DaL's user avatar
DaL's user avatar
DaL's user avatar
DaL
  • Member for 8 years, 8 months
  • Last seen more than a month ago
108 votes

Why is accuracy not the best measure for assessing classification models?

12 votes
Accepted

Classification with a neural network when one class has disproportionately many entries

9 votes

Techniques to detect overfitting

8 votes

Is machine learning only about estimating programs?

7 votes

What balancing method can I apply to an imbalanced data set?

7 votes

Is it pointless to use Bagging with nearest neighbor classifiers?

7 votes

Neural Network: Why can't I overfit?

6 votes
Accepted

Training Neural Net with examples it misclassified

6 votes
Accepted

Is there an intuitive interpretation of mutual information values (bits, nits)?

5 votes

What is the origin of the term confusion matrix?

5 votes

Best approaches for feature engineering?

5 votes
Accepted

How to check for the distribution stability?

5 votes
Accepted

What is a Hamming Loss ? will we consider it for an Imbalanced Binary classifier

4 votes

What base classifiers to use with bagging?

4 votes

Defining the usability / applicability of the data for machine learning

4 votes

Arguments against model or forecast combination?

4 votes

lifetime of fraud detection models

4 votes
Accepted

Semi supervised classification with unseen classes

4 votes

What is the benefit of breaking up a continuous predictor variable?

3 votes

What is your favorite "data analysis" cartoon?

3 votes
Accepted

Why isn't Hosmer-Lemeshow test used for other models besides Logistic Regression?

3 votes
Accepted

Comparing 0/10 to 0/20

3 votes
Accepted

Is it true that "from an information theory point of view, lexical white noise is just about the opposite of interesting"?

3 votes

Could a predictor with a zero correlation to a response be significant in a multiple regression?

2 votes

Who is credited with inventing the term "high dimensional data"?

2 votes
Accepted

Changing classifier input features from unigrams to bigrams

2 votes
Accepted

Classification problem: compare results of a decision tree, naive bayes and 1-nearest-neighbor classifier

2 votes

Machine learning or statistical models that account for time evolution and underlying system changes

2 votes
Accepted

Churn Prediction: why are churners usually selected as the positive class and not vice versa?

2 votes

A statistical approach to determine if data are missing at random