CSCI 662
Advanced Natural Language Processing

Time:

Tu/Th, 11:00am–12:20pm

Location:

KAP 140

Instructors:

Prof. David Chiang (chiang@isi.edu)
Prof. Kevin Knight (knight@isi.edu)

Teaching Assistant:

Qing Dou (qdou@isi.edu)

Course Description

Computers process massive quantities of information every day in the form of human language, yet machine understanding of human language remains one of the great challenges of computer science. How can advances in computing technology enable more intelligent processing of all this language data? Will computers ever be able to use this data to learn language like humans do? This course provides a systematic introduction to statistical models of human language, with particular attention to the structures of human language that inform them and the structured learning and inference algorithms that drive them. This is a lecture course, not a seminar course, but aims to cover both fundamental and cutting-edge research issues.

Audience This graduate course is intended for PhD students (or undergraduate or masters students who want to continue to a PhD program) who want to deepen their understanding of current natural language processing (NLP) research.

Prerequisites Students are expected to be proficient in programming, basic algorithms and data structures, discrete math, and basic probability theory.

Related Courses This course is part of USC's curriculum in natural language processing. There is a sister course, Applied Natural Language Processing (CSCI 544), offered in the Spring semester. You can take these two in either order.

Textbooks

• Class notes (to be distributed in class).
• Recommended but optional: Jurafsky and Martin, Speech and Language Processing (2nd ed.), Prentice Hall, 2008.

Requirements and policies (subject to change)

CourseworkStudents will experiment with existing NLP software toolkits and write their own programs. Grades will be based on:

• Quizzes (6 × 3.33% of grade): at the end of each unit.
• Programming assignments (5 × 12%): graded according to the correctness and clarity of the solutions. Sometimes a small part of the grade will depend on the performance of a system relative to the rest of the class.
• Research project (20%): graded according to the project's substantiality, correctness, and relevance to the course, as well as the clarity and depth of the project report.

All assignments and the project will be due at the beginning of class (11am) on the due date. Late assignments will be accepted with a 30% penalty up to one week late. No exceptions can be made except for a grave reason.

Students are expected to submit only their own work for homework assignments. They may discuss the assignments with one another but may not collaborate with or copy from one another.

Statement on Academic Integrity USC seeks to maintain an optimal learning environment. General principles of academic honesty include the concept of respect for the intellectual property of others, the expectation that individual work will be submitted unless otherwise allowed by an instructor, and the obligations both to protect one's own academic work from misuse by others as well as to avoid using another's work as one's own. All students are expected to understand and abide by these principles. Scampus, the Student Guidebook, contains the Student Conduct Code in Section 11.00, while the recommended sanctions are located in Appendix A: http://www.usc.edu/dept/publications/SCAMPUS/gov. Students will be referred to the Office of Student Judicial Affairs and Community Standards for further review, should there be any suspicion of academic dishonesty. The Review process can be found at: http://www.usc.edu/student-affairs/SJACS.

Syllabus

Suggested Readings

Some useful Python modules

• bigfloat: floating point numbers with very large range