Precision & Recall - Intern Gill needs help! #machinelearning #datascience

Accuracy is not always the best metric to evaluate imbalanced data. ❎ ❎
- Especially in classification models like Loan Prediction. 💵 💵
- Accuracy basically is the number of correct predictions out of total predictions.🔍 🔎
- In these cases, Using accuracy here is like using a telescope to examine microorganisms.🔭.🔭

What do you think is the right parameter?❓❓
- Precision and Recall are better metrics to evaluate both classes in classification models. ✅ ✅
- Precision is of all the loans that we labeled as 'Bad', how many were actually 'Bad'? 🔬🔬
- Recall is of all the 'Bad Loans' that truly exist, how many did you correctly label as 'Bad'. 📏📏

🔥🔥Sometimes selection of accurate metrics to see how your model works perfectly is extremely important!💬 ✅

🔥 Free Learning Resources

#evaluationmetric #accuracy #precision #recall #confusionmatrix #mlshorts #datashorts #ExplainedIn60

[BheezHandle]

I suggest F1 score, since it uses both Recall and Precision.

[gouravnandy]

Its recall, which will define how many bad loans were actually labelled as bad . If we use, precision we are looking into the good loans that are predicted as bad. This might affect the business.

[classicemmaeasy]

Recall will give him out of all his data, how many of them are right, and F1 for it's performance

[yoyovatsa]

AAyla, Nice video. This guy even sounds like tendulkar

[pavithranarayanan]

Recall would be correct here because we need to be able to correctly identify the bad loans

[tejaswisaituraga]

KS is better if you don’t want to have any cutoff values

[ajaiar]

Recall, we need to reduce the false negatives

[anubhavrathi-un]

Recall. Because as a bank, I would not want make sure that I identify as much bad loans as possible. Otherwise, I risk losing money.

[muhammadkawthar]

We have to choose Recall as a good metric in this situation

[bamise]

recall. because we want to be able to catch the bad loans.

[Analyticsvidhya]

Accuracy is not always the best metric to evaluate imbalanced data. ❎ ❎
- Especially in classification models like Loan Prediction. 💵 💵
- Accuracy basically is the number of correct predictions out of total predictions.🔍 🔎
- In these cases, Using accuracy here is like using a telescope to examine microorganisms.🔭.🔭

What do you think is the right parameter?❓❓
- Precision and Recall are better metrics to evaluate both classes in classification models. ✅ ✅
- Precision is of all the loans that we labeled as 'Bad', how many were actually 'Bad'? 🔬🔬
- Recall is of all the 'Bad Loans' that truly exist, how many did you correctly label as 'Bad'. 📏📏

🔥🔥Sometimes selection of accurate metrics to see how your model works perfectly is extremely important!💬 ✅


🔥 Subscribe to our channel now: bit.ly/42tER5t