user avatar
user avatar
user avatar
The Great
  • Member for 3 years, 5 months
  • Last seen this week
20 votes
4 answers
1k views

Which comes first - domain expertise or an experimental approach?

9 votes
1 answer
2k views

Why regularization parameter called as lambda in theory and alpha in python?

9 votes
1 answer
436 views

How to calibrate models if we don't have enough data?

5 votes
2 answers
269 views

Interpreting table 1 in clinical research papers

4 votes
3 answers
2k views

Why significant variable doesn't improve model performance?

4 votes
1 answer
345 views

Why and how to match variables using logistic regression?

4 votes
2 answers
580 views

Why use supervised binning on train data if it leaks data?

4 votes
1 answer
145 views

Rule based label - random split vs time-based split

3 votes
1 answer
263 views

What is the use of expected value in machine learning models?

3 votes
1 answer
168 views

Why ML involves trial and error and no overarching rules?

3 votes
1 answer
124 views

Usefulness of KS tests and other similar distribution comparing tests

3 votes
3 answers
109 views

MOOC on Causal analysis - Recommendations for software engineer

3 votes
1 answer
3k views

Understanding the 10:1 events per variable rule

3 votes
3 answers
7k views

How to adjust confounders in Logistic regression?

3 votes
0 answers
42 views

Does low frequent items require data preparation as high freq items?

2 votes
1 answer
228 views

Plotting categorical data along with numerical variable

2 votes
1 answer
150 views

What is noise-tolerant learning?

2 votes
1 answer
74 views

Why does my model produce unrealistic output?

2 votes
1 answer
35 views

How to make a decision - when there is a tie and no human expert

2 votes
1 answer
2k views

how to calculate median follow-up time?

2 votes
1 answer
620 views

how is confounders different from interaction terms?

2 votes
1 answer
159 views

How to select best feature set?

2 votes
2 answers
470 views

When to drop correlated features?

2 votes
1 answer
88 views

How to know if I have enough controllable info for ML modelling?

2 votes
1 answer
241 views

likelihood vs confidence in layman terms

2 votes
1 answer
305 views

Imbalanced classification vs anomaly detection

2 votes
2 answers
73 views

Generating interesting/useful features from existing useless features

2 votes
1 answer
135 views

How to interpret feature hashing ouptut?

2 votes
1 answer
44 views

Using outcome label to create feature qualifies as data leakage?

2 votes
1 answer
102 views

1D cluster - Jenks optimization - Finding optimal number

1
2 3 4 5