Suppose we have a binary classification problem. 99.8% of individuals in our sample = 0, 0.2% = 1. We downsample: we randomly sample say 1% of the individuals which =0, but keep all which =1. We fit our model to this new smaller sample, and now we have a model which produces probabilities that an individual = 1. How do we adjust our output probabilities to use this model on the population as a whole?
It was a mix of both technical and psychometric questions, e.g Tell us you experience as a data analyst.
1. Tell me about yourself 2. Did you have experiences about customer success?
Behavioural questions, Background, past experience
What kind of job are you most interested in?
SQL joins
There were a couple of questions but these are the ones that I remember: 1. Tell me about yourself 2. Why apply for this position? 3. What are your strengths & weaknesses? 4. Present one of your projects 5. What do you expect from the company?
Describe how you’ve experienced diversity in the workplace.
What is data science do?
Do you question data after seeing it?
Viewing 661 - 670 interview questions