COM5508 Media Data Analytics

A graduate course on computational methods for media research, covering data collection, text mining, network analysis, and machine learning applications in communication studies.

Instructor Xiaofan Liu (Teaching Assistant: Zhicong Chen)
Term Spring 2019
Location City University of Hong Kong

Course Overview

This graduate seminar introduces students to computational and quantitative methods for media and communication research. Topics span the full research pipeline: data collection (APIs, scraping), text analysis (sentiment, topic modeling, word embeddings), social network analysis, and machine learning classification.

Zhicong Chen served as Teaching Assistant for this course.

Learning Objectives

By the end of this course, students will be able to:

  • Collect large-scale media and social data programmatically using APIs and web scraping
  • Apply text mining techniques (TF-IDF, topic modeling, sentiment analysis) to communication research questions
  • Construct and analyse social networks from media interaction data
  • Train and evaluate machine learning classifiers on media content
  • Critically assess the validity and ethics of computational media research

Level

Graduate (PhD and Research Masters)

Institution

Department of Media and Communication, City University of Hong Kong

Key Software and Libraries

Tool Purpose Link
Python 3 Core language python.org
NLTK NLP preprocessing nltk.org
Gensim Topic models & word vectors radimrehurek.com/gensim
scikit-learn Machine learning scikit-learn.org
NetworkX Network analysis networkx.org
Gephi Network visualisation gephi.org
Beautiful Soup Web scraping crummy.com/software/BeautifulSoup

Assessment

Component Weight
Weekly lab exercises 30%
Literature review 20%
Final research project 45%
Participation 5%

Schedule

Week Date Topic Materials
1 Week 1 Introduction to Computational Communication Research

Overview of big data in media studies; the promise and limitations of computational methods; research workflow.
2 Week 2 Python for Media Research

Python environment setup, Jupyter notebooks, pandas basics, data types and structures.
3 Week 3 Collecting Data from APIs

REST APIs, authentication, rate limiting; collecting data from social media and news APIs.
4 Week 4 Web Scraping for Media Research

Scraping news sites, HTML parsing with BeautifulSoup, handling dynamic pages with Selenium.
5 Week 5 Text Preprocessing and Representation

Tokenization, stopwords, stemming, TF-IDF; building document-term matrices.
6 Week 6 Content Analysis and Dictionary Methods

Automated content analysis; sentiment dictionaries; LIWC; validation strategies.
7 Week 7 Topic Modeling with LDA

Latent Dirichlet Allocation; selecting K; interpreting and labelling topics; stability.
8 Week 8 Social Network Analysis I: Basics

Graph theory fundamentals; degree, betweenness and closeness centrality; NetworkX.
9 Week 9 Social Network Analysis II: Communities

Community detection algorithms; modularity; information diffusion; visualisation with Gephi.
10 Week 10 Machine Learning for Media Classification

Naive Bayes, SVM, random forests; train/test split; precision, recall, F1.
11 Week 11 Word Embeddings and Semantic Analysis

Word2Vec, GloVe; using pre-trained embeddings; semantic similarity in media text.
12 Week 12 Research Design and Reproducibility

Pre-registration, open data, replication; writing up computational studies; ethical considerations.
13 Week 13 Student Project Presentations I

Student teams present computational media research projects.
14 Week 14 Student Project Presentations II

Remaining presentations; course wrap-up and reflections.