1. CS410: Text Information Systems (Spring 2018)
Instructor: ChengXiang “Cheng” Zhai
Full-Time TAs:
Bingjie Jiang
Qihao Shao
Part-Time TAs:
Chase Geigle
Eddie Huang
Dominic Seyler
Sheng Wang

2. Motivation: Harnessing Big Text Data
Text data is ubiquitous and growing rapidly
Internet
Blogs
News
Literature
Twitter …
Many
applications!
Knowledge

3. Humans as Subjective & Intelligent “Sensors”
Sense
Report
Real World
Sensor
Data
Weather
Thermometer
3C , 15F, …
Perceive
Express
“Human Sensor”
Locations
Geo Sensor
41°N and 120°W ….
Network Sensor
Networks

4. Unique Value of Text Data
Useful to all big data applications
Especially useful for mining knowledge about people’s behavior, attitude, and opinions
Directly express knowledge about our world Small text data are also useful!
Data  Information  Knowledge
Text Data

5. Main Techniques for Harnessing Big Text Data: Text Retrieval + Text Mining
Small
Relevant Data
Small Relevant Data
Knowledge
Many Applications

6. Big Design of CS410: Overview MOOC 2 MOOC 1 Course ProjectHi
Online Videos + High Engagement
MOOC 1
MOOC 2
Course
Project
Text Retrieval
Text Mining
Big
Text Data
Big Text Data
Small
Relevant Data
Small Relevant Data
Knowledge
Many Applications

7. Design of CS410: Goals Emphasize both theory and practice
Theory: basic concepts and general principles are applicable to all applications  Lectures + Quizzes
Practice: specific practical skills are immediately useful  Programming assignments
Integration of theory and practice  Course projects

8. Design of CS410: Goals
Personalized learning  Self paced + Choices of project
Collaborative learning  Forum-based interactions and collaboration + Group projects

9. Prerequisites Required: Optional:
Proficiency in programming (needed for assignments and projects)
Comfortable with programming (ideally C++ or Java, but Python would also be okay)
Optional:
Knowledge of basic probability & statistics (helpful for understanding algorithms deeply)
Contact the instructor if you aren’t sure

10. Textbook & Readings Textbook (available online)
Text Data Management and Analysis: A practical introduction to text mining and information retrieval, by ChengXiang Zhai, Sean Massung, ACM and Morgan & Claypool Publishers, 2016
Notes and additional readings will be available online as needed

11. Design of CS410: Format & Grading
Extra
Credit:
+ 5%
Synchronous Weekly Class Meetings & Office Hours
Asychronous Question Answering & Discussion via Forums Hi
MOOC 1
MOOC 2
Course
Project
Text Retrieval
Text Mining
25%
Proposal
15%
25%
60% 5%
45%
25%
Lecture Videos
Lecture Videos
Quizzes
Quizzes
Presentation
MP Assignment
MP Assignment
Report

12. You have Complete Control over Your Grade!
B+: [80, 84]
B: [75, 79]
B-: [70,74]
C: [60, 69]
D: [55,59]
F: <55
5% Extra Credit
would help move your grade up
by one bracket

13. Format Lecture Videos + In-class Discussions & Problem Solving
No class meetings on most Thursdays, but we meet and give a quiz on almost every Tuesday
Watch video before Thur Submit summary (questions/quizzes) before next Tue  Get questions answered in class next Tue Take quizzes the Tue after
Assignments: 4~5 assignments (experiments + coding)
No exam
Course project
Literature review (only required for 4 credit hours)

14. Weekly Schedule You watch videos before Thursday
You submit a summary of what you’ve learned in the subsequent weekend
The difficult lecture segments + Specific questions (if you have difficulty in understanding some part), or
A quiz question to test knowledge about the lecture (if you mastered the materials well)
Summaries will be published to help all of you prepare for the quiz
We’ll review the difficult concepts and answer your questions in the class meeting next Tuesday and give a quiz to test topics in a previous lecture
We also discuss assignments and projects to help you finish them

15. Forum Discussion
Forum (Piazza) is the primary way of interactions and engagement
Asynchronous discussion enables participation of everyone at any time
Enables faster question answering without waiting until a class meeting or office hour
Facilitates identification of difficult concepts to be covered in class meetings and office hours

16. Protocol of Question Answering
As soon as you have a question or issue to discuss, post it immediately on Forum and submit the question
If the question is not answered in a timely manner on Forum or addressed adequately, the question to all of us (i.e., the instructor and 4 TAs) using a subject line containing the keyword “CS410S18”
If you don’t receive a reply from us by in a timely manner, come to an office-hour.

17. Format of Office Hours
TAs and the instructor will hold weekly office hours at published time slots
Special office hour by instructor
Thursdays, 11am-12:15pm, in 1404 SC (classroom)
Priority list in descending order:
High: Issues posted on Forum, but unresolved even after communications with the TAs/Instructor
Medium: Other unresolved issues on Forum
Low: Any questions or issues not posted on Forum, brought by a student joining an office hour (first come, first serve)

18. How to Get the Most out of CS410?
Watch every lecture video in a timely manner!
Identify and ask questions before the Tuesday meeting
Read in advance if possible
Collaborative learning: help each other to get an “A”!
Actively participate in forum discussions (you’ll learn from reading posts on Forums)
Earn up to 5% extra credit by making effort to answer others’ questions on Forums (your effort will be logged on the Forum)
Post questions on Forum immediately whenever you have difficulty in understanding any part of the course materials

19. Your Work Load … … Jan Feb Mar Apr May 1/19 Last day of instruction
Spring
Break
Video
Watching
&Q A
& Quiz
Assign #1
Assign #2 … …
Final Week
Assign #k
Project
Literature Review

20. Questions? Course website: Course Piazza:
Course Piazza: 
Course Compass space: