Machine Learning: Capstone Project

IBM via edX

Go to class Write review

Details

Go to class

Provider

edX
Pricing

Free Online Course (Audit)
Languages

English
Certificate

$149.00 Certificate Available
Duration & workload

1 week, 20-24 hours a week
Sessions

On-Demand
Subtitles

English

Found in

Overview

This Machine Learning Capstone is designed to showcase and solidify your expertise in Python-based machine learning. In this hands-on course, you’ll bring together everything you’ve learned in previous courses in the program and apply it to real-world problems using libraries such as Pandas, Scikit-learn, and TensorFlow/Keras.

Your main project will focus on building a course recommender system. You’ll work with course-related datasets, calculate cosine similarity, create similarity matrices, and experiment with multiple algorithms. By applying K-Nearest Neighbors (KNN), Principal Component Analysis (PCA), and non-negative matrix collaborative filtering, you will compare and contrast the performance of different machine learning approaches to recommendation systems.

Beyond recommendation systems, you'll also train a neural network to predict course ratings and build regression and classification models to enhance your predictive analytics skills. This project gives you the opportunity to demonstrate not just technical proficiency, but also critical thinking in evaluating and selecting the most effective models.

By the end of the course, you’ll have a portfolio-worthy project, practical experience with advanced machine learning techniques, and the confidence to apply your skills to real-world challenges.

Syllabus

Welcome

Video: Introduction to Machine Learning Capstone
Ungraded Plugin: Capstone Overview

Module 1: Machine Learning Capstone Overview

Reading: Learning Objectives
Video: Introduction to Recommender Systems
Reading: Text Analysis
Reading: Stopwords and WordCloud
App Item: Exploratory Data Analysis on Online Course Enrollment Data
App Item: Extract Bag of Words (BoW) Features from Course Textual Content
Reading: Sparse and Dense Bag of Words (BOW) Vectors
Reading: Similarity Measures in Recommender Systems
App Item: Calculate Course Similarity using BoW Features
Practice Assignment: Checkpoints: Exploratory Data Analysis on Online Course Enrollment Data
Graded Assignment: Exploratory Data Analysis and Feature Engineering

Module 2: Unsupervised-Learning Based Recommender System

Reading: Learning Objectives
Video: Content-based Recommender Systems
Reading: Evaluation Metrics of Recommender Systems
App Item: Content-based Course Recommender System using User Profile and Course Genres
Reading: Heatmaps
App Item: Content-based Course Recommender System using Course Similarities
App Item: Clustering-based Course Recommender System
Practice Assignment: Checkpoints: Unsupervised-Learning Based Recommender System
Graded Assignment: Unsupervised-Learning Based Recommendation Systems

Module 3: Supervised-Learning Based Recommender Systems

Reading: Learning Objectives
Video: Collaborative Filtering-Based Recommender Systems
Reading: Exploring Surprise Library and KNN Model
App Item: Collaborative Filtering-based Recommender System using K Nearest Neighbor
App Item: Collaborative Filtering-based Recommender System using Non-negative Matrix Factorization
App Item: Course Rating Prediction using Neural Networks
App Item: Regression-based Rating Score Prediction Using Embedding Features
App Item: Classification-based Rating Mode Prediction using Embedding Features
Practice Assignment: Checkpoints: Supervised-Learning Based Recommender Systems
Graded Assignment: Supervised-Learning Based Recommendation Methods