Today's Interview QnAs

Company Name - Fractal Analytics

Role- Data Scientist



Q. What is residual error?

A. The difference between a group of values observed and their arithmetical mean. The difference between what was expected and what was predicted is called the residual error.



Q. Select an appropriate value of k in k-means?

A. The elbow approach is a popular way for determining the ideal value of K when using the K-Means Clustering Algorithm. The essential concept behind this method is that it plots various cost values as k changes. There will be fewer elements in the cluster as the value of K grows.



Q. How do you deal missing values in datasets?

A. Deleting Rows with missing values, Impute missing values for continuous variable, Impute missing values for categorical variable, Using Algorithms that support missing values, Prediction of missing values, Imputation using Deep Learning Library — Datawig, Imputation using Machine Learning libraries like SimpleImputer, KNNImputer,etc.



Q. What is a confusion matrix?

A. A confusion matrix is a method of summarising a classification algorithm's performance. Calculating a confusion matrix can help you understand what your classification model is getting right and where it is going wrong. It gives us: “true positive” for correctly predicted event values, “false positive” for incorrectly predicted event values, “true negative” for correctly predicted no-event values, “false negative” for incorrectly predicted no-event values.

..................

P.S: If you wanna crack interviews of Data Science and become Job Ready, join our" Assignment based Data Science course" today!

Features

✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Frequently Asked Interview QnAs PDF collection
✅ Course completion Certificate
✅ Resume and LinkedIn Review
✅ Job Referrals after course completion
✅ Job Opening Mails, Quiz, Newsletters and more
✅ 1 Month Internship Opportunity
✅ Self-Paced course with Lifetime Access

Get everything at best affordable price

Enroll Now 👉🏻https://www.cloudyml.com/cloudyml-data-science-course/

Happy Learning With CloudyML 💙

Check our blog which contains Data Science Interview Questions collection with companies name that you can go for.

Blog Link 👉🏻https://ai.cloudyml.com/DS_Interview_question

Today's Interview QnAs

Company Name - Accenture

Role: Data Scientist



Q. What is skewed Distribution & uniform distribution?

A. The term "uniform distribution" refers to a situation in which all of the observations in a dataset are evenly distributed over the distribution range. When one side of a graph has more dataset than the other, the circumstance is known as skewed distribution. A uniform distribution, as opposed to a skewed distribution, contains relatively consistent data, with the frequency of each class being similar to the others.



Q. Find the positions of numbers that are multiples of 4 from a series?

A. import pandas as pd

import numpy as np

num_series = pd.Series(np.random.randint(1, 10, 9))

result = np.argwhere(num_series % 4==0)

print(result)



Q. Solve curse of dimensionality?

A. The curse of dimensionality states that as the number of characteristics grows, the error grows as well. It refers to the fact that high-dimensional algorithms are more difficult to build and often have a running duration that is proportional to the dimensions. Dimensionality reduction is a technique for transforming high-dimensional variables into lower-dimensional variables while preserving the variables' particular information. The curse of dimensionality is caused by unstable parameter estimates; thus, regularising these estimates will assist the parameters in making accurate estimations. One of the most common tools for dimension reduction is PCA (Principal Component Analysis).



Q. What are lambda functions? Give example?

A. In Python programming, an anonymous function or a function with no name is known as a Lambda Function. It's a simple and limited function with only one line of code. A Lambda function, like any other function, can have numerous arguments in a single statement. Lambda expressions (or lambda forms) are used to create anonymous functions in Python. You'll achieve this by using the lambda keyword (just as you use def to define normal functions).

Example:

adder = lambda x, y: x + y
print (adder (1, 2))




Q. Get third highest scored student in class table sql?

A. SELECT * FROM student_table ORDER BY marks DESC LIMIT 2,1



Q. Difference between type I and type II errors?

A. In statistics, a Type I error is a false positive conclusion, while a Type II error is a false negative conclusion. Type I vs. Type II Error as an Example Based on your modest symptoms, you decide to get tested for COVID-19. There are two types of errors that could occur: Type I error (false positive): the test result indicates that you have coronavirus when you don't. Type II error (false negative): the test result indicates that you are free of coronavirus, while you are in fact infected.

