(after describing a project on my resume) explain recall in plain english? why did you use recall as evaluation metric? how do you calculate recall for mutli-label data? how did you decide the number of hidden layers and number of neurons in each layer of the fully connected network that you have designed for this project? were there imbalances in the labels in your data? what have you done to handle that huge imbalance?

Question

Accepted Answer

Recall is a metric used to measure the ability of a model to find all relevant instances in a dataset.

Recall is the ratio of true positives to the sum of true positives and false negatives.
It is used as an evaluation metric to assess the model's ability to identify all positive instances correctly.
In multi-label data, recall can be calculated for each label separately and then averaged.
The number of hidden layers and neurons in each layer of the fully connected network is determined through experimentation and tuning.
Imbalances in labels refer to unequal distribution of different classes in the dataset.
Techniques like oversampling, undersampling, or using class weights can be employed to handle label imbalances.

Top Chubb Data Scientist interview questions & answers

Popular interview questions of Data Scientist

Top HR questions asked in Chubb Data Scientist