Module SGIA40K15: Natural Language Processing for Public Policy

Department: Government and International Affairs

SGIA40K15: Natural Language Processing for Public Policy

Prerequisites

• None

Corequisites

• SGIA49915 Quantitative Methods and Analysis OR SGIA40G15 Full Stack Research Design

Excluded Combination of Modules

• None

Aims

• To furnish students with the skills to apply natural language processing methods (methods of collecting, pre-processing and analysing text data) to answer empirical questions in public policy.
• To furnish students with the skills to communicate results from text analysis to a policy audience.

Content

• Indicative content may include:
• Use of web scraping and API to collect text data
• Methods of pre-processing text data
• Unsupervised learning methods
• Supervised learning methods
• Language models
• How to use natural language processing methods to:
• relate policy documents to the performance of policy actors
• study how policy documents are used
• investigate public sentiment towards policy actors
• analyse the changing meaning of key terms in public policy

Learning Outcomes

• Advanced knowledge of methods of collecting text data;
• Advanced knowledge of methods to pre-process text data to make it suitable for analysis;
• Advanced knowledge of natural language processing techniques, including how to appropriately use these methods to answer empirical questions in public policy, and the strengths and limitations of these methods.

• Apply methods of collecting, pre-processing and analysing text data to answer an empirical question in public policy;
• The ability to effectively communicate the results of natural language processing techniques to a policy audience.

• Research skills: the ability to do independent research, including the ability to design research, execute research, and critically evaluate its strengths and weaknesses.
• Communication skills: the ability to effectively communicate results from research.
• The ability to use data analysis software, including the ability to autonomously trouble-shoot using online resources.

Modes of Teaching, Learning and Assessment and how these contribute to the learning outcomes of the module

• Teaching and learning are through a series of 1-hour lectures and computer labs.
• Lectures will familiarise students with methods of collecting, pre-processing and analysing text data, and furnish students with examples of the use of these methods in public policy.
• Computer labs will allow students to practice applying methods of collecting, pre-processing and analysing text data to the field of public policy, and to work on and receive feedback on their formative assessment.
• Summative assessment is a 3,000-word policy briefing (accounting for 100% of the final grade), which must apply natural language processing techniques to answer an empirical question in public policy.
• Formative assessment is a 750-word outline for the policy briefing, allowing students to receive feedback on their plans for the briefing.

Teaching Methods and Learning Hours

Summative Assessment

Formative Assessment:

Formative assessment is a 750-word outline for the policy briefing.

■ Attendance at all activities marked with this symbol will be monitored. Students who fail to attend these activities, or to complete the summative or formative assessment specified above, will be subject to the procedures defined in the University's General Regulation V, and may be required to leave the University