FIT5147 Data Exploration and Visualisation

Monash University
FIT5147 Data Exploration and Visualisation
Semester 1, 2025
Data Exploration Project
Part 1: Data Exploration Project Proposal
Part 2: Data Exploration Project Report
You are asked to explore and analyse data about a topic of your choice. It is an individual assignment and
worth 35% of your total mark for FIT5147. Part 1 Project Proposal contributes 2% and Part 2 Project Report
contributes 33%.
Relevant Learning Outcome
● Perform exploratory data analysis using a range of visualisation tools.
Overview of the Assessment Tasks

1. Identify the project topic, some related questions that you want to address, and the data source(s)
   that you will be using to answer those questions.
2. Submit your Project Proposal (Part 1) in the Assessments section of Moodle in Week 3.
3. Discuss with your tutor in your Week 3 Applied Session (after the submission in Moodle) and wait
   for approval from your tutor before proceeding further. Do not seek approval from the lecturer.
4. Collect data and wrangle it into a suitable form for analysis using whatever tools you like (e.g., Excel,
   R, Python).
5. Explore the data visually to answer your original questions and/or to find other interesting insights
   using Tableau or R. The exploration must rely on visualisations and visual analysis, but can analytical
   methods or statistical analysis where appropriate.
6. Write a report detailing your findings and the methods that you used. This must include properly
   captioned figures demonstrating your visual analysis (i.e. your visualisations must be referred to
   correctly in your report).
7. The Project Report (Part 2) is due in Week 7.
   Read the rest of this document before deciding on your project topic, as the proposal is for the entire Data
   Exploration Project and Data Visualisation Project, which is the second major assignment of this unit. See
   the end of this document 代写FIT5147 Data Exploration and Visualisationfor an example proposal and potential data sources to get started. Be careful not
   to copy this proposal; it is an example proposal, not template text.
   Choosing a Topic and Data
   The choice of topic, data, and the questions you seek to answer should allow for interesting and detailed
   analysis in the Data Exploration Project (DEP) and the subsequent Data Visualisation Project (DVP, due at the
   end of semester), which involves presenting the findings from your DEP in a specifically designed narrative
   interactive visualisation format.
   Good questions are general and not linked to specific parts of the data, allowing for more open-ended and
   exploratory analysis. For instance, asking “Where is the safest part of the network?”is a good question that
   lets you explore various interpretations of how to link terms like “where” and “safest” to the data about a
   network, whereas “Which region has the lowest value of number-of-deaths?” is not a very good question as
   it is very specific to the data, is easy to answer with one visualisation and therefore limits the exploration
   and visualisation possibilities.
   It is strongly recommended that you avoid questions that are:
   ● too easy to answer (e.g., what is the correlation between x and y, what is the average value of z
   variable, what are the top/bottom N values), or
   ● too difficult to answer (the work would take longer than the time available in the unit), or
   ● not relevant to the unit (e.g., training a machine learning model), or
   ● are not possible to answer from the available data.
   Proposals with such questions will be rejected. If you are in doubt, talk to teaching staff during face-to-face
   teaching times or ask for confirmation on Ed.
   How do you know if you have appropriate data? This depends on your topic and questions. You should
   ensure your data is big enough, i.e., has enough breadth and depth to invite interesting exploration.
   Combining data from different data sources is an ideal way to help add to the originality of the topic. To
   encourage different visualisation techniques your data will likely have a mixture of different data types.
   Time series (whether this be aggregated or detailed, such as months and years, or milliseconds) may be
   useful for your topic, and spatial, relational or text based data add useful complexity. If in doubt, talk to
   teaching staff during face-to-face teaching times or in a consultation before the due date.
   The chosen topic should be topical and some of the data should be recently collected, ideally from the last
   two or three years. The data must be accessible to the teaching staff, so the use of open data is
   encouraged (see the list of suggested data sources at the end of this document). Use of closed or
   proprietary data is allowed as long as explicit permission for use in this assignment is granted by the
   original authors or copyright holders. If you have closed data, you must still make it available to your
   teaching staff to access, i.e., via a shared Google Drive.
