Today's Interview QnAs
Company Name - IBM
Role - Data Scientist
Q1. What is P-value?
Ans. The p-value is the probability that the null hypothesis is true. (1 – the p-value) is the probability that the alternative hypothesis is true. A low p-value shows that the results are replicable. A low p-value shows that the effect is large or that the result is of major theoretical, clinical or practical importance.
Q2. Forward and Backward selection? It's working?
Ans. Forward Selection chooses a subset of the predictor variables for the final model. Unlike forward stepwise selection, it begins with the full least squares model containing all p predictors, and then iteratively removes the least useful predictor, one-at-a-time. forward selection starts with a null model (with no predictors) and proceeds to add variables one at a time, and so unlike backward selection, it DOES NOT have to consider the full model (which includes all the predictors).
Q3. Fix multicollinearity?
Ans. To remove multicollinearities, we can do two things.
1. We can create new features
2. remove those features from our data.
Q4. Difference between Gradient boosting and random forest?
Ans. The two main differences are: How trees are built: random forests builds each tree independently while gradient boosting builds one tree at a time. Combining results: random forests combine results at the end of the process (by averaging or "majority rules") while gradient boosting combines results along the way.
Q5. Precision and Recall? How they are related to ROC curve?
Ans. The recall is the ratio of the relevant results returned by the search engine to the total number of the relevant results that could have been returned. The precision is the proportion of relevant results in the list of all returned search results. When dealing with highly skewed datasets, Precision-Recall (PR) curves in give a more informative picture of an algorithm's performance. We show that a deep connection exists between ROC space and PR space, such that a curve dominates in ROC space if and only if it dominates in PR space.
P.S.: If you wanna learn Python and become a Job-Ready Data Scientist from scratch, check out our Assignments based Complete Data Science course
✅Hands on Guided assignments
✅Capstone End-to-end Projects
✅1-1 Skype Live Doubt Clearance
✅Frequently Asked Interview QnAs PDF collection
✅Course completion Certificate
✅Course completion Certificate
It is a Self-Paced course with lifetime access.
👉🏻Check the Course link in our Telegram Description
We are offering 70% Discount till 31st Dec.
Happy Learning With CloudyML 💙
Company Name - IBM
Role - Data Scientist
Q1. What is P-value?
Ans. The p-value is the probability that the null hypothesis is true. (1 – the p-value) is the probability that the alternative hypothesis is true. A low p-value shows that the results are replicable. A low p-value shows that the effect is large or that the result is of major theoretical, clinical or practical importance.
Q2. Forward and Backward selection? It's working?
Ans. Forward Selection chooses a subset of the predictor variables for the final model. Unlike forward stepwise selection, it begins with the full least squares model containing all p predictors, and then iteratively removes the least useful predictor, one-at-a-time. forward selection starts with a null model (with no predictors) and proceeds to add variables one at a time, and so unlike backward selection, it DOES NOT have to consider the full model (which includes all the predictors).
Q3. Fix multicollinearity?
Ans. To remove multicollinearities, we can do two things.
1. We can create new features
2. remove those features from our data.
Q4. Difference between Gradient boosting and random forest?
Ans. The two main differences are: How trees are built: random forests builds each tree independently while gradient boosting builds one tree at a time. Combining results: random forests combine results at the end of the process (by averaging or "majority rules") while gradient boosting combines results along the way.
Q5. Precision and Recall? How they are related to ROC curve?
Ans. The recall is the ratio of the relevant results returned by the search engine to the total number of the relevant results that could have been returned. The precision is the proportion of relevant results in the list of all returned search results. When dealing with highly skewed datasets, Precision-Recall (PR) curves in give a more informative picture of an algorithm's performance. We show that a deep connection exists between ROC space and PR space, such that a curve dominates in ROC space if and only if it dominates in PR space.
P.S.: If you wanna learn Python and become a Job-Ready Data Scientist from scratch, check out our Assignments based Complete Data Science course
✅Hands on Guided assignments
✅Capstone End-to-end Projects
✅1-1 Skype Live Doubt Clearance
✅Frequently Asked Interview QnAs PDF collection
✅Course completion Certificate
✅Course completion Certificate
It is a Self-Paced course with lifetime access.
👉🏻Check the Course link in our Telegram Description
We are offering 70% Discount till 31st Dec.
Happy Learning With CloudyML 💙
No matter what you do now, if you wanna be a Data Scientist, you can do it.
Learn everything from scratch.
This course covers Python, SQL, Machine Learning, Statistics, Model deployment, AWS Sagemaker and capstone end-to-end projects
Enroll Now 👉🏻https://bit.ly/Switch-to-Data-at-999
So long running offer will end on 31st, Dec 11:59PM
Note: It is a Self-Paced course with lifetime access, you can learn according to your time availability.
Learn everything from scratch.
This course covers Python, SQL, Machine Learning, Statistics, Model deployment, AWS Sagemaker and capstone end-to-end projects
Enroll Now 👉🏻https://bit.ly/Switch-to-Data-at-999
So long running offer will end on 31st, Dec 11:59PM
Note: It is a Self-Paced course with lifetime access, you can learn according to your time availability.
Our so long running Data Science course offer gonna end today.
