from the data.
• 🔄 Epoch refers to a complete pass through the entire training dataset in machine learning, with the number of epochs affecting the model's training.
• 🔄 Variance Inflation Factor (VIF) is used to estimate the amount of multicollinearity in regression variables, helping to identify and manage it.
• 🔢 Confusion Matrix is a tool used to evaluate the performance of classification models by summarizing the counts of correct and incorrect predictions.
• 🚫 Type 1 and Type 2 errors refer to False Positives and False Negatives respectively, which are critical to understand when evaluating the accuracy of predictive models.
• 🏠 The choice between using Classification or Regression depends on the nature of the prediction task, with regression used for numerical predictions and classification for categorical outcomes.
• 📈 Logistic Regression is used for binary or categorical dependent variables, predicting the probability of an event occurring.
• 🧩 Handling Missing Values in datasets can be done using methods like detecting with `isnull()`, removing with `dropna()`, or filling with placeholder values using `fillna()` in Python's pandas library.

Q & A

What is the average salary of a machine learning engineer in the United States according to the video?

According to the video, the average salary of a machine learning engineer in the United States is around $112,742 per year.

How much does a machine learning engineer typically earn in India per year?

The video states that the average salary of a machine learning engineer in India is around 9 LPA (Lakhs per Annum) per year.

What is the relationship between machine learning, artificial intelligence, and deep learning?

As explained in the video, deep learning is a subset of machine learning, and machine learning is a subset of artificial intelligence. These technologies are interrelated but distinct, with overlapping terms and techniques.

What is the difference between bias and variance in machine learning?

Bias in machine learning is the difference between the average prediction of a model and the correct value, while variance is the difference of predictions over a training set and anticipated value of another training set. High bias can lead to inaccurate predictions, and high variance can lead to large fluctuations in the output.

Can you explain what clustering is in the context of machine learning?

Clustering, as mentioned in the video, is an unsupervised learning technique used for grouping data points with similar features and properties into distinct categories. Algorithms like k-means and mean shift clustering help in classifying data points into their respective groups.

What is linear regression and how is it used in machine learning?

Linear regression is a supervised machine learning algorithm used to find the linear relationship between dependent and independent variables for predictive analysis. It is represented by the equation y = a + b * x, where 'a' is the intercept, 'b' is the coefficient, 'x' is the independent variable, and 'y' is the dependent variable.

What is a decision tree in machine learning and how does it work?

A decision tree in machine learning is a hierarchical diagram used to explain a sequence of actions that must be performed to get a desired output. It helps in making decisions by breaking down a complex problem into simpler steps based on a set of conditions.

What is overfitting in machine learning and how can it be avoided?

Overfitting occurs when a machine learning model learns the training data too well, including its noise and outliers, leading to poor generalization on new data. It can be avoided by using techniques like cross-validation, which divides the data set into training and testing subsets to ensure the model performs well on unseen data.

What is hypothesis testing in machine learning and what is its purpose?

Hypothesis testing in machine learning involves using a dataset to understand a specific function that maps inputs to outputs in the best possible way, known as function approximation. The goal is to find a model that approximates the target function and performs necessary input-output mappings.

What is the main difference between supervised and unsupervised learning in machine learning?

Supervised learning uses labeled data to train the model, providing both input and output data, with the aim of predicting outputs for new data. Unsupervised learning, on the other hand, uses unlabeled data to identify hidden trends without any feedback, aiming to extract information from unknown datasets.

What is the purpose of Principal Component Analysis (PCA) in machine learning?

PCA is used in machine learning to reduce the dimensions of multi-dimensional data by removing irrelevant dimensions and keeping only the most relevant ones. It finds a new set of uncorrelated dimensions or orthogonal dimensions and ranks them based on variance.

What is a Support Vector Machine (SVM) and how is it used in machine learning?

A Support Vector Machine (SVM) is a machine learning algorithm primarily used for classification tasks. It operates on high-dimensional feature spaces and is designed to find the optimal hyperplane that separates data points into different classes.

What are the different techniques of cross-validation in machine learning?