Happy Learning With CloudyML 💙

Today's Interview QnAs

Company Name - Ford

Role: Data Scientist



Q. What is the the difference between normal distribution, standard normal distribution and uniform distribution?

A. A uniform distribution is one in which all values are equally likely within a range (and impossible beyond that range). A Normal distribution is one in which values cluster around the mean, or average, a outlying values are very unlikely. The standard normal distribution, also called the z-distribution, is a special normal distribution where the mean is 0 and the standard deviation is 1.


Q. Which should be preferred among Gini impurity and Entropy?

A. When using sklearn, the Gini impurity is a good default because it is slightly faster to compute. When they work in opposite directions, Gini impurity isolates the most common class in its own branch of the Tree, whereas entropy produces slightly more balanced Trees.


Q. Explain the time and space complexity of training and testing in the case of a Decision Tree.

A. In the training stage for features (dimensions) in the dataset, we sort the data which takes O(n log n) time following which we traverse the data points to find the right threshold which takes O(n) time. Subsequently, for d dimensions, the total time complexity would be: O(nlogn*d). Moreover, the testing time complexity is O(depth) as we have to traverse from the root to a leaf node of the decision tree i.e., testing space complexity is O(nodes) .


Q. What do you mean by Bootstrap Sample in random forest?

A. Each tree in a random forest learns from a random selection of data points during training. Because the samples are drawn via replacement, or bootstrapping, certain samples will be utilised numerous times in a single tree. The theory is that by training each tree on various samples, even if each tree has a lot of variance with regard to a specific set of training data, the overall variance of the forest will be minimised, but not at the expense of bias.


Q. What are the Limitations of Bagging Trees?

A. The loss of interpretability of a model is one of the drawbacks of bagging. When the right technique is ignored, the resulting model can have a lot of bias. Despite its great accuracy, bagging can be computationally expensive, which may deter its adoption in some situations.

..................

P.S: If you wanna crack interviews of Data Science and become Job Ready, join our" Assignment based Data Science course" today!

Features

✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Frequently Asked Interview QnAs PDF collection
✅ Course completion Certificate
✅ Resume and LinkedIn Review
✅ Job Referrals after course completion
✅ Job Opening Mails, Quiz, Newsletters and more
✅ 1 Month Internship Opportunity
✅ Self-Paced course with Lifetime Access

Get everything at best affordable price

Enroll Now 👉🏻https://www.cloudyml.com/cloudyml-data-science-course/

Happy Learning With CloudyML 💙

#Job_Openings


1. Company: Fractal Analytics

Role: Data Science research and development intern

Experience:0-2 years

Job site: https://ai.cloudyml.com/3Gz3yS8



2. Company: Eduvogue

Role: Data Science Trainer

Experience: Fresher

Job site: https://ai.cloudyml.com/3uySQsy



3. Company: Applied Data Finance

Role: Junior Data Scientist

Experience: 1-2 years

Job site: https://ai.cloudyml.com/3Llec2t



4. Company: Myna Solutions

Role: Data Science intern

Experience : Fresher

Job site: https://ai.cloudyml.com/364Yb0z



5. Company: Toolyt

Role: Product Analyst

Experience: Fresher ( Full time)

Job site: https://ai.cloudyml.com/3uKneQV

Hiring Hackathon by Tredence Analytics

Criteria: BE / B. Tech from a reputed University (eligible for 2019, 2020, 2021 pass outs) .6 to 24 Months of any prior IT experience

Role: Business Analyst

Work Location:
Bangalore, Chennai and Gurgaon

Apply Now👉🏻 https://ai.cloudyml.com/apply2tredence

Note: Candidates will be trained on Analytics/Data Engineering and will become a successful professional in next 12 - 18 months.

Link👉🏻https://ai.cloudyml.com/Register-for-FREE

Today's Interview QnAs

Company Name - TCS

Role: Data Scientist



Q. What is a bias-variance trade-off?

A. If the algorithm is too basic (hypothesis with linear eq. ), it may be prone to errors due to strong bias and low variance. If the algorithms are too sophisticated (hypothesis with a high degree eq. ), the variance and bias may be considerable. The new entries will not fare well in the latter scenario. Trade-off, also known as Bias Variance Trade-off, is something that exists between these two situations. There is a tradeoff between bias and variance because of this tradeoff in complexity. It's impossible for an algorithm to be both more complex and less complex at the same time.


Q. What are the support vectors in SVM?

A. Support vectors are data points that are closer to the hyperplane and influence the position and orientation of the hyperplane. Using these support vectors, we maximize the margin of the classifier. Deleting the support vectors will change the position of the hyperplane. These are the points that help us build our SVM.



Q. Describe different regularization methods, such as L1 and L2 regularization?

A. L1 regularization, also known as L1 norm or Lasso (in regression problems), combats overfitting by shrinking the parameters towards 0. This makes some features obsolete. L2 regularization, or the L2 norm, or Ridge (in regression problems), combats overfitting by forcing weights to be small, but not making them exactly 0.


Q. What is correlation and covariance in statistics?

A. Both covariance and correlation measure the relationship and the dependency between two variables. Covariance indicates the direction of the linear relationship between variables. Correlation measures both the strength and direction of the linear relationship between two variables. Correlation values are standardized. Covariance values are not standardized.

..................

P.S: If you wanna crack interviews of Data Science and become Job Ready, join our" Assignment based Data Science course" today!

Features

✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Frequently Asked Interview QnAs PDF collection
✅ Course completion Certificate
✅ Resume and LinkedIn Review
✅ Job Referrals after course completion
✅ Job Opening Mails, Quiz, Newsletters and more
✅ 1 Month Internship Opportunity
✅ Self-Paced course with Lifetime Access

Get everything at best affordable price

Enroll Now 👉🏻https://www.cloudyml.com/cloudyml-data-science-course/

Happy Learning With CloudyML 💙

Today you gonna get a complete understanding of DEEP LEARNING and its importance.

In today's webinar, we gonna cover the following topics:

✅Why you should learn deep learning ?
✅Deep Learning application in real world
✅Its different sections
✅Best approach to learn deep learning

Join us today and get a clear vision about this highly on demand skill of 21st century.

Webinar is OVER NOW, you can check our Deep learning course

👉🏻https://www.cloudyml.com/deep-learning/

Deep Learning Course Webinar offer is extended till 11:59PM tonight.

Course Link 👉🏻 https://www.cloudyml.com/deep-learning/

Today's Interview QnAs

Company Name - Accenture

Role: Data Scientist



Q. What are the conditions for Overfitting and Underfitting?

A. The model has not overfitted if both the training and test accuracy are close. The model has overfitted if the training result is excellent but the test result is bad. The model is underfit if the training and test accuracy are both low.


Q. How do you clean raw data ?

A. Data cleaning is the process of fixing or removing incorrect, corrupted, incorrectly formatted, duplicate, or incomplete data within a dataset. While the techniques used for data cleaning may vary according to the types of data your company stores, you can follow these basic steps to map out a framework for your organization. Step 1: Remove duplicate or irrelevant observations, Step 2: Fix structural errors, Step 3: Filter unwanted outliers and Step 4: Handle missing data.


Q. Explain any feature selection method?

A. The goal of feature selection in machine learning is to find the best set of features that allows one to build useful models of studied phenomena. Regularization consists of adding a penalty to the different parameters of the machine learning model to reduce the freedom of the model, i.e. to avoid over-fitting. In linear model regularization, the penalty is applied over the coefficients that multiply each of the predictors. From the different types of regularization, Lasso or L1 has the property that is able to shrink some of the coefficients to zero. Therefore, that feature can be removed from the model.


Q. Why random forest better than multiple decided trees?

A. Overfitting, error due to variation, and error due to bias are all important difficulties with decision trees. A Random Forest is a group of decision trees that produce a single, aggregated outcome. Overfitting is less likely when numerous trees are used in the random forest. They're also difficult to comprehend. A decision tree is simple to read and comprehend, whereas a random forest is more difficult to comprehend. A single decision tree isn't very good at forecasting outcomes, but it's simple to set up. More trees result in a more robust model that avoids overfitting. Each tree in the forest must be generated, processed, and analysed. As a result, this is a lengthy process that might take hours or even days.

..................

P.S: If you wanna crack interviews of Data Science and become Job Ready, join our "Assignment based Data Science course" today!