It's your last chance to learn from scratch and become a Job-Ready Data Scientist at just 999/-
Enroll Now👉🏻https://bit.ly/Switch-to-Data-at-999
Note: Extended offer for one more day. Offer Valid till 1st Jan, 2022
It's your last chance to learn from scratch and become a Job-Ready Data Scientist at just 999/-
Enroll Now👉🏻https://bit.ly/Switch-to-Data-at-999
Note: Extended offer for one more day. Offer Valid till 1st Jan, 2022
Today's Interview Quest
Company Name - Cognizant
Role - Data Scientist
Q. Describe how Gradient Boosting works.
A. Gradient boosting is a type of machine learning boosting. It relies on the intuition that the best possible next model, when combined with previous models, minimizes the overall prediction error. If a small change in the prediction for a case causes no change in error, then next target outcome of the case is zero. Gradient boosting produces a prediction model in the form of an ensemble of weak prediction models, typically decision trees.
Q. Describe the decision tree model.
A. Decision Trees are a type of Supervised Machine Learning where the data is continuously split according to a certain parameter. The leaves are the decisions or the final outcomes. A decision tree is a machine learning algorithm that partitions the data into subsets.
Q. What is a neural network?
A. Neural networks are a set of algorithms, modeled loosely after the human brain, that are designed to recognize patterns. They interpret sensory data through a kind of machine perception, labeling or clustering raw input. They, also known as Artificial Neural Networks, are the subset of Deep Learning.
Q. Explain the Bias-Variance Tradeoff
A. The bias–variance tradeoff is the property of a model that the variance of the parameter estimated across samples can be reduced by increasing the bias in the estimated parameters.
Q. What’s the difference between L1 and L2 regularization?
A. The main intuitive difference between the L1 and L2 regularization is that L1 regularization tries to estimate the median of the data while the L2 regularization tries to estimate the mean of the data to avoid overfitting. That value will also be the median of the data distribution mathematically.
P.S.: If you wanna learn Python and become a Job-Ready Data Scientist from scratch, check out our Assignments based Complete Data Science course
✅Hands on Guided assignments
✅Capstone End-to-end Projects
✅1-1 Skype Live Doubt Clearance
✅Frequently Asked Interview QnAs PDF collection
✅Course completion Certificate
✅Job Referrals
It is a Self-Paced course with lifetime access.
👉🏻Check the Course link in our Telegram Description
We have extended the 70% discount offer till 1st Jan, 2022.
Happy Learning With CloudyML 💙
Company Name - Cognizant
Role - Data Scientist
Q. Describe how Gradient Boosting works.
A. Gradient boosting is a type of machine learning boosting. It relies on the intuition that the best possible next model, when combined with previous models, minimizes the overall prediction error. If a small change in the prediction for a case causes no change in error, then next target outcome of the case is zero. Gradient boosting produces a prediction model in the form of an ensemble of weak prediction models, typically decision trees.
Q. Describe the decision tree model.
A. Decision Trees are a type of Supervised Machine Learning where the data is continuously split according to a certain parameter. The leaves are the decisions or the final outcomes. A decision tree is a machine learning algorithm that partitions the data into subsets.
Q. What is a neural network?
A. Neural networks are a set of algorithms, modeled loosely after the human brain, that are designed to recognize patterns. They interpret sensory data through a kind of machine perception, labeling or clustering raw input. They, also known as Artificial Neural Networks, are the subset of Deep Learning.
Q. Explain the Bias-Variance Tradeoff
A. The bias–variance tradeoff is the property of a model that the variance of the parameter estimated across samples can be reduced by increasing the bias in the estimated parameters.
Q. What’s the difference between L1 and L2 regularization?
A. The main intuitive difference between the L1 and L2 regularization is that L1 regularization tries to estimate the median of the data while the L2 regularization tries to estimate the mean of the data to avoid overfitting. That value will also be the median of the data distribution mathematically.
P.S.: If you wanna learn Python and become a Job-Ready Data Scientist from scratch, check out our Assignments based Complete Data Science course
✅Hands on Guided assignments
✅Capstone End-to-end Projects
✅1-1 Skype Live Doubt Clearance
✅Frequently Asked Interview QnAs PDF collection
✅Course completion Certificate
✅Job Referrals
It is a Self-Paced course with lifetime access.
👉🏻Check the Course link in our Telegram Description
We have extended the 70% discount offer till 1st Jan, 2022.
Happy Learning With CloudyML 💙
Today's Interview QnAs
Company Name - JP Morgan
Role: Data Scientist
Q. Explain How a System Can Play a Game of Chess Using Reinforcement Learning?
A. The chess-playing reinforcement learning model starts with a clean slate, and is only given the basic rules of moving the pieces and the ultimate goal, which is to drive the opponent into check mate. At the beginning, the AI knows nothing about the tactics of the game and makes random moves.
But after playing against itself thousands and millions of times, it starts to develop a statistical model of what sequences of moves are likely to win each situation.
With reinforced learning, It will make a move (decision), check if it’s the right move (feedback), and keep the outcomes in memory for the next step it takes (learning). There is a reward for every correct decision the system takes and punishment for the wrong one.
