Skip to content

Syllabus

C S 5293-001: Natural Language Processing — Spring 2026 (3 credits) University of Oklahoma · School of Computer Science / Graduate Program

Course at a glance

  • Instructor: Jie Cao · jie.cao@ou.edu
  • TA: none this semester
  • Meetings: In person, TR 1:30–2:45 pm, Sarkeys Energy Center A0236
  • Office hours: TBA
  • Communication: Discord (course staff reachable via a private post); Canvas
  • Prerequisites: C S 2413 / 2414 / 5005; C S 2813 or MATH 2513; and an approved statistics course

Course description

This course covers the fundamental algorithms and methods in natural language processing (NLP). Topics include n-gram models, text classification, statistical models over linguistic structures (sequences, trees, and graphs), vector semantics, neural networks, large language models, and more. Applications include parsing, information extraction, machine translation, topic modeling, and dialog systems. Students build computational models across different areas of NLP.

Textbook & materials

  • Speech and Language Processing, Dan Jurafsky & James H. Martin, 3rd Edition (2024). Free online at web.stanford.edu/~jurafsky/slp3 — required reading.
  • Extra lecture notes on special topics (LLMs, Agentic AI), shared on Canvas.
  • This interactive notebook book for hands-on tutorials and assignments — see the Table of Contents.

Learning outcomes

By the end of the course you will be able to:

  • Understand the problems, algorithms, and methods of NLP and their relationship to linguistics and statistics — including basic linguistic phenomena, statistical NLP models, neural representations, deep learning, and large language models.
  • Design, implement, and evaluate practical NLP systems — analyze text with n-gram models and linguistic features; train and evaluate text classifiers and improve them with neural networks; implement structured prediction for sequences, trees, and graphs; improve performance with pretrained language models; and prompt-engineer and post-train LLMs for specific tasks.
  • Read NLP research papers and contribute initial findings through collaboration — survey a research area, and design initial experiments to compare existing methods and explore new ideas.

Grading

Component Weight
Assignments (5, individual) 50%
Course Project (groups of ≤ 2) 30%
In-Class Paper Quizzes 15%
Participation 5%

Letter grades: A ≥ 90, B 80–89, C 70–79, D 60–69, F < 60.

Participation reflects your engagement in class discussions — live (lectures, sessions, in-class quizzes) or online (public/anonymous/private posts on Discord).

See the Assignments and Schedule pages for the full list of deliverables and due dates.

Course policies

Generative AI policy

This course encourages thoughtful, creative use of generative-AI tools — within clear guardrails:

  • Never use AI tools for in-class quizzes (violation → 0 on the quiz).
  • You may use any free generative-AI tools (e.g., Copilot, open LLMs) for assignments and the project, so everyone has equal access. Do not fully delegate code writing to AI — prefer its autocompletion features.
  • You must cite AI usage — model version/date and any quoted or paraphrased ideas/content — per the class citation guidelines.
  • You must include an AI-usage statement at the end of each assignment, written without AI. It must state the platform(s) and prompt(s) used, summarize how AI helped you meet the learning objectives, and describe what you did to verify the output and make the work your own.

Deviating from these acknowledgement, citation, and reflection guidelines may be treated as an academic-integrity violation. The instructor will model appropriate AI use by disclosing when and why it is used, and will never use AI to grade your work.

Attendance

You are expected to attend or view all lectures. Participation grades reflect engagement in lectures, office hours, check-in quizzes, and online forums. If you are ill, notify the staff in advance so an alternative can be arranged.

Readings

The schedule lists readings for each lecture, mainly from Jurafsky & Martin (3rd ed.). You are responsible for the material before class. For LLM and Agentic-AI content, free lecture notes and notebooks are provided.

Assignments & submission

Written and programming assignments are released and submitted through Canvas; programming is in Python.

Late policy

Assignments are due at 12:00 pm (noon). Work is accepted up to 24 hours late with a 10% penalty; nothing is accepted beyond 24 hours. The Canvas timestamp is authoritative and the policy is enforced strictly (12:01 pm → 10% penalty). Documented extenuating circumstances are handled individually.

Course project

Group project (≤ 2 students; individual by prior approval). Choose a shared task or propose your own research project — no double-dipping across classes; data and evaluation must be clearly defined and ready to use. Graded across three milestones (proposal, mid-term demo, final report). Sync closely with the instructor.

Quizzes & exams

  • Scheduled quizzes (15%): in-class, on paper, closed book (paper notes allowed), scope announced ≥ 1 week ahead.
  • Unannounced check-ins (participation only): short, device-based.
  • No final exam this semester.

Communication: Canvas & Discord

Canvas hosts readings, assignments, and submissions. Discord is the fastest way to ask questions (public channels, or direct messages for private matters). Use email for personal matters such as extended illness. The Discord sign-up link is on Canvas.

Academic conduct

You may discuss assignments with staff and discuss ideas with classmates, but you may not look at or share code (similar code scores zero). Document collaborators and all external resources (Discord, StackOverflow, ChatGPT, or other AI tools). Programming work is checked with collaboration-detection software; first violations are reported per the OU Academic Misconduct Code. Providers and receivers of a solution are treated equally.

Classroom conduct

Cell phones and laptops should not be used during class except for classroom exercises. Disruptive behavior may result in removal and a Student Code violation.

Incompletes

A grade of "I" is reserved for rare cases and requires all three: within two weeks of the end of the semester, a current grade of C or better, and a properly documented, compelling reason.

Course reflection survey

A confidential end-of-semester survey is strongly encouraged; results are shared with the instructor only after final grades and inform future offerings.

University policies

The University of Oklahoma's standard policies apply to this course, including:

  • Mental Health Support — University Counseling Center, 405-325-2911; 24/7 crisis support via TimelyCare.
  • Title IX — faculty and TAs are mandatory reporters; confidential support via OU Advocates (24/7, 405-615-0013) and the Institutional Equity Office (405-325-3546).
  • Reasonable Accommodations — register with the Accessibility and Disability Resource Center (ADRC), 405-325-3852, adrc@ou.edu, ou.edu/adrc.
  • Religious Observance, Final Exam Preparation Period, Adjustments for Pregnancy and Related Issues, and Emergency / Active-Threat Protocols per University policy.

This page summarizes the official syllabus. The PDF distributed on Canvas is the authoritative version.