The video mentions several cross-validation techniques: hold-out method, k-fold cross-validation, stratified k-fold cross-validation, and leave-p-out cross-validation. These methods help in evaluating the performance of a machine learning model by using different subsets of the data for training and testing.

What does entropy measure in the context of machine learning?

In machine learning, entropy measures the randomness or unpredictability in the data. The higher the entropy, the more difficult it is to draw useful conclusions from the data, as it indicates a higher level of disorder or randomness.

What is an Epoch in machine learning and how is it related to training a model?

An Epoch in machine learning refers to a complete pass through the entire training dataset. It indicates the number of times the training process has worked through the entire dataset. The relationship between epochs, dataset size, iterations, and batch size can be understood through the formula D * E = I * B, where D is the dataset, E is the number of epochs, I is the number of iterations, and B is the batch size.

What is a confusion matrix and how does it help in evaluating a classification model?

A confusion matrix is a tool used to evaluate the performance of a classification model by summarizing the predictions and comparing them with the actual outcomes. It provides counts of correct and incorrect predictions and helps identify the uncertainty between classes, contributing to the calculation of accuracy and other performance metrics.

What are Type 1 and Type 2 errors in the context of testing and evaluation?

Type 1 error, also known as a false positive, occurs when a test incorrectly indicates that a condition is present when it is not. Type 2 error, or false negative, happens when a test fails to detect a condition that is actually present. These errors are important considerations in the evaluation and interpretation of test results.

When should classification be used over regression in predictive modeling?

Classification should be used over regression when the task involves predicting categorical or discrete outcomes, such as determining whether an event belongs to a specific category. Regression, on the other hand, is used for predicting continuous numerical values, like the price of a house.

What is logistic regression and how is it different from linear regression?

Logistic regression is a type of regression analysis used when the dependent variable is categorical or binary. Unlike linear regression, which predicts continuous outcomes, logistic regression is used to predict the probability of a certain class or event occurring and is particularly useful for binary classification tasks.

How can missing values in a dataset be handled using Python's pandas library?

In Python's pandas library, missing values can be handled using functions like 'isnull()' to detect missing values, 'dropna()' to remove rows or columns with null values, and 'fillna()' to fill missing values with placeholder values or statistics like mean or median.

Outlines

00:00

🌟 Introduction to Machine Learning Interview Questions

The video begins with an introduction to machine learning as an exciting field that enables computers to learn from data and make decisions without explicit programming. It highlights the popularity and high demand for machine learning engineers across various industries and mentions average salaries in the United States and India. The video then invites viewers to subscribe and prepare for a series of interview questions starting with the basic definitions of machine learning, artificial intelligence, and deep learning, emphasizing their differences and relationships.

05:02

📚 Understanding Bias and Variance in Machine Learning

This section delves into the concepts of bias and variance in machine learning. Bias refers to the difference between a model's average predictions and the correct value, with a lower bias indicating more accurate predictions. Variance is the difference between predictions over a training set and the expected value from other training sets, with a lower variance desired to avoid large fluctuations in output. The video explains the trade-off between bias and variance and provides a visual representation to help viewers understand the balance needed for optimal model performance.

10:03

🤖 Clustering Techniques in Machine Learning

The script explains clustering as an unsupervised learning technique that groups similar data points together. It introduces k-means clustering as a common algorithm used to find hidden patterns in data and classify it into groups based on feature similarity. The mean shift clustering is also mentioned, which differs from k-means by automatically discovering the number of clusters without pre-specifying it. The video aims to provide a brief understanding of clustering algorithms and their applications in machine learning.

15:04

📈 Linear Regression and Decision Trees in Machine Learning

The video discusses linear regression, a supervised learning algorithm used to find linear relationships between dependent and independent variables for predictive analysis. It uses the equation y = a + b*x to illustrate this relationship and explains the process of finding the best fit line by adjusting coefficients. The decision tree is also introduced as a hierarchical diagram that represents a sequence of actions to achieve a desired output, using an example of driving with or without a license to demonstrate its application in machine learning.