Q. Difference between Gini Impurity and Entropy in a Decision Tree?
A. The criteria typically used to decide which feature to split on are the Gini index and information entropy. Both of these measures are pretty similar numerically.
*The Gini index is used by the CART (classification and regression tree) algorithm, whereas information gain via entropy reduction is used by algorithms like C4.5.
*Entropy takes slightly more computation time than Gini Index because of the log calculation, maybe that's why Gini Index has become the default option for many ML algorithms.
*The range of Entropy lies in between 0 to 1 and the range of Gini Impurity lies in between 0 to 0.5.
*Gini measurement is the probability of a random sample being classified incorrectly if we randomly pick a label according to the distribution in a branch.Whereas entropy is a measure of information that indicates the disorder of the features with the target.
Q. Suppose you found that your model is suffering from low bias and high variance. How would you tackle it?
A. Low bias and High variance is the classic term of overfitting which means your model is performing very good in the training dataset, but not good in the test data or unseen data
regularization, creating a random forest or any bagging technique work as well. Rationale of bagging is as simple as following - if X1 , ..., Xn are independently distributed with variance v , variance of their average becomes v/n and thereby reducing the variance.
Q. Difference between stochastic gradient descent (SGD) and gradient descent (GD)?
A. In both gradient descent (GD) and stochastic gradient descent (SGD), you update a set of parameters in an iterative manner to minimize an error function.
While in GD, you have to run through ALL the samples in your training set to do a single update for a parameter in a particular iteration, in SGD, on the other hand, you use ONLY ONE or SUBSET of training sample from your training set to do the update for a parameter in a particular iteration. If you use SUBSET, it is called Minibatch Stochastic gradient Descent.
Thus, if the number of training samples are large, in fact very large, then using gradient descent may take too long because in every iteration when you are updating the values of the parameters, you are running through the complete training set. On the other hand, using SGD will be faster because you use only one training sample and it starts improving itself right away from the first sample.
SGD often converges much faster compared to GD but the error function is not as well minimized as in the case of GD. Often in most cases, the close approximation that you get in SGD for the parameter values are enough because they reach the optimal values and keep oscillating there.
Q. What are crosstab in python?
A. Crosstab function builds a cross-tabulation table that can show the frequency with which certain groups of data appear.
🔴P.S.- Learn Data Science from zero. Get complete hands on practical learning experience with 1-1 Skype live Doubt Clearance Support in our Data Science course.
👉🏻Link in Telegram Description
Happy Learning With CloudyML 💙
Company Name - JP Morgan
Role: Data Scientist
Q. Explain How a System Can Play a Game of Chess Using Reinforcement Learning?
A. The chess-playing reinforcement learning model starts with a clean slate, and is only given the basic rules of moving the pieces and the ultimate goal, which is to drive the opponent into check mate. At the beginning, the AI knows nothing about the tactics of the game and makes random moves.
But after playing against itself thousands and millions of times, it starts to develop a statistical model of what sequences of moves are likely to win each situation.
With reinforced learning, It will make a move (decision), check if it’s the right move (feedback), and keep the outcomes in memory for the next step it takes (learning). There is a reward for every correct decision the system takes and punishment for the wrong one.
Q. Difference between Gini Impurity and Entropy in a Decision Tree?
A. The criteria typically used to decide which feature to split on are the Gini index and information entropy. Both of these measures are pretty similar numerically.
*The Gini index is used by the CART (classification and regression tree) algorithm, whereas information gain via entropy reduction is used by algorithms like C4.5.
*Entropy takes slightly more computation time than Gini Index because of the log calculation, maybe that's why Gini Index has become the default option for many ML algorithms.
*The range of Entropy lies in between 0 to 1 and the range of Gini Impurity lies in between 0 to 0.5.
*Gini measurement is the probability of a random sample being classified incorrectly if we randomly pick a label according to the distribution in a branch.Whereas entropy is a measure of information that indicates the disorder of the features with the target.
Q. Suppose you found that your model is suffering from low bias and high variance. How would you tackle it?
A. Low bias and High variance is the classic term of overfitting which means your model is performing very good in the training dataset, but not good in the test data or unseen data
regularization, creating a random forest or any bagging technique work as well. Rationale of bagging is as simple as following - if X1 , ..., Xn are independently distributed with variance v , variance of their average becomes v/n and thereby reducing the variance.
Q. Difference between stochastic gradient descent (SGD) and gradient descent (GD)?
A. In both gradient descent (GD) and stochastic gradient descent (SGD), you update a set of parameters in an iterative manner to minimize an error function.
While in GD, you have to run through ALL the samples in your training set to do a single update for a parameter in a particular iteration, in SGD, on the other hand, you use ONLY ONE or SUBSET of training sample from your training set to do the update for a parameter in a particular iteration. If you use SUBSET, it is called Minibatch Stochastic gradient Descent.
Thus, if the number of training samples are large, in fact very large, then using gradient descent may take too long because in every iteration when you are updating the values of the parameters, you are running through the complete training set. On the other hand, using SGD will be faster because you use only one training sample and it starts improving itself right away from the first sample.
SGD often converges much faster compared to GD but the error function is not as well minimized as in the case of GD. Often in most cases, the close approximation that you get in SGD for the parameter values are enough because they reach the optimal values and keep oscillating there.
Q. What are crosstab in python?
A. Crosstab function builds a cross-tabulation table that can show the frequency with which certain groups of data appear.
🔴P.S.- Learn Data Science from zero. Get complete hands on practical learning experience with 1-1 Skype live Doubt Clearance Support in our Data Science course.
👉🏻Link in Telegram Description
Happy Learning With CloudyML 💙
#Job #Openings
Role - Data Science Developer
Company - Quickxpert Infotech
Exp - 0-5 yr
Link - https://bit.ly/32H5olW
Role - Data Scientist (Python)
Company - Spark Wave
Exp - 0-5 yr
Link - https://bit.ly/3pOKLgE
Role - AI/ML Architect
Company - Winrar Engineering Solutions
Exp - 0-4 yr
Link - https://bit.ly/3eLbg03
Role - Financial Data Analyst
company - Moody's
Exp - 0-2 yr
Link - https://bit.ly/34iH5vd
Role - Business Analyst
Company - O2b Technologies Private Ltd
Exp - 0-1 yr
Link - https://bit.ly/3JBC2WT
All The Best💙
Role - Data Science Developer
Company - Quickxpert Infotech
Exp - 0-5 yr
Link - https://bit.ly/32H5olW
Role - Data Scientist (Python)
Company - Spark Wave
Exp - 0-5 yr
Link - https://bit.ly/3pOKLgE
Role - AI/ML Architect
Company - Winrar Engineering Solutions
Exp - 0-4 yr
Link - https://bit.ly/3eLbg03
Role - Financial Data Analyst
company - Moody's
Exp - 0-2 yr
Link - https://bit.ly/34iH5vd
Role - Business Analyst
Company - O2b Technologies Private Ltd
Exp - 0-1 yr
Link - https://bit.ly/3JBC2WT
All The Best💙
Media is too big
VIEW IN TELEGRAM
You have the best chance to learn Data Science from scratch with job Referrals after Course completion just at 999/- Offer Price.
Enroll Now👉🏻https://www.cloudyml.com/cloudyml-data-science-course/
Enroll Now👉🏻https://www.cloudyml.com/cloudyml-data-science-course/
🆒1
Anonymous Quiz
4%
Lists
10%
Dictionary
8%
Tuples
77%
Class
Today's Interview QnAs
Company Name - Honeywell
Role: Data Scientist
Q. Significance of Gamma and Regularization in SVM?
A. Gamma and C(behaves as a regularization parameter in svm) are the two parameters which is very important in SVM. The behavior of the model is very sensitive to the gamma parameter. If gamma is too large, the radius of the area of influence of the support vectors only includes the support vector itself and no amount of regularization with C will be able to prevent overfitting. When gamma is very small, the model is too constrained and cannot capture the complexity or “shape” of the data. The resulting model will behave similarly to a linear model with a set of hyperplanes that separate the centers of high density of any pair of two classes.
Q. Is ARIMA model a good fit for every time series problem?
A. No we cannot use it for every problem although it's a popular method in time series. ARIMA models are applied in some cases where data show evidence of non-stationarity in the sense of mean (but not variance/autocovariance), where an initial differencing step (corresponding to the "integrated" part of the model) can be applied one or more times to eliminate the non-stationarity of the mean function.
Q. Difference between stochastic gradient descent (SGD) and gradient descent (GD)?
A. In both gradient descent (GD) and stochastic gradient descent (SGD), you update a set of parameters in an iterative manner to minimize an error function.
While in GD, you have to run through ALL the samples in your training set to do a single update for a parameter in a particular iteration, in SGD, on the other hand, you use ONLY ONE or SUBSET of training sample from your training set to do the update for a parameter in a particular iteration. If you use SUBSET, it is called Minibatch Stochastic gradient Descent.
Thus, if the number of training samples are large, in fact very large, then using gradient descent may take too long because in every iteration when you are updating the values of the parameters, you are running through the complete training set. On the other hand, using SGD will be faster because you use only one training sample and it starts improving itself right away from the first sample.
SGD often converges much faster compared to GD but the error function is not as well minimized as in the case of GD. Often in most cases, the close approximation that you get in SGD for the parameter values are enough because they reach the optimal values and keep oscillating there.
P.S.: Learn Data Science from Zero and become Interview ready at best affordable price with personal doubt clearance support
✅Hands on Guided assignments
✅Capstone End-to-end Projects
✅1-1 Skype Live Doubt Clearance
✅Frequently Asked Interview QnAs PDF collection
✅Course completion Certificate
It is a Self-Paced course with lifetime access.
Enroll Now👉🏻https://bit.ly/Switch-to-Data-at-999
Happy Learning With CloudyML 💙
Company Name - Honeywell
Role: Data Scientist
Q. Significance of Gamma and Regularization in SVM?
A. Gamma and C(behaves as a regularization parameter in svm) are the two parameters which is very important in SVM. The behavior of the model is very sensitive to the gamma parameter. If gamma is too large, the radius of the area of influence of the support vectors only includes the support vector itself and no amount of regularization with C will be able to prevent overfitting. When gamma is very small, the model is too constrained and cannot capture the complexity or “shape” of the data. The resulting model will behave similarly to a linear model with a set of hyperplanes that separate the centers of high density of any pair of two classes.
Q. Is ARIMA model a good fit for every time series problem?
A. No we cannot use it for every problem although it's a popular method in time series. ARIMA models are applied in some cases where data show evidence of non-stationarity in the sense of mean (but not variance/autocovariance), where an initial differencing step (corresponding to the "integrated" part of the model) can be applied one or more times to eliminate the non-stationarity of the mean function.
Q. Difference between stochastic gradient descent (SGD) and gradient descent (GD)?
A. In both gradient descent (GD) and stochastic gradient descent (SGD), you update a set of parameters in an iterative manner to minimize an error function.
While in GD, you have to run through ALL the samples in your training set to do a single update for a parameter in a particular iteration, in SGD, on the other hand, you use ONLY ONE or SUBSET of training sample from your training set to do the update for a parameter in a particular iteration. If you use SUBSET, it is called Minibatch Stochastic gradient Descent.
Thus, if the number of training samples are large, in fact very large, then using gradient descent may take too long because in every iteration when you are updating the values of the parameters, you are running through the complete training set. On the other hand, using SGD will be faster because you use only one training sample and it starts improving itself right away from the first sample.
SGD often converges much faster compared to GD but the error function is not as well minimized as in the case of GD. Often in most cases, the close approximation that you get in SGD for the parameter values are enough because they reach the optimal values and keep oscillating there.
P.S.: Learn Data Science from Zero and become Interview ready at best affordable price with personal doubt clearance support
✅Hands on Guided assignments
✅Capstone End-to-end Projects
✅1-1 Skype Live Doubt Clearance
✅Frequently Asked Interview QnAs PDF collection
✅Course completion Certificate
It is a Self-Paced course with lifetime access.
Enroll Now👉🏻https://bit.ly/Switch-to-Data-at-999
Happy Learning With CloudyML 💙
𝐓𝐨𝐝𝐚𝐲'𝐬 𝐈𝐧𝐭𝐞𝐫𝐯𝐢𝐞𝐰 𝐐𝐮𝐞𝐬𝐭 𝐍 𝐀𝐧𝐬
𝐂𝐨𝐦𝐩𝐚𝐧𝐲 𝐍𝐚𝐦𝐞 - 𝐂𝐚𝐩𝐠𝐞𝐦𝐢𝐧𝐢
𝐑𝐨𝐥𝐞 - 𝐃𝐚𝐭𝐚 𝐒𝐜𝐢𝐞𝐧𝐭𝐢𝐬𝐭
--------------------------------------------
𝐐1. 𝐩-𝐯𝐚𝐥𝐮𝐞?
𝐀ns. p-value is a measure of the probability that an observed difference could have occurred just by random chance. The lower the p-value, the greater the statistical significance of the observed difference. P-value can be used as an alternative to or in addition to pre-selected confidence levels for hypothesis testing.
𝐐2. 𝐈𝐧𝐭𝐞𝐫𝐩𝐨𝐥𝐚𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐄𝐱𝐭𝐫𝐚𝐩𝐨𝐥𝐚𝐭𝐢𝐨𝐧?
𝐀ns. Interpolation is the process of calculating the unknown value from known given values whereas extrapolation is the process of calculating unknown values beyond the given data points.
𝐐3. 𝐔𝐧𝐢𝐟𝐨𝐫𝐦𝐞𝐝 𝐃𝐢𝐬𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧 & 𝐧𝐨𝐫𝐦𝐚𝐥 𝐝𝐢𝐬𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧?
𝐀ns. The normal distribution is bell-shaped, which means value near the center of the distribution are more likely to occur as opposed to values on the tails of the distribution. The uniform distribution is rectangular-shaped, which means every value in the distribution is equally likely to occur.
𝐐4. 𝐑𝐞𝐜𝐨𝐦𝐦𝐞𝐧𝐝𝐞𝐫 𝐒𝐲𝐬𝐭𝐞𝐦𝐬?
𝐀ns. The recommender system mainly deals with the likes and dislikes of the users. Its major objective is to recommend an item to a user which has a high chance of liking or is in need of a particular user based on his previous purchases. It is like having a personalized team who can understand our likes and dislikes and help us in making the decisions regarding a particular item without being biased by any means by making use of a large amount of data in the repositories which are generated day by day.
𝐐5. 𝐉𝐎𝐈𝐍 𝐟𝐮𝐧𝐜𝐭𝐢𝐨𝐧 𝐢𝐧 𝐒𝐐𝐋
𝐀ns. The SQL Joins clause is used to combine records from two or more tables in a database.
𝐐6. 𝐒𝐪𝐮𝐚𝐫𝐞𝐝 𝐞𝐫𝐫𝐨𝐫 𝐚𝐧𝐝 𝐚𝐛𝐬𝐨𝐥𝐮𝐭𝐞 𝐞𝐫𝐫𝐨𝐫?
𝐀ns. mean squared error (MSE), and mean absolute error (MAE) are used to evaluate the regression problem's accuracy. The squared error is everywhere differentiable, while the absolute error is not (its derivative is undefined at 0). This makes the squared error more amenable to the techniques of mathematical optimization.
-----------------------
P.S.: If you wanna crack interviews of Data Scientist, Data Analyst, ML Engineer, Business Analyst and Research analyst, join our Assignments based "AI For ALL" Course.
✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Frequently Asked Interview QnAs PDF collection
✅ Course completion Certificate
It is a Self-Paced course with lifetime access.
👉🏻Check the Course link in our Telegram Description
Happy Learning With CloudyML 💙
𝐂𝐨𝐦𝐩𝐚𝐧𝐲 𝐍𝐚𝐦𝐞 - 𝐂𝐚𝐩𝐠𝐞𝐦𝐢𝐧𝐢
𝐑𝐨𝐥𝐞 - 𝐃𝐚𝐭𝐚 𝐒𝐜𝐢𝐞𝐧𝐭𝐢𝐬𝐭
--------------------------------------------
𝐐1. 𝐩-𝐯𝐚𝐥𝐮𝐞?
𝐀ns. p-value is a measure of the probability that an observed difference could have occurred just by random chance. The lower the p-value, the greater the statistical significance of the observed difference. P-value can be used as an alternative to or in addition to pre-selected confidence levels for hypothesis testing.
𝐐2. 𝐈𝐧𝐭𝐞𝐫𝐩𝐨𝐥𝐚𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐄𝐱𝐭𝐫𝐚𝐩𝐨𝐥𝐚𝐭𝐢𝐨𝐧?
𝐀ns. Interpolation is the process of calculating the unknown value from known given values whereas extrapolation is the process of calculating unknown values beyond the given data points.
𝐐3. 𝐔𝐧𝐢𝐟𝐨𝐫𝐦𝐞𝐝 𝐃𝐢𝐬𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧 & 𝐧𝐨𝐫𝐦𝐚𝐥 𝐝𝐢𝐬𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧?
𝐀ns. The normal distribution is bell-shaped, which means value near the center of the distribution are more likely to occur as opposed to values on the tails of the distribution. The uniform distribution is rectangular-shaped, which means every value in the distribution is equally likely to occur.
𝐐4. 𝐑𝐞𝐜𝐨𝐦𝐦𝐞𝐧𝐝𝐞𝐫 𝐒𝐲𝐬𝐭𝐞𝐦𝐬?
𝐀ns. The recommender system mainly deals with the likes and dislikes of the users. Its major objective is to recommend an item to a user which has a high chance of liking or is in need of a particular user based on his previous purchases. It is like having a personalized team who can understand our likes and dislikes and help us in making the decisions regarding a particular item without being biased by any means by making use of a large amount of data in the repositories which are generated day by day.
𝐐5. 𝐉𝐎𝐈𝐍 𝐟𝐮𝐧𝐜𝐭𝐢𝐨𝐧 𝐢𝐧 𝐒𝐐𝐋
𝐀ns. The SQL Joins clause is used to combine records from two or more tables in a database.
𝐐6. 𝐒𝐪𝐮𝐚𝐫𝐞𝐝 𝐞𝐫𝐫𝐨𝐫 𝐚𝐧𝐝 𝐚𝐛𝐬𝐨𝐥𝐮𝐭𝐞 𝐞𝐫𝐫𝐨𝐫?
𝐀ns. mean squared error (MSE), and mean absolute error (MAE) are used to evaluate the regression problem's accuracy. The squared error is everywhere differentiable, while the absolute error is not (its derivative is undefined at 0). This makes the squared error more amenable to the techniques of mathematical optimization.
-----------------------
P.S.: If you wanna crack interviews of Data Scientist, Data Analyst, ML Engineer, Business Analyst and Research analyst, join our Assignments based "AI For ALL" Course.
✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Frequently Asked Interview QnAs PDF collection
✅ Course completion Certificate
It is a Self-Paced course with lifetime access.
👉🏻Check the Course link in our Telegram Description
Happy Learning With CloudyML 💙
Anonymous Quiz
75%
List
9%
Dictionary
9%
Tuple
7%
Array
Today's Interview QnAs
Company Name - HCL
Role: Data Scientist
Q. What is the loss function SVM tries to minimize?
A. Although there is no "loss function" for hard-margin SVMs, the loss does exist when solving soft-margin SVMs. The hinge loss is a loss function used in machine learning to train classifiers. For "maximum-margin" classification, the hinge loss is utilised, most notably for support vector machines (SVMs).
Q. Detect heteroscedasticity in simple linear regression?
A. Heteroscedasticity refers to the situation where the spread of the residuals changes in a systematic way over the range of observed values. A fitted value vs. residual plot is the simplest technique to determine heteroscedasticity. The "cone" form is a clear marker of heteroscedasticity if the residuals become significantly more spread out as the fitted values get greater. The Breusch-Pagan test is a more formal, mathematical method of determining heteroscedasticity.
Q. Explain ANOVA?
A. The analysis of variance (ANOVA) is a statistical technique for determining if the means of two or more groups differ significantly. One-way ANOVA, two-way ANOVA, and multivariate ANOVA are the three types. An ANOVA's null hypothesis is that there is no significant difference between the groups. The alternative hypothesis proposes that the groups have at least one substantial difference. The null hypothesis is rejected and the alternative hypothesis is validated if the p-value associated with the F is less than.05. If the null hypothesis is rejected, one concludes that the means of all the groups are not equal.
