Michael McGowan's user avatar
Michael McGowan's user avatar
Michael McGowan's user avatar
Michael McGowan
  • Member for 12 years, 5 months
  • Last seen more than 1 year ago
  • West Chester, PA, United States
69 votes
9 answers
15k views

How can I help ensure testing data does not leak into training data?

37 votes
5 answers
37k views

Are decision trees almost always binary trees?

31 votes
4 answers
79k views

Predicting with both continuous and categorical features

25 votes
3 answers
7k views

Visualizing the calibration of predicted probability of a model

18 votes
5 answers
1k views

Does this quantity related to independence have a name?

12 votes
2 answers
8k views

Predicting multiple targets or classes?

12 votes
2 answers
838 views

Domain-agnostic feature engineering that retains semantic meaning?

9 votes
1 answer
4k views

Bayesian inference for multinomial distribution with asymmetric prior knowledge?

6 votes
2 answers
371 views

Is there a generic term for measures of correctness like "precision" and "recall"?

4 votes
0 answers
82 views

How can I help ensure testing data does not leak into training data? [duplicate]

4 votes
1 answer
1k views

Are there terms to distinguish between the two types of outliers?

3 votes
2 answers
6k views

Can the mutual information of a "cell" be negative?

3 votes
1 answer
513 views

What is the association rule learning approach to the logical XOR problem?

2 votes
0 answers
1k views

Is it OK to do additive smoothing before applying Pearson's chi-square test for independence?

2 votes
2 answers
4k views

Does "x per y" always refer to the arithmetic mean?

2 votes
1 answer
175 views

Can I define an $R^2$-like measure in this way when predicting with exotic loss functions?

1 vote
1 answer
561 views

How can I better predict with (g)lmer with missing values?