20:06

🔍 Overfitting, Hypothesis Testing, and Learning Types

This part of the video addresses the issue of overfitting, which occurs when a model learns from an inadequate dataset, and how it can be mitigated using cross-validation techniques. Hypothesis testing is introduced as a method to understand the function that maps inputs to outputs, with the model approximating this function. The difference between supervised and unsupervised learning is also explained, with supervised learning using labeled data for training and unsupervised learning identifying hidden trends in unlabeled data.

📊 Principal Component Analysis (PCA) and Support Vector Machines (SVM)

The script introduces principal component analysis (PCA) as a method to reduce the dimensions of multi-dimensional data by removing irrelevant dimensions and keeping the most relevant ones, which is essential for data visualization and analysis. Support vector machines (SVM) are then discussed as a classification algorithm used in high-dimensional spaces, highlighting their role in machine learning for classification tasks.

🔄 Cross-Validation and Entropy in Machine Learning

Cross-validation is explained as a technique to enhance the performance of a machine learning algorithm by breaking the dataset into smaller parts for training and testing. Various cross-validation techniques are mentioned, including hold-out method, k-fold cross-validation, and others. Entropy is introduced as a measure of randomness in data, with higher entropy indicating more difficulty in drawing conclusions from the data.

🔢 Epoch, Variance Inflation Factor, and Confusion Matrix

The concept of an epoch in machine learning is defined as the count of passes over a training dataset, with the relationship between epochs, iterations, and batch size explained through a formula. The variance inflation factor is introduced as an estimate of multicollinearity in regression variables. The confusion matrix is also discussed as a tool to evaluate the performance of classification models by providing a summary of correct and incorrect predictions.

🚫 Type 1 and Type 2 Errors, and Choosing Between Classification and Regression

The video clarifies type 1 and type 2 errors, with type 1 being a false positive and type 2 being a false negative, using examples to illustrate each. It also discusses when to use classification over regression, explaining that classification is for identifying groups while regression is for predicting numerical outcomes, with examples provided to distinguish between the two.

🧮 Logistic Regression and Handling Missing Values

Logistic regression is introduced as a method for predictive analysis when the dependent variable is categorical or binary, used to predict probabilities of categorical outcomes. The video provides examples of its application, such as predicting seniority or disease presence. The final topic is handling missing values in a dataset using Python's pandas library, with methods like 'isnull' for detection, 'dropna' for removal, and 'fillna' for replacing missing values.

Keywords

💡Machine Learning

Machine learning involves developing algorithms and statistical models that enable computers to learn from data and make predictions or decisions without being explicitly programmed. It is increasingly popular in industries such as finance, healthcare, and e-commerce. The video introduces machine learning as a core topic, explaining its importance and applications.

💡Artificial Intelligence

Artificial intelligence (AI) uses machine learning and deep learning techniques to enable computer systems to perform tasks that normally require human intelligence, such as reasoning, learning, and self-correction. The video explains AI as a broader field that encompasses machine learning and highlights its ability to perform tasks using human-like intelligence.

💡Deep Learning

Deep learning is a subset of machine learning that involves neural networks with many layers. It enables software to learn from vast amounts of data and perform complex tasks like image and speech recognition. The video differentiates deep learning from machine learning and AI, describing its use of multi-layered neural networks.

💡Bias

Bias in machine learning is the difference between the average prediction of a model and the correct value. High bias indicates that the model's predictions are not accurate. The video explains that bias should be minimized to improve the accuracy of machine learning models.

💡Variance

Variance measures the variability of model predictions for different training sets. High variance can lead to large fluctuations in the model's output, making it less reliable. The video discusses the trade-off between bias and variance and the importance of achieving a balance to avoid overfitting.

💡Clustering

Clustering is an unsupervised learning technique used to group data points with similar features into clusters. Examples include k-means clustering and mean-shift clustering. The video describes clustering as a method to classify data points into groups based on their properties.

💡Linear Regression

Linear regression is a supervised machine learning algorithm used to find the linear relationship between dependent and independent variables for predictive analysis. The video explains the equation of linear regression and its application in predicting outcomes based on input data.