Q. Determine no. of neighbors in KNN?
A. The number of neighbors(K) in KNN is a hyperparameter that must be chosen during model construction. According to research, there is no ideal number of neighbors for all types of data sets. A small number of neighbors is the most flexible fit, resulting in low bias but high variation, whereas a big number of neighbors results in a smoother decision boundary, resulting in reduced variance but higher bias. If the number of classes is even, data scientists usually choose an odd number. You can also test the model's performance by creating it with different values of k and comparing the results. Elbow technique is another option.
Q. What do you mean by central trend?
A. The central trend is a description of a dataset represented by a single value that represents the data distribution's center. The following measurements can be used to describe the central tendency of a dataset. The sum of all values in a dataset divided by the total number of values is the mean. The middle value in an ascending-ordered dataset is called the median. The most often occurring value in a dataset is defined by the mode. Despite the fact that the measures listed above are the most generally employed to describe central tendency, there are others, such as geometric mean, harmonic mean, midrange, and geometric median.
P.S.: If you wanna crack interviews of Data Scientist, Data Analyst, ML Engineer, Business Analyst and Research analyst, join our Assignments based "AI For ALL" Course.
✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Frequently Asked Interview QnAs PDF collection
✅ Job Referrals
✅ Weekly Live Webinar on various topics
✅ Course completion Certificate
It is a Self-Paced course with lifetime access.
Enroll Now👉🏻https://ai.cloudyml.com/Switch-to-Data-Science-at-999
Happy Learning With CloudyML 💙
Company Name - HCL
Role: Data Scientist
Q. What is the loss function SVM tries to minimize?
A. Although there is no "loss function" for hard-margin SVMs, the loss does exist when solving soft-margin SVMs. The hinge loss is a loss function used in machine learning to train classifiers. For "maximum-margin" classification, the hinge loss is utilised, most notably for support vector machines (SVMs).
Q. Detect heteroscedasticity in simple linear regression?
A. Heteroscedasticity refers to the situation where the spread of the residuals changes in a systematic way over the range of observed values. A fitted value vs. residual plot is the simplest technique to determine heteroscedasticity. The "cone" form is a clear marker of heteroscedasticity if the residuals become significantly more spread out as the fitted values get greater. The Breusch-Pagan test is a more formal, mathematical method of determining heteroscedasticity.
Q. Explain ANOVA?
A. The analysis of variance (ANOVA) is a statistical technique for determining if the means of two or more groups differ significantly. One-way ANOVA, two-way ANOVA, and multivariate ANOVA are the three types. An ANOVA's null hypothesis is that there is no significant difference between the groups. The alternative hypothesis proposes that the groups have at least one substantial difference. The null hypothesis is rejected and the alternative hypothesis is validated if the p-value associated with the F is less than.05. If the null hypothesis is rejected, one concludes that the means of all the groups are not equal.
Q. Determine no. of neighbors in KNN?
A. The number of neighbors(K) in KNN is a hyperparameter that must be chosen during model construction. According to research, there is no ideal number of neighbors for all types of data sets. A small number of neighbors is the most flexible fit, resulting in low bias but high variation, whereas a big number of neighbors results in a smoother decision boundary, resulting in reduced variance but higher bias. If the number of classes is even, data scientists usually choose an odd number. You can also test the model's performance by creating it with different values of k and comparing the results. Elbow technique is another option.
Q. What do you mean by central trend?
A. The central trend is a description of a dataset represented by a single value that represents the data distribution's center. The following measurements can be used to describe the central tendency of a dataset. The sum of all values in a dataset divided by the total number of values is the mean. The middle value in an ascending-ordered dataset is called the median. The most often occurring value in a dataset is defined by the mode. Despite the fact that the measures listed above are the most generally employed to describe central tendency, there are others, such as geometric mean, harmonic mean, midrange, and geometric median.
P.S.: If you wanna crack interviews of Data Scientist, Data Analyst, ML Engineer, Business Analyst and Research analyst, join our Assignments based "AI For ALL" Course.
✅ Hands on Guided assignments
✅ Capstone End-to-end Projects
✅ 1-1 Skype Live Doubt Clearance
✅ Frequently Asked Interview QnAs PDF collection
✅ Job Referrals
✅ Weekly Live Webinar on various topics
✅ Course completion Certificate
It is a Self-Paced course with lifetime access.
Enroll Now👉🏻https://ai.cloudyml.com/Switch-to-Data-Science-at-999
Happy Learning With CloudyML 💙
Learn from Scratch and Switch to Data Science from any Non-Coding Background
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
Bonus(You gonna get these only if you enroll today)
🟢Resume Review
🟢LinkedIn Profile Review
🟢Job Referrals after course completion
It is a Self-Paced course with lifetime access.
Don't Miss this opportunity. You Never gonna get all these features at such price.
Enroll Now👉🏻https://ai.cloudyml.com/Data-Science-at-999
Offer Valid till tonight by 11:59PM ! Hurry Up!
