Learning Outcomes

Understand how to use decision tree models in classification

You understand

• how to model data using decision models
• what are decision tree and their underlying assumptions
• how to contruct decision from data
• how to use decision trees for prediction
• how to use an ensemble of decision trees, aka, a random forest

Referencing modules: Decision Trees & Forests

Understand how to use SVM for classification

You understand

• what is linear separability
• what is a hyperplane
• what is the margin to nearest data points (support vectors)
• what is support vector machine
• how support vectors are computed from data

Referencing modules: SVM

Understand how to use probabilistic models in classification

You understand

• how to model data using probabilistic models
• what are probabilistic models and their underlying assumptions
• how to contruct probabilistic models from data
• how to use probabilistic models for prediction

Referencing modules: Probabilistic Classification

Understand how to use logistic regression for classification

You understand

• what is logistic regression
• how logistic regression is different from linear regression

Referencing modules: Logistic Regression

Understand how to use linear discriminant analysis

You understand

• what is linear discriminant analysis
• how it is related to dimensionality reduction
• how linear discriminant analysis work and their limitations

Referencing modules: Linear Discriminant Analysis

Understand how to use linear discriminant analysis

You understand

• what is a perceptron, multi-layer perceptron
• how gradient descent and backpropagation is used to train a neural network
• recurrent neural networks and long short term memory networks
• autoencoders and convolution neural networks
• how to apply neural networks to various classification problems

Referencing modules: Deep Learning

Understand how to use dimensionality reduction techniques

You understand

• the problem associated with high dimensions
• when dimensionality reduction techniques can be used
• what dimensionality reduction techniques are
• how dimensionality reduction techniques work and their limitations

Referencing modules: Dimensionality Reduction

Understand what features to use

You understand

• what are features and how they affect the analysis
• what are the common features for text data
• what are the common features for audio data
• what are the common features for video data

Referencing modules: Feature Engineering

Understand how this course is organized and basic concepts in big data analysis

You understand

• how this course is organized
• how to succeed in this course
• the life cycle of data and the data analysis workflow
• how to conceptualize data as tables/dataframes/matrices.
• how data can be categorized into types
• how to view data in geometric & linear algebra terms
• how to view data in probabilistic terms
• linear regression and the least squares method
• k-nearest neighbor classification
• how to perform data analysis in python
• how to visualize data using python

Referencing modules: Introduction

Understand how to think about data

You understand

• what are the different types of data
• how data is structured (or unstructured)
• the different ways of thinking about data mathematically
• how to analyze data in geometric & linear algebra terms
• how to analyze data in probabilistic terms
• how the mathematical view of the data affects how we address data quality issues

Referencing modules: Thinking about Data

Understand how to analyze co-occurrences

You understand

• what is frequent itemset mining
• how to define a transaction/event duration for observing co-occurrences
• how frequent itemset mining algorithms work
• how to extend co-occurrences to association rules
• how to apply frequent itemset analysis to real world scenarios
• how to validate analysis results

Referencing modules: Frequent Itemset Analysis

Understand how to analyse user rating data

You understand

• what are the different types of user rating data
• how to analyze user rating data using collaborative filtering
• how to use collaborative filtering techniques to make recommendations
• the limitations of collaborative filtering
• how to apply collaborative filtering to real world scenarios

Referencing modules: Collaborative Filtering

Understand how to perform cluster analysis

You understand

• what problems are addressed by cluster analysis
• how to analyze data using clustering techniques
• what are the limitations of clustering techniques
• how the k-means technique works
• how the agglomerative hierarchical clustering technique works
• how density-based clustering works
• how to apply clustering to real world scenarios

Referencing modules: Cluster Analysis

Understand how to process very large data sets

You understand

• what are the problems caused by very large data sets
• what are some of the technical approaches to processing large data sets
• what are some of the platforms and tools for processing large data sets
• how does virtualization help with processing large data sets
• the differences between virtual machines and containers
• what is a compute cluster and how does it work
• how to setup a computer cluster
• what are some of the programming paradigms for cluster computing

Referencing modules: Big Data Platforms

Understand how big data analytics affect society

You understand

• what are the societal and ethical issues associated with the use of big data analytics
• what are the key privacy issues
• what are the key bias
• how to use big data analytics responsibly

Referencing modules: The Dark Side

Apply big data processing and mining techniques

You are able to

• install, configure and use data management tools
• ethically acquire data for use in the project
• extract, clean, transform, load large data sets
• query and analyze large data sets to extract relevant information/features
• apply and compare different data mining techniques quantitatively

Referencing modules: Project