💡Decision Tree

A decision tree is a hierarchical diagram used to explain a sequence of actions to achieve a desired outcome. It is a popular algorithm in machine learning for making decisions based on data. The video illustrates how decision trees work and their role in machine learning.

💡Overfitting

Overfitting occurs when a machine learning model learns from an inadequate data set, leading to poor generalization to new data. The video explains how overfitting can be avoided using cross-validation, where the data set is divided into training and testing subsets.

💡Hypothesis Testing

Hypothesis testing in machine learning involves using data to determine if a specific hypothesis about a target function is true. The video explains that hypothesis testing helps in approximating the target function and performing necessary input-to-output mappings.

💡Supervised Learning

Supervised learning uses labeled data to train models, which receive direct feedback to improve their predictions. The video contrasts supervised learning with unsupervised learning, emphasizing the use of input-output pairs to train models.

💡Unsupervised Learning

Unsupervised learning uses unlabeled data to identify hidden patterns without direct feedback. The video explains that unsupervised learning aims to extract information from unknown data sets and can result in less accurate outcomes compared to supervised learning.

💡PCA (Principal Component Analysis)

PCA is a technique used to reduce the dimensionality of data by keeping only the most relevant dimensions. The video describes PCA as a method to simplify data visualization and analysis by focusing on the dimensions with the most variance.

💡SVM (Support Vector Machine)

SVM is a machine learning algorithm used primarily for classification tasks. It works by finding the hyperplane that best separates different classes in high-dimensional space. The video highlights SVM's role in classification problems.

💡Cross-Validation

Cross-validation is a technique to evaluate the performance of a machine learning model by splitting the data into training and testing sets. The video discusses various cross-validation methods, such as k-fold and stratified k-fold, to ensure robust model evaluation.

💡Entropy

Entropy measures the randomness in a data set. In machine learning, higher entropy indicates more unpredictability in the data, making it harder to draw conclusions. The video uses the example of coin flipping to illustrate the concept of entropy.

💡Epoch

An epoch in machine learning refers to one complete pass through the entire training data set. The video explains how epochs relate to iterations and batch size in the training process of machine learning models.

💡Variance Inflation Factor (VIF)

VIF measures the level of multicollinearity in regression variables. High VIF indicates that the independent variables are highly correlated, which can affect the model's performance. The video briefly introduces VIF as an important metric in regression analysis.

💡Confusion Matrix

A confusion matrix is a tool used to evaluate the performance of classification models by comparing the predicted and actual values. The video describes how the confusion matrix provides a summary of prediction accuracy and error types.

💡Type 1 and Type 2 Errors

Type 1 error (false positive) occurs when a test incorrectly rejects a true condition, while Type 2 error (false negative) occurs when a test incorrectly accepts a false condition. The video provides examples of these errors in the context of medical diagnoses.

💡Logistic Regression

Logistic regression is used for predictive analysis when the dependent variable is categorical or binary. The video explains how logistic regression models the probability of a binary outcome and provides examples of its application.

💡Handling Missing Values

Handling missing values in data sets can be done using methods like isnull(), dropna(), and fillna() in Python pandas. The video explains these methods for detecting and managing missing data to ensure data quality.

Highlights

Machine learning is an exciting field involving algorithms and statistical models that enable computers to learn from data.

Machine learning is popular in various industries such as Finance, Healthcare, and e-commerce.

The average salary of a machine learning engineer in the United States is around $112,742 per year.

In India, the average salary for a machine learning engineer is approximately 9 LPA per year.

Machine learning, artificial intelligence, and deep learning are distinct yet interrelated technologies.

Deep learning is a subset of machine learning, which in turn is a subset of artificial intelligence.

Bias in machine learning refers to the difference between a model's average prediction and the correct value.

Variance in machine learning is the difference of predictions over a training set and anticipated value of other training sets.

Clustering is an unsupervised learning technique used for grouping similar data points.

K-means clustering is a popular algorithm for finding hidden patterns in data and classifying it into various groups.

Linear regression is a supervised machine learning algorithm used to find the linear relationship between variables for predictive analysis.

