Humans have taken their voices online
en-masse. Twitter publishes half a billion tweets each day. Over one billion
people engage their friends via Facebook. So what makes Facebook worth $170
billion - about $150 per active user? Mainly, it's the ability to sell things
to those users, based on analysis of their data and behavior.
That's what a course at Statistics.com, Social
Data Mining with Python, is about.
You will use Python to extract valuable signals from these huge, chaotic
datasets to explain collective behavior and create computational knowledge
bases. Using Python, you will analyze user-generated content such as movie
ratings, online comments, status updates, and friendship networks. You will
learn algorithms from the fields of social network analysis, text analysis, and
recommender systems. Finally, you will gain experience with pragmatic workflows
that leverage social APIs to reveal human insights in your own projects. This course includes substantial Python
instruction, and is a good way for those with some Python familiarity to extend
their knowledge. The instructor of this course is Dr. Shilad Sen. For more details
about course and instructor please visit at http://www.statistics.com/python-social-data/.
Who Should Take
This Course:
Programmers and statisticians familiar with
Python who want to learn how to do analysis of text and social network date;
analysts who know some Python and who want to deepen their Python knowledge by
learning how to mine social data.
Course Program:
Week 1:
Recommendation algorithms
- Refresher
on Python data structures
- Streaming
large datasets
- Filtering
noise in long-tailed datasets
- Algorithmic
time complexity in a nutshell
- Introduction
to recommendation algorithms
- Case
study: If you liked “Star Wars” you’ll like ??
Week 2:
Introduction to text analysis in Python
- Python’s
“yield” keyword
- Tf/Idf
weighting
- Algorithms
that identify distinctive language
- K-means
clustering
- Case
study: Distinctive language in subreddit communities
Week 3: Social
APIs
- A
taxonomy of social APIs and Python interfaces to them
- Practical
workflows when using social APIs
- Case
study: Datasift & sentiment analysis on Twitter
Week 4: Social
network analysis
- Network
data structures in Python
- Your
Facebook graph: visualization and community detection
- Using
the Gephi network visualization tool
- Case
study: Inducing network graphs - the landscape of movies
The instructor, Dr. Shilad Sen is an
Assistant Professor at Macalester College in St. Paul, and he builds systems
that empower people to be better contributors to online communities. His
professional experience also includes Sourcelight Technologies, Google, IBM
Research, and Thomson Reuters R & D.
Course participants will ask questions and exchange comments with Dr.
Sen via a private discussion forum.
You will be able to ask questions and exchange
comments with the instructors via a private discussion board throughout the
course. The courses take place online
at statistics.com in a series of 4 weekly lessons and assignments, and require
about 15 hours/week. Participate at your
own convenience; there are no set times when you must be online. You have the
flexibility to work a bit every day, if that is your preference, or concentrate
your work in just a couple of days.
We, the Center for eLearning and Training
(C-eLT), Pune, partner with Statistics.com and offer these courses to Indian
participants at special prices payable in INR.
Call: 020 6680 0322
No comments:
Post a Comment