Machine learning is a technology that is breaking ground at new speeds every day. Technically, it should be improving faster and faster, given that ML is essentially supposed to be learning itself.

Modeling & Machine Learning Interview

Introduction

Overview

How Much Machine Learning?

Types of Machine Learning Interview Questions

Introduction to Machine Learning

The machine learning and modeling case study is the most common type of interview question that tests a combination of modeling intuition and business application.

Case Study Framework

Modeling Case Study

Data processing and analysis is the first step that we need to consider once we've clarified details and started down the path of building the model.

Missing Housing Data

Variable Error

Skewed Pricing

Bank Fraud Model

Data Pre-Processing

Feature selection and feature engineering is the second part of the data processing step. Once we've understood what our data looks like, we need to begin to theorize the kinds of features we would use to build the model.

Ride Requests Model

Multicollinearity in Regression

Keyword Bidding

Feature Selection

Model selection is usually the crux of any modeling case study problem. We want to be able to select a model or machine learning algorithm that will combine a bunch of factors to become the most optimal algorithm for the problem.

Booking Regression

Evaluate News

Credit Card Fraud Model

Model Selection

We have touched on the different machine learning algorithms throughout this lesson, but haven't yet dived deep into each one. The prior for this course is that you, as a candidate, have an idea of basic machine learning concepts, and the different modeling algorithms are one such example of them.

Machine Learning Algorithms Framework

Regression, Regularization, and Gradient Descent

Tree Algorithms

Coefficients of Logistic Regression

Encoding Categorical Features

Assumptions of Linear Regression

Machine Learning Algorithms

Recommendation and search engines are a subclass of the information filtering domain. They are used in almost every single modern web application or platform. Implementing recommendation and search engines usually requires a complex team of data scientists, data and software engineers, and machine learning engineers to collaborate and build at every company.

Comparing Search Engines

Job Recommendation

Newsfeed Model

Restaurant Recommender

Recommendation and Search Engines

Most machine learning model deployment requires some technical details and implementation to doing so. But we can abstract away from that in an interview when we’re focusing on the model roll out.

Cross Validation

Evaluation Metrics

Testing and Roll Out

Loan Model

Food Delivery Times

Spam Classifier

Delivery Estimate Model

85% vs 82%

Model Evaluation

Applied modeling is a type of case question asked about practical machine learning. The most common type of question framework is: Given an example scenario with a machine learning system or model, how would you analyze and fix the problem?

Rejection Reason

Active User Probability

One Million Rides

Parents Joining Teens

Applied Modeling

Machine learning system design focuses more heavily on the engineering aspects of model deployment. While slightly out of scope, we still wanted to cover some of the basics.

Type-ahead Search

Youtube Recommendations

Spotify Discover Weekly

Dynamic Demand Pricing

Machine Learning System Design

Regression models are used to predict the value of a dependent variable from one or more independent variables.

Introduction to GLMs

Linear Regression

Lasso vs Ridge

Using R Squared

Logistic Regression

Perfectly Separable

Softmax vs Logistic

Time Series Model

Why Do We Need Time Series Models?

AR and MA

Significance Time Series

Generalized Linear Models and Regression

Understanding of how to build, deploy, and test machine learning models and tackle business problems while applying modeling concepts. 

Learn how to apply machine learning to influence business decisions

What are time series models? Why do we need them when we have less complicated regression models?

What can happen if a traditional regression model is used on data that has autocorrelation?

What can happen if a traditional regression model is used on data that has autocorrelation?

Why Do We Need Time Series Models? I

The model may overestimate the parameters, leading to an extremely large R^2 value

The model may underestimate the parameters, leading to a low R^2 value

The model will not be able to process the data

The model will ignore the autocorrelation, producing accurate results

The model will automatically adjust to accommodate the autocorrelation

Why can't standard regression models be used effectively for data points where current values are influenced by past values?

Why can't standard regression models be used effectively for data points where current values are influenced by past values?

Why Do We Need Time Series Models? II

Standard regression models assume no autocorrelation, i.e., past values do not influence present values

Standard regression models are too complex for such data points

Standard regression models cannot handle large data sets

Standard regression models do not work with time-dependent data

Standard regression models require data to be in a specific format

Modeling & Machine Learning Interview

Why Do We Need Time Series Models?
Go to question details page

31%

Modeling & Machine Learning Interview

Why Do We Need Time Series Models?Go to question details page

31%

Why Do We Need Time Series Models?
Go to question details page