大數據分析:文字探勘

This course aims to provide the methodological foundation of big data analytics in the field of communication and social sciences (computational communication research) through a hand-on approach. Real world dataset will be used to discuss the application of big data analytics like text mining, including data collection (data scraping), data cleaning, transformation, and annotation (data wrangling), model selection and analysis, and interpretation and data visualization. Popular open source machine learning tools, such as scikit-learn (https://scikit-learn.org/), TensorFlow (https://www.tensorflow.org/), and Weka (https://www.cs.waikato.ac.nz/ml/weka/) may be briefly introduced if the time allows. There is one point to be emphasized. Communication researchers and social scientists face a challenge that the surge of big data claims to revolutionize the process of scientific invention and discovery. The hype sparks off intense debates. On the one hand, some researchers argue that big data simply mean more data, and looking for patterns in observable data merely produces spurious correlation which is very difficult to be replicated from and generalized to other datasets. On the other hand, other researchers content that big data has demonstrated its excellence in prediction. The key issue behind these controversies is that the data-driven, inductive, and predictive feature of big data seems to be incompatible with the theory-driven, deductive, and explanatory nature of scientific method. This course will use R and Python. Please bring a laptop with the latest version of R, RStudio, and Python using the Anaconda distribution to each class.

Students are expected to have the following background: - Familiarity with basic statistics and R programing (at a minimum, at the level of DCT1038, or equivalent with permission of the instructor) - Basic knowledge of HTML and CSS (at a minimum, at the level of DCT2045, or equivalent with permission of the instructor)

Assignment 50% Midterm 20% Term project 30%

Matthes, E. (2019). Python crash course (2nd ed.). San Francisco, CA: No Starch Press. Mitchell, R. (2018). Web Scraping with Python: Collecting Data from the Modern Web. O’Reilly. Silge, J., & Robinson, D. (2019). Text mining with R: A tidy approach. Retrieved from https://www.tidytextmining.com/

地點

Office: HK 222

時間

Thursday 12:00-14:00

聯絡方式

Office phone: 31540 Email: taoc@mail.nctu.edu.tw