   Avoid common topics. Common topics including COVID-19, Netflix, AirBnB, car accidents, crime, house
   sales, car sales, world cup soccer, or electric vehicle sales should be avoided. Topics similar to the proposal
   example at the end of this document, i.e., traffic accidents and poor weather, must also be avoided. If you
   do have personal motivation for any of these mentioned common topics, you will need to propose a
   completely new angle to exploring the theme through novel questions with a mixture of new data sources.
   It is highly recommended to discuss your intentions with the tutor of your Applied Session prior to the
   proposal submission to avoid immediate rejection of the proposal.
   Part 1: Project Proposal (2%)
   Write a one-page PDF document consisting of the following sections:
8. Project Title
   A descriptive title for your project.
9. Topic Introduction
   One paragraph introducing the topic. This should include why it is a topical subject (for example,
   has it been in the news recently), and who might benefit from the insights you seek from your
   questions.
10. Motivation
    One paragraph describing why you personally are motivated to study this topic.
11. Questions
    Three questions you wish to answer using the data.
12. Data source(s)
    Briefly describe the data source(s) you will use. This should include: URLs of data source(s) and a
    description for each source: what is the data about, what is the size of the data (e.g., number of
    rows, number of columns), the type of data (e.g., tabular, spatial, relational, or textual), the type of
    attributes (e.g., categorical, ordinal, etc.) and the temporal intervals and period (e.g., monthly
    between 2019 and 2023).
13. References
    The bibliographical details of any references you have cited in the previous sections.
    Include your full name, student ID, tutor names, and Applied Session class number. This can be in the
    document header or footer. There should be no cover page.
    Part 2: Data Exploration (33%)
    The report should have the following structure:
14. Introduction
    Topic detail, problem description, questions, and brief motivation.
15. Data Wrangling and Checking
    Description of the data and data sources with URLs of the data, the steps in data wrangling
    (including data cleaning and data transformations) and tools that you used. The data checking that
    you performed, errors that you found, your method and justification for how you corrected errors,
    and the tools that you used. A comprehensive checking process is expected to justify data
    correctness, even if the data set is believed to be clean.
16. Data Exploration
    Description of the data exploration process with details of the visualisations (including figures and
    descriptions of findings) and statistical tests (if applicable) you used, what you discovered, and what
    tools you used.
17. Conclusion
    Summary of what you learned from the data and how your data exploration process answered (or
    didn’t answer) your original questions.
18. Reflection
    Brief description of what lessons you learnt in this project and what you might have done differently
    in hindsight.
19. Bibliography
    Appropriate references and bibliography (this includes acknowledgements to online references or
    sources that have influenced your exploration) using either the APA or IEEE referencing system.
    Include your full name, student ID, tutor names, and Applied Session class number. This may be on a cover
    page, or in the header or footer of the first page.
    The written report should be not longer than 10 pages for all sections mentioned above, excluding cover
    page, table of contents and appendix. Your written report will be the sole basis for judging the quality of the
    data checking, data wrangling, data exploration, as well as the degree of difficulty. Thus, include sufficient
    information in the report. It should, for instance, contain images of visualisations used for exploration and
    the results of any statistical analysis. You should include any analysis that you carry out even if it is
    incomplete or inconclusive as it demonstrates that you have thoroughly explored the data set.
    If you wish to provide additional material, an Appendix of up to 5 pages may be added at the end of the
    document. However, the Appendix will not be marked. Therefore, you should only use it to provide
    supplementary material that is not essential to the report or the reader's understanding. Be sure to clearly
    title this section as Appendix.
    Marking Rubric
    Part 1: Project Proposal (2%)
    ● Completeness and Timeliness [1%]: All components of the Proposal are included and it is submitted
    on time.
    ● Suitability and Clarity [1%]: Motivation, Questions and Data Sources.
    Motivation: A well-formulated project description with detailed information; a compelling and worthwhile topic to
    explore and visualise as a real-world problem.
    Questions: Three well-crafted questions that can be clearly answered through data visualisations. Each question
    requires sophisticated analysis of relationