Paul
  • Member for 6 years, 11 months
  • Last seen more than a month ago
Multi-class neural net always predicting 1 class after optimization
5 votes

You learn a lot by comparing to a naive model. A naive model is one without any features. As a default, it will always predict the most likely Target. Note that this is exactly what your model is ...

View answer
How to approach a Data Science case study question?
Accepted answer
2 votes

This question (something I've asked variants of several times in interviews) has absolutely nothing to do with statistical or other quantitative procedures. What is being asked here is for an ...

View answer
Is R2 score a reasonable regression measure on huge datasets?
1 votes

There is no general answer of what to expect for an $R^2$ score. And there is no general answer for whether a model with this $R^2$ score is a "good" model. There are many cases where (1) this kind ...

View answer
Translating a business problem into a machine learning solution: job-adds website
1 votes

With this kind of general problem there are many possible approaches, and I can't list them all. In general, you want to go with the simplest model that creates value over a more naive approach. So ...

View answer
Re-bucket weekly sales data and calculate descriptive statistics
Accepted answer
1 votes

If the week ID is given as you state, calculate a bucket variable $w_x=int(weekID/x)$. Then use a SQL statement to summarize the volume to levels of $w_x$.

View answer
How clustering is used in data management?
0 votes

This is a wide open question, so I am not sure which aspects you are most interested in. Here are some Wikipedia articles you can get started with. Metadata Harvesting Metadata Discovery

View answer
Is training in Biostats sufficient to get into Data Science?
0 votes

a question, and then some comments. First, what kind of market analyst do you want to be? This field is rapidly becoming specialized, and the answer will affect the decision. If you want to ...

View answer
How much of a background in programming is necessary to become a data scientist?
0 votes

The R language is the best place to start. Grab some open datasets, and start programming with r. R has many different analytical functions that you can learn a lot with it.

View answer
Can distribution values of a target variable be used as features in cross-validation?
0 votes

I agree that there is nothing wrong with using these type of features. I have used for inter-arrival times for example in modeling work. I have noticed however that many of these kind of features ...

View answer