Programming Tools for Urban Analytics Summative Assessment
Lecturer in Urban Analytics
1. Overview The aim of this course is to familiarize students with programming tools which allow them to access, collect, manage, analyze, visualize, and understand urban big data efficiently and effectively. It will cover techniques required to extract data from on-line Application Protocol Interfaces (APIs), set up a database for holding data in a way which enables efficient analysis, and statistic/machine learning and visualization tools.
It will cover best practices in relation to coding (Python, SQL/NoSQL queries), collaborating on coding projects (Unix Shell, Git, and GitHub), and reproducibility of analyses (Jupyter Lab/Notebook).
Students will undertake a data science project which requires them to demonstrate the skills which they have acquired during the course.
You will be required to submit a single Jupyter Notebook and a HTML file generated from the Jupyter Notebook including all your Python code and markdown-style written report, with a maximum of 3,000 words (not including code and reference). You will need to submit your final assignment through Moodle.
2. Content and Format The final data science report should broadly follow the style of a quantitative journal article, with the exception that you should focus on the data analysis and explanation of your data analysis. It is not necessary to include a detailed literature review, though you may choose to cite papers to support some of your choices e.g., your research question, your choice of variable, the assumptions you make, and so on.
The data science report should outline what your research question is and what data you will use to address it.
You will analyze your data by using the tools and packages you have learnt in the classroom though using extra Python packages to achieve your project goals is highly favorable.
Your analysis should include:
• Research questions and project objectives with the support of academic literature
• Data collection methods, either through API, online scraping, or explaining the data sources
• Understanding your data (data types, summary statistics, data visualization, etc.)
• Data cleaning (missing values, outliers, date/time transformation, data errors, etc.)
• Feature engineering (categorical variables to dummy variables, normalization/standardization, feature combination, etc.)
• Data analysis (time series analysis, machine learning, spatial analysis, advanced regression analysis, etc.) • A summary of your findings and suggestion from your data analysis
Students are required to keep to within an additional 10 percent of the word limit given for an assignment – there are penalties on assignments that are longer than this. Submissions that go 10-14% over the word limit on an assessment will be subject to a 1 point deduction; 15-19% over a 2 points deduction; 20-24% over a 3 points deduction and 25% or more over will be awarded a fail (zero) and required to resubmit as a second attempt.
3. Marking As usual, the purpose of the assignment is to give students the opportunity to demonstrate
their learning in relation to the course’s intended learning outcomes. The outcomes of this course are:
• Set up, connect to and query a simple relational and non-relational database by Python
• Retrieve and analyze data from an Application Protocol Interface (API)
• Perform basic machine learning tasks
• Write code according to best practice and produce tidy data
• Collaborate effectively with other analysts using appropriate tools
• Produce documentation for their work which makes the processes behind analyses transparent and reproducible.
For more information about assessment at the University of Glasgow, please consult the Code of Assessment.
4. Getting help There is time available during the last class lecture sessions for you to ask any related questions for your final assessment.
An essay proposal or outline (maximum 1-page, formative assessment) if you would like to get feedback from your instructor.
This should identify the data science project you want to do for this course, and outline the objectives, programming tools you want to use, and expected outcomes.
You can also post any questions on the assignment forum on Moodle or email the instructor if you have specific questions.