Features

✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Interview QnAs PDF collection
✅ Course completion Certificate
✅ Resume and LinkedIn Review
✅ Job Referrals and Opening Mails
✅ 1 Month Internship Opportunity
✅ Self-Paced course with Lifetime Access

Get everything at best affordable price

Enroll Now 👉🏻https://www.cloudyml.com/cloudyml-data-science-course/

Happy Learning With CloudyML 💙

Today's Interview QnAs

Company Name - Cognizant

Role: Data Scientist



Q. How will you remove duplicates from dataframe based on particular column?

A. An important part of Data analysis is analyzing Duplicate Values and removing them. Pandas drop_duplicates() method helps in removing duplicates from the data frame. Subset parameter takes a column or list of column label. After passing columns, it will consider them only for duplicates.



Q. Explain ROC curve.

A. A Receiver Operator Characteristic (ROC) curve is a graphical plot used to show the diagnostic ability of binary classifiers.A ROC curve is constructed by plotting the true positive rate (TPR) against the false positive rate (FPR). The true positive rate is the proportion of observations that were correctly predicted to be positive out of all positive observations (TP/(TP + FN)). Similarly, the false positive rate is the proportion of observations that are incorrectly predicted to be positive out of all negative observations (FP/(TN + FP)).


Q. What is learning rate in gradient descent?

A. Gradient descent is the popular optimization algorithm used in machine learning to estimate the model parameters. Learning rate, generally represented by the symbol ‘α’, is a hyper-parameter used to control the rate at which an algorithm updates the parameter estimates or learns the values of the parameters.


Q. What do you understand by Recall and Precision?

A. Precision is defined as the fraction of relevant instances among all retrieved instances. Recall, sometimes referred to as ‘sensitivity, is the fraction of retrieved instances among all relevant instances. A perfect classifier has precision and recall both equal to 1.

..................

P.S: If you wanna crack interviews of Data Science and become Job Ready, join our "Assignment based Data Science course" today!

Features

✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Interview QnAs PDF collection
✅ Course completion Certificate
✅ Resume and LinkedIn Review
✅ Job Referrals and Opening Mails
✅ 1 Month Internship Opportunity
✅ Self-Paced course with Lifetime Access

Get everything at best affordable price

Enroll Now 👉🏻https://www.cloudyml.com/cloudyml-data-science-course/

Happy Learning With CloudyML 💙

Today's Interview QnAs

Company Name - Capgemini

Role: Data Scientist


Q. Conditions for Overfitting and Underfitting.

A. If both the training accuracy and test accuracy are close then the model has not overfit. If the training result is very good and the test result is poor then the model has overfitted. If the training accuracy and test accuracy is low then the model has underfit.


Q. What does it mean when the p-values are high and low?

A. The p-value is a number between 0 and 1 and interpreted in the following way: A small p-value (typically ≤ 0.05) indicates strong evidence against the null hypothesis, so you reject the null hypothesis.
A large p-value (> 0.05) indicates weak evidence against the null hypothesis, so you fail to reject the null hypothesis. p-values very close to the cutoff (0.05) are considered to be marginal (could go either way).



Q. What are the differences between correlation and covariance?

A. Covariance is a measure to indicate the extent to which two random variables change in tandem.Correlation is a measure used to represent how strongly two random variables are related to each other.Covariance is nothing but a measure of correlation.Correlation refers to the scaled form of covariance.


Q. How to create a sparse Matrix in Python?

A. Python’s SciPy gives tools for creating sparse matrices using multiple data structures, as well as tools for converting a dense matrix to a sparse matrix. The function csr_matrix() is used to create a sparse matrix of compressed sparse row format whereas csc_matrix() is used to create a sparse matrix of compressed sparse column format.

..................

P.S: If you wanna crack interviews of Data Science and become Job Ready, join our "Assignment based Data Science course" today!

Features

✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Interview QnAs PDF collection
✅ Course completion Certificate
✅ Resume and LinkedIn Review
✅ Job Referrals and Opening Mails
✅ 1 Month Internship Opportunity
✅ Self-Paced course with Lifetime Access

Get everything at best affordable price

Enroll Now 👉🏻https://www.cloudyml.com/cloudyml-data-science-course/

Happy Learning With CloudyML 💙

Dear All, we are here with a super discount offer for you.

AI for All - Data Science Course = ₹1249
Demystify Deep Learning = ₹3499
SQL For Data Science = ₹749

All of our courses are on discount and if you buy this combo package, you can get all these courses total worth price Rs. 5497 at extra discounted Combo Price of Rs. 3999.

Enroll Now and get this super deadly discount.

Combo Link👉🏻 https://ai.cloudyml.com/combo-package

This Offer is only valid for tonight.