Decision trees are used to explain a sequence of actions that must be performed to get the desired output.

Overfitting occurs when a machine learning model learns from an inadequate dataset.

Hypothesis testing in machine learning involves using a dataset to understand a specific function that maps input to output.

Supervised learning uses labeled data to train models and confirm the correctness of predictions.

Unsupervised learning uses unlabeled data to identify hidden trends without feedback.

Bias theorem provides the probability of an event occurring using prior knowledge.

Principal Component Analysis (PCA) is used to reduce the dimension of data by keeping only the most relevant dimensions.

Support Vector Machines (SVM) are used for classification on high dimensionality of characteristic vectors.

Cross-validation is used to increase the performance of a machine learning algorithm by using sample data.

Entropy in machine learning measures the randomness in data, affecting the ease of drawing conclusions.

An Epoch in machine learning indicates the count of passes in a training dataset by the algorithm.

Variance Inflation Factor (VIF) estimates the volume of multi-collinearity in regression variables.

A confusion matrix is used to explain a model's performance and summarize predictions of classification problems.

Type 1 error (false positive) and Type 2 error (false negative) are significant concepts in understanding test outcomes.

Classification should be used over regression when predicting categorical outcomes, while regression is for numerical predictions.

Logistic regression is used for binary or categorical dependent variables and predicts probabilities of outcomes.

Handling missing values in a dataset can be done using methods like isnull, dropna, and fillna in Python pandas.

Transcripts

[Music]

