TEN famous questions for basic Data science interview, Part V

  • بادئ الموضوع Hicham AMAR
  • تاريخ البدء
H

Hicham AMAR

Guest
The Data science interview is an essential step with the technical staff of the new company that you want to integrate. So, it is important to prepare yourself to answer to the theoretical questions and technical situations and some time to write a code using the language that you master (python or R).

You can find the parts I, II, III and IV in the section Data science of the Geoinfo4all.com blog.

  • Define AUC (Area under the ROC Curve)?

An evaluation metric that considers all possible classification thresholds. The area under the ROC curve is the probability that a classifier will be more confident that a randomly chosen positive example in actually positive then that randomly chosen negative example is positive.

  • What is backpropagation?

The primary algorithm for performing gradient descent a neural network. First, the output values of each node are calculated (and cached) in a forward pass, the n partial derivate of the error with respect to each parameter in calculated in backward pass through the graph.

  • What is bivariate analysis?

Bivariate analysis attempts to understand the difference between two variables at a time as in a scatterplot.

  • What is multivariate analysis?

Multivariate analysis deals with the study of more than two variables to understand the effect of variables on the response.

  • What do you understand by automation bias?

When a human decision maker favours recommendation made by an automated decision-making system over information made without automation, even when the automated decision-making system makes errors.

  • What is baseline?

A simple model or heuristic used as reference point for comparing how well a model is performing.

  • What do you mean by batch?

The set of examples used in one iteration (that is one gradient upade) of model training.

  • Define the term accuracy?

The fraction of predictions that a classification got to right. In multi*class classification, accuracy is defined as follows:

Accuracy = correct predictions/ total number of examples

In binary classification, accuracy has the flowing definition

Accuracy = true positive + true negative / total N of examples

  • What is activation function?

A function that takes in the weighted sum of all inputs from the previous layers and then generates and passes an output value (typically nonlinear) to the next layers.

  • What do you mean by Ada Grad?

A sophisticated gradient descent algorithm that rescales the gradient of each parameter, effectively given each parameter an independent learning rate.

Hicham AMAR, Ing in Geomatic Sciences, Co-founder of Geoinfo4all.com

متابعة القراءة...
 
أعلى