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
Bonus(You gonna get these only if you enroll today)
🟢Resume Review
🟢LinkedIn Profile Review
🟢Job Referrals after course completion
It is a Self-Paced course with lifetime access.
Don't Miss this opportunity. You Never gonna get all these features at such price.
Enroll Now👉🏻https://ai.cloudyml.com/Data-Science-at-999
Offer Valid till tonight by 11:59PM ! Hurry Up!
Today's Interview QnAs
Company Name - Ford
Role: Data Scientist
Q. In experimental design, is it necessary to do randomization? If yes, why?
A. Yes, in experimental design, randomization is required. Randomization eliminates biases and ensures that the outcomes are balanced. The sample that is randomly picked is designed to be representative of the population, and it is fairly selected because it is not influenced by the researcher. You can achieve the best cause-effect linkages between the variables by randomizing the experiments.
Q. What is the benefit of weight initialisation in neural networks?
A. The fundamental goal of weight initialization is to prevent exploding or vanishing gradients in layer activation outputs during forward propagation. If either of these issues arises, loss gradients will be either too great or too little, and the network will take longer to converge, if it can at all. If we appropriately set the weights, we will achieve our goal of optimising the loss function in the shortest time possible; otherwise, converging to a minimum via gradient descent will be impossible.
Q. How will you evaluate the performance of a logistic regression model?
A. The confusion matrix can be used to evaluate a logistic regression model. The accuracy, sensitivity, and specificity of the model can be useful indicators of what you want to do with it - focusing on true positives or false negatives. We can also utilize precision and recall to evaluate your model, as well as the f1 score.
Q. How can you make data normal using Box-Cox transformation?
A. A Box Cox transformation turns non-normal dependent variables into normal shapes. The Lambda value specifies the level of data that should be raised to. The Box-Cox power transformation does this by searching from Lambda = -5 to Lambda = +5 until the best value is discovered.
P.S: Learn from Scratch and Switch to Data Science from any Non-Coding Background
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
Bonus(You gonna get these only if you enroll today)
🟢Resume Review
🟢LinkedIn Profile Review
🟢Job Referrals after course completion
It is a Self-Paced course with lifetime access.
Enroll Now👉🏻https://ai.cloudyml.com/Data-Science-at-999
Happy Learning With CloudyML 💙
Company Name - Ford
Role: Data Scientist
Q. In experimental design, is it necessary to do randomization? If yes, why?
A. Yes, in experimental design, randomization is required. Randomization eliminates biases and ensures that the outcomes are balanced. The sample that is randomly picked is designed to be representative of the population, and it is fairly selected because it is not influenced by the researcher. You can achieve the best cause-effect linkages between the variables by randomizing the experiments.
Q. What is the benefit of weight initialisation in neural networks?
A. The fundamental goal of weight initialization is to prevent exploding or vanishing gradients in layer activation outputs during forward propagation. If either of these issues arises, loss gradients will be either too great or too little, and the network will take longer to converge, if it can at all. If we appropriately set the weights, we will achieve our goal of optimising the loss function in the shortest time possible; otherwise, converging to a minimum via gradient descent will be impossible.
Q. How will you evaluate the performance of a logistic regression model?
A. The confusion matrix can be used to evaluate a logistic regression model. The accuracy, sensitivity, and specificity of the model can be useful indicators of what you want to do with it - focusing on true positives or false negatives. We can also utilize precision and recall to evaluate your model, as well as the f1 score.
Q. How can you make data normal using Box-Cox transformation?
A. A Box Cox transformation turns non-normal dependent variables into normal shapes. The Lambda value specifies the level of data that should be raised to. The Box-Cox power transformation does this by searching from Lambda = -5 to Lambda = +5 until the best value is discovered.
P.S: Learn from Scratch and Switch to Data Science from any Non-Coding Background
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
Bonus(You gonna get these only if you enroll today)
🟢Resume Review
🟢LinkedIn Profile Review
🟢Job Referrals after course completion
It is a Self-Paced course with lifetime access.
Enroll Now👉🏻https://ai.cloudyml.com/Data-Science-at-999
Happy Learning With CloudyML 💙
Learn from Scratch and Switch to Data Science from any Non-Coding Background
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
✅ Job openings updates, newsletter
Bonus
🟢Resume Review
🟢LinkedIn Profile Review
🟢Job Referrals after course completion
🟢Live Webinar on various Data Science topics
A Self-Paced course with lifetime access.
🔴So long running offer gonna end Tonight by 11:59PM
Don't Miss this opportunity. You Never gonna get all these features at such price.
Enroll Now👉🏻https://ai.cloudyml.com/Data-Science-at-999
Note: From tomorrow, price will become 3499/- for the same package, so Hurry up !
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
✅ Job openings updates, newsletter
Bonus
🟢Resume Review
🟢LinkedIn Profile Review
🟢Job Referrals after course completion
🟢Live Webinar on various Data Science topics
A Self-Paced course with lifetime access.
🔴So long running offer gonna end Tonight by 11:59PM
Don't Miss this opportunity. You Never gonna get all these features at such price.
Enroll Now👉🏻https://ai.cloudyml.com/Data-Science-at-999
Note: From tomorrow, price will become 3499/- for the same package, so Hurry up !