hello everyone and welcome to today's

video on machine learning interview

questions on intellipart

do you know friends that machine

learning is an exciting field that

involves developing algorithms and

statistical models that enables

computers to learn from data and make

predictions or decisions without being

explicitly programmed machine learning

is becoming increasingly popular in wide

range of Industries including Finance

Healthcare and e-commerce

according to the pay scale the average

salary of a machine learning engineer in

the United States is around 112 742

dollars per year while going upwards

it's around 160 000 per year in India

the average salary of a machine learning

engineer is around 9 LPA per year

so without further Ado let's dig dive

and discuss our interview questions but

before that do not forget to hit the

Subscribe button and click the Bell icon

so let's start with machine learning

interview question here is your first

question it is a kind of pretty basic

question and the question is explain

machine learning artificial intelligence

and deep learning

it is very common to get confused

between the three in-demand Technologies

which are machine learning artificial

intelligence and deep learning these

three Technologies throw a little

different from one another but are

quietly interrelated while deep learning

is a subset of machine learning machine

learning is a subset of artificial

intelligence some terms and techniques

May overlap in these Technologies and it

is quite easy to get confused among them

so let's learn about these Technologies

if I talk about machine learning machine

learning involves various statisticals

and deep learning techniques that allows

machines to use their past experiences

and get better at performing specific

tasks without having been to be

monitored if I talk about artificial

intelligence artificial intelligence

uses numerous machine learning and deep

learning techniques that enable computer

systems to perform tasks using

human-like intelligence with logic and

rules

if I talk about deep learning then deep

learning comprises of several algorithms

that enable softwares to learn from

themselves and perform various business

tasks including image and speech

recognition deep learning is a possible

when systems expose their multi-layered

neural networks to a large volume of

data

I hope so guys you would have got brief

idea regarding machine learning

artificial intelligence and deep

learning

so our next question is what is the

difference between bias and variance in

machine learning

the answer to the same question is that

bias is a difference between the average

prediction of a model and the correct

value of the model if the bias value is

high the prediction of the model is not

accurate hence the bias value should be

as low as possible to make the desired

predictions if I talk about the variance

variance is a number that gives a

difference of predictions over a

training set and anticipated value of

another training sets High variance may

lead to large fluctuations in the output

therefore a model's output should have a

low variance if you could see in a

diagram you could see the following

trade-off so here is a bias and variance

trade-off here is a desired result in

the blue circle at the center if you get

off from the blue section then the

prediction goes on wrong I hope so guys

you would have got a brief idea

regarding the difference between bias

and variance in the machine learning now

let's move on to our next question

our next question is what is clustering

in machine learning

if I talk about clustering clustering is

a technique which is used in

unsupervised learning that involves

grouping data points the clustering

algorithms can be used with a set of

data points this technique will allow

you to classify all the data points into

their particular groups the data points

that are thrown into the same category

have similar features and properties

while the data points that belong to a

different group have distinct features

and properties statistical data analysis

can be performed by this method let us

take at some of the examples some of the

examples can be k-means clustering if I

talk about the k-means clustering this

algorithm will be commonly used when

there is a data with no specific group

or category K means clustering allows

you to find the hidden patterns in the

data which can be used to classify the

data into the various groups the

variable key is used to represent the

number of groups the data is divided

into and the data points are clustered

using the similarity of features here

the centroids of the Clusters are used

for labeling new data another clustering

algorithm can be mean shaped clustering

if I talk about the mean shift

clustering the mean name of this

algorithm will be to update the center

point and the candidates to be mean and

find the center points of all the groups

in mean shift clustering unlike k-means

clustering the possible number of

clusters need not to be selected as it

can be automatically be discovered by

the mean shift so here are some of the

examples of the clustering algorithms I

hope so guys you would have got a brief

idea regarding what is clustering in

machine learning now let's move on and

discuss our next question

which is what is linear regression in

machine learning this is the most

popular questions which is asked in the

machine learning interview now let's

discuss the answer of the same

layer regression is a supervised machine

learning algorithm which is used to find

the linear relationship between the

dependent and the independent variables

for predictive analysis

here the equation can be consider y

equals to A plus b dot X where X is the

input or the independent variable where

Y is the output or a dependent variable

and a is the intercept and B is the

coefficient of x you can see here the

diagram with best fit shows that the

data of the weight y or a dependent

variable and the heter of the X of the

independent variable

here the straight line shows that the

best linear relationship that would help

in predicting the weight of the

candidates according to their height to

get this best fit line the best values

of A and B should be found by adjusting

the values of A and B the errors in the

prediction of Y can be reduced this is

how the linear regression helps in

finding the linear relationship and

predicting the output I hope so guys you

would have got a brief idea regarding

what is linear regression in machine

learning now let's move forward and

discuss our next question

our next question is what is decision

Tree in machine learning if I talk about

