@@ -76,4 +76,39 @@ triple equals compares both content and data types of LHS & RHS.
7676Difference between Breadth-first search & Depth first search.
7777
7878
79+ ****************************************************************************
80+ Explanation of the past project. What were the features used and how did you determine performance?
81+ 82+ What is the difference between linear regression and logistic regression?
83+ 84+ What is the internal working of logistic regression (LR)?
85+ 86+ What is the loss function of LR?
87+ 88+ Name some hyperparameters used in LR? Why do we use regularization?
89+ 90+ When do we use accuracy as a metric? When should we not use accuracy?
91+ 92+ How do you deal with imbalance data?
93+ 94+ What is SMOTE and how is it different from stratified sampling?
95+ 96+ Watch this video to understand how SMOTE works [ https://www.youtube.com/watch?v=U3X98xZ4_no ]
97+ 98+ What is better 0.51 AUC (Area Under the Curve) or 0.43 F1 score? Which one should you present to a client?
99+ 100+ Watch this video to understand how AUC is interpreted [ https://www.youtube.com/watch?v=mUMd_cKU0VM ]
101+ 102+ What does the ROC AUC value signify?
103+ 104+ Do we only use the threshold of 0.5 or can we use other thresholds in LR? If yes, how do we find them?
105+ 106+ Can I use a sales forecasting model built using pencils data to be used in erasers data?
107+ 108+ How would you compare the performance of two forecasting models?
109+ 110+ What are the different metrics used in regression analysis? Which metric should be used where?
111+ 112+ How do you build a testing pipeline for a data science model? [ https://www.kdnuggets.com/2020/08/unit-test-data-pipeline-thank-yourself-later.html ]
113+ 79114****************************************************************************
0 commit comments