Building Intelligent Web Applications (AA-TTAML0001)

This course is an applied artificial intelligence (AI) course that teaches you how to create machine learning applications that crunch and wrangle data collected from users, web applications, and website logs. Leveraging current standards, skills, and practices, you'll explore intelligent algorithms that extract value from data. Important machine learning concepts are taught using code examples in Python's scikit-learn. This course guides you through algorithms to capture, store, and structure data streams from the web. You'll explore recommendation engines and jump into classification via statistical algorithms, neural networks, and deep learning.

Building applications for the intelligent web

• An intelligent algorithm in action: Google Now
• The intelligent-algorithm lifecycle
• Further examples of intelligent algorithms
• Things that intelligent applications are not
• Classes of intelligent algorithm
• Evaluating the performance of intelligent algorithms
• Important notes about intelligent algorithms

Extracting structure from data: clustering and transforming your data

• Data, structure, bias, and noise
• The curse of dimensionality
• K-means
• The relationship between k-means and GMM
• Transforming the data axis

Recommending relevant content

• Setting the scene: an online movie store
• Distance and similarity
• How do recommendation engines work?
• User-based collaborative filtering
• Model-based recommendation using singular value decomposition
• The Netflix Prize
• Evaluating your recommender

Classification: placing things where they belong

• The need for classification
• An overview of classifiers
• algorithms
• Fraud detection with logistic regression
• Are your results credible?
• Classification with very large datasets

Case study: click prediction for online advertising

• History and background
• The exchange
• What is a bidder?
• What is a decisioning engine?
• Click prediction with Vowpal Wabbit
• Complexities of building a decisioning engine
• The future of real-time prediction

Deep learning and neural networks

• An intuitive approach to deep learning
• Neural networks
• The perceptron
• Multilayer perceptronís
• backpropagation
• Going deeper: from multilayer neural networks to deep learning

Making the right choice

• A/B testing
• Multi-armed bandits
• Bayesian bandits in the wild
• A/B vs. the Bayesian bandit
• Extensions to multi-armed bandits

The future of the intelligent web

• Future applications of the intelligent web
• Social implications of the intelligent web

Join an engaging hands-on learning environment, where you'll learn:

• Machine learning essentials, as well as deep learning and neural networks
• How recommendation engines work
• Building applications for the intelligent web
• Extracting structure from data: clustering and transforming your data
• Recommending relevant content
• Classification: placing things where they belong
• Relevant Case Study: click prediction for online advertising
• Making the right Machine Learning choices for your web apps
• The future of the intelligent web

This course has a 50% hands-on labs to 50% lecture ratio with engaging instruction, demos, group discussions, labs, and project work. This is not a basic class.

Before attending this course, you should have:

• Basic to Intermediate IT Skills, with some prior Python exposure if able (attendees without a programming background like Python may view labs as follow along exercises or team with others to complete them)
• Good foundational mathematics or logic skills
• Basic Linux skills, including familiarity with command-line options such as ls, cd, cp, and su

Do you have more questions? We're delighted to assist you!

Developer, Data Scientist, Architect, and Software Engineer