the decision tree a decision tree is

used to explain a sequence of actions

that must be performed to get the

desired output it is a hierarchical

diagram that shows the action this

algorithm can be created for a decision

Tree on the basis of the set of

hierarchy of actions in the above

decision tree diagram a sequence of

actions has been made for driving a

vehicle with or without license so you

can see how the decision tree algorithm

works in the machine learning domain I

hope so guys you would have got a brief

idea regarding what is decision Tree in

machine learning now let's move forward

and discuss our next question

our next question is what is overfitting

in machine learning

actually overfitting happens when a

machine learning has an inadequate data

set and tries to learn from it so

overfitting is inversely proportional to

the amount of data for small databases

overfitting can be bypassed by the cross

validation method in this approach a

data set is divided into two sections

these two sections will comprise the

testing and training data set to train

the model the training data set is used

and for testing of the model for new

inputs the testing data set is used this

is how we can avoid the overfitting I

hope so guys you would have got a brief

idea regarding what is overfitting in

machine learning now let's move forward

and discuss our next question

our next question is what is hypothesis

testing if I talk about the hypothesis

testing machine learning allows the use

of available data set to understand a

specific function that Maps the input to

the output in the best possible way this

problem is known as a function

approximation here the approximation

need to be used for the unknown Target

function that maps all plausible

observations based on a given problem in

the best manner hypothesis in machine

learning is a model that helps in

approximating the Target function and

Performing the necessary input to Output

mappings the choice and configuration of

algorithms allow defining the space of

the plausible hypothesis that may be

represented by the model in hypothesis

the lower Edge is used for a specific

hypothesis where the uppercase Edge or

the Capital Edge is used for hypothesis

space that is being searched so this is

what exactly the hypothesis testing is I

hope so guys you would have got a fair

idea regarding what exactly is

hypothesis testing now let's move

forward and discuss our next question

so the next question is what is the

difference between supervised and

unsupervised learning if I talk about

the supervised learning the algorithms

of the supervised learnings used the

label data to get trained the model

takes the direct feedback to confirm

with the output that is being predicted

is indeed correct moreover both the

input data and the output data are

provided to the model and the main aim

here is to train the model to predict

the output upon receiving the new data

supervoice learning offers accurate

results and can largely be divided into

two parts which is classification and

regression if I talk about the

unsupervised learning the algorithms of

the unsupervised learning use unlabeled

data for training purposes in

unsupervised learning the models

identify hidden data Trends and do not

take any feedback the unsupervised

learning model is only provided with

input data unsupervised learnings main

aim is to identify hidden patterns to

extract information from the unknown

sets of data it can also be classified

into two parts which is clustering and

Association unfortunately unsupervised

learnings can offer results that are

comparatively less accurate I hope so

guys you would have got a brief idea

regarding what is the difference between

supervised and unsupervised learning now

let's move on and discuss our next

question

so our next question is what is bias

theorem if I talk about the bias theorem

bias theorem offers the probability of

Any Given amount to occur using the

prior knowledge in mathematical terms it

can be defined as a true positive rate

of the given sample conditions divided

by the sum of the true positive rate of

the set conditions and false positive

rate of the entire population

two of the most significant applications

of bias theorem in machine learning are

and byzene belief networks this theorem

is also the foundation behind the

machine learning brand that involves

need bias classifier so as you can see

the formula here P of a by b equals to P

of B by a DOT p a divided by PB where P

of a by B is a probability of occurring

B given the evidence B has already

occurred where P of B by a is equals to

probability of B occurring given the

evidence a has already occurred here PA

is a probability of a occurring and PB

is a probability of B occurring I hope

so guys you would have got the idea

regarding what is bias theorem now let's

move on and discuss our next question

so our next question is what is PCA in

machine learning

if I talk about the multi-dimensional

data it plays an important role in the

real world data visualization and

computations become more challenging

with increase in the dimension in such

scenarios the dimension of data might

have to be reduced to analyze and

visualize it easily this is done by

removing irrelevant dimensions and

keeping only the most relevant Dimension

this is where the principal component

analysis is used the goal of the PCA is

to find a fresh collection of

uncorrelated Dimension or orthogonal and

rank them on the basis of variance

which defines the process of PCA in

machine learning I hope so guys you

would have got a brief idea regarding

what is PCA in machine learning now

let's move on and discuss our next

question

our next question is what is svm in

machine learning if I talk about svm or

support Vector machines it is a machine

learning algorithm that is majorly used

for classification it is used on the top

of the high dimensionality of the

characteristic vector which basically

defines what svm is in machine learning

I hope so guys you would have got a fair

idea what are support Vector machines in

machine learning now let's move on and

discuss our next question

so next question is what is cross

validation in machine learning if I talk

about the cross validation cross

validation allows a system to increase

the performance of a given machine

learning algorithm which is fed a number

of sample data from the data set this

sampling process is done to break the

data set into smaller parts that have

the same number of rows out of which a

random part is selected as a test set

and rest of the parts are kept as a

train sets cross validation consists of

the following techniques which can be

hold out method k-fold cross validation

stratified k-fold cross-validation and

leave P out cross validation I hope so

guys you would have got a fair idea

regarding what is cross validation in

machine learning

our next question is what is entropy in

machine learning the answer to the same

question is entropy in machine learning

measures the randomness in the data that

needs to be processed the more entropy

in the given data the more difficult it

becomes to draw any useful conclusion

from the data for example let us take

the flipping of a coin the result of

this act is random and it does not favor

heads or tails here the result of any

number of tosses cannot be predicted

easily as there is no definite

relationship between the action of

flipping and the possible outcomes I

hope so guys you would have got a brief

idea regarding entropy in machine

learning now let's move forward and

discuss our next question our next

question is what is Epoch in machine

learning

if I talk about Epoch in machine

learning which is basically used to

indicate the count of passes in a given

training data set where the machine

learning algorithm has done its job

generally when there is last chunk of

data it is grouped into several batches

and all these batches go through the

given model and this process is referred

to as iteration

now if the batch size comprises the

complete training data set the count of

iteration is same as that of epochs in a

case there is more than one batch which

is equals to D dot E equals to I star B

which is a Formula where D is a data set

and E is the number of epochs and I is

the number of iterations and B is the

batch size so you can remember d dot E

equals to I dot b where D is a data set

e is the number of epochs and I is the

number of iteration and where b equals

to batch size so this equation generally

defines the relationship between Epoch

data set iteration and number of batches

I hope so guys you would have got a fair

idea regarding what is Epoch in machine

learning now let's move on and discuss

our next question

our next question is what is the

variance inflation Factor the various

inflation factor is the estimate of the

volume of multi-collinearity in a

collection of many regression variables

where vif equals to variance of the

model divided by variance of the model

with single independent variable

that's it for variance inflation factor

I also guys you would have got a brief

idea regarding what is variation

inflation factor and it is one of the

most important questions which can be

asked in a machine learning interview

now let's move on and discuss our next

question

so our next question is what is a

confusion Matrix if I talk about the

confusion metric it is used to explain

modern's performance and gives the

summary of predictions of classification

problems it assists in identifying the

uncertainty between classes confusion

Matrix gives the count of correct and

incorrect values and error types

according to the model where you can see

accuracy is defined as TP plus TN

divided by TP plus TN plus FP plus FL

I hope so guys you would have got a fair

idea regarding what is confusion Matrix

now let's move on and discuss our next

question so our next question is what is

the type 1 and type 2 error type 1 error

is false positive is an error where the

outcome of a test shows the

non-acceptance of a true condition for

example suppose a person gets diagnosed

with depression even when they are not

suffering from the same it is a case of

false positive if I talk about the type

2 error type 2 error or false negative

is an error where the outcome of a test

shows the acceptance of a false

condition for example the CT scan of a

person shows that they do not have a

disease but in fact they do have a

disease here the test accepts the false

condition that the person does not have

the disease this is a case of false

negative I hope so guys you would have

got the idea regarding what are type 1

and type 2 error

now let's move on and discuss our next

question

so our next question is when should

classification be used over regression

both classification and regression are

associated with prediction

classification involves the

identification of values or entities

that lie in a specific group regression

entails predicting a response value from

the consecutive set of outcomes for

example if you want to predict the price

of a house you should use regression

since it is a numerical variable however

if you are trying to predict whether the

house situated in a particular area is

going to be high medium or low price

then the classification model should be

used

I hope so guys you would have got a fair

idea that when should you use

classification over regression now let's

move forward and discuss our next

question

so next question is explain logistic

regression

is also one of the most asked question

in a machine learning interview

so logistic regression is a proper

regression analysis when the dependent

variable is categorial or binary like

all regression analysis logistic

regression is a technique for predictive

analysis logistic regression is used to

explain data and a relationship between

one dependent binary variable or one or

more independent variable logistic

regression is also employed to predict

the probability of categorial dependent

variables logistic regressions can be

used in the following scenarios so here

are some of the examples such as to

predict whether a citizen is senior or

not or to check whether a person has a

disease or not

I hope so guys you will have got a fair

idea what exactly is logistic regression

now let's move on and discuss our next

question

so our final question is how do I handle

missing values in a data set so if I

correctly answer this question in Python

pandas there are two possible methods to

locate the lost or corrupted data and

discard those values the first function

is is null it can be used for detecting

the missing values the second one is

dropner where it can be used for

removing a column or row with null

values and there is also another which

is called filner which can be used to

fill all the word values with the

placeholder values I hope so guys you

would have got a fair idea regarding how

to handle the missing values in a data

set

thank you guys for watching this video

that was all for today's session I hope

so you would have enjoyed our today's

video on machine learning interview

questions if you want to make a career

in data science then intellipat has IIT

Madras Advanced Data science and AI

certification program

this course is of very high quality and

cost effective as it is taught by IIT

professors and Industry experts

thank you