Learning Objectives -- Practical Data Science - Fall 2013

By the end of this course, you will (as time permits):

1. Be familiar with common tools for programming, development and data management:

1. unix command line and utilities
2. the Python programming language
3. basic database querying
4. for data access, data processing, visualization, and machine learning

2. Understand what data is: objects, relationships, & information
3. Know how to represent (store and retrieve) data in a variety of common formats
4. Use grep and regular expressions to process data: clean noise from raw data, extract exactly what is needed from raw data for your task
5. Interact with databases to query for relevant info, store data, provide a storage point for model results
6. Deal with big data: using hadoop to mine massive amounts of information
7. Using web APIs to query for diverse information, possibly setting up a simple API to act as an end point for a system you’ve developed
8. Find correlations between (attributes of) objects
9. Visualize data for exploratory and confirmatory analysis
10. Build models to make predictions given data, categorize objects
11. Transforming raw data into features that are useful for predictive models
12. Evaluate predictive models: how well do models predict a phenomenon?

1. produce quantitative evaluations
2. visualize model evaluations to assess business value / usefulness

13. Understand controlled experiments in the wild;

1. deploying models / data systems
2. comparing and understanding treatments

14. Understand some of the main applications of data science

1. Recommender Systems
2. Applications to online advertising
3. Others depending on case studies and guest speakers