Text Mining and Natural Language Processing

Course Description

Students collect and analyze unstructured text data using web/API scraping methods, and then analyze their corpus using text mining and natural language processing. Additionally, students conduct a survey of relevant issues pertaining to privacy rights and intellectual property rights for web scraping and text mining methods.

Syllabus

Student Learning Outcomes, Goals, Objectives:

Students successfully completing this course will be able to:

  • Explain foundational concepts and applications of text mining and NLP.

  • Collect, clean, tokenize, normalize, and linguistically annotate text data.

  • Represent text using Bag-of-Words, TF-IDF, n-grams, and embeddings.

  • Build and evaluate machine-learning models for text classification and sentiment analysis.

  • Apply topic modeling, clustering, named entity recognition, and information extraction.

  • Explain neural sequence models, attention, transformers, and pretrained language models.

  • Use Generative AI and LLM tools for prompting, semantic search, and basic RAG applications.

  • Evaluate NLP systems for accuracy, bias, privacy, hallucination, and responsible use.

Course Grading Information:

Activity/Performance Measure

Percentage/Points

Programming Assignments

30%

Quizzes and Discussions

15%

Midterm Exam

20%

Final Project Report and Code

25%

Final Presentation

10%

Attendance/Participation:

Students are expected to complete weekly Canvas modules, readings, videos, discussions, quizzes, and assignments. Regular participation is expected.

Work submitted within seven calendar days after the deadline may receive a 25% deduction. Work more than seven days late is normally not accepted. Documented emergencies should be communicated promptly.

Course Materials Purchased by the Students:

No course materials are required.

Scholarly Perspectives

This course engages diverse scholarly perspectives to develop critical thinking, analysis, and debate and inclusion of a reading does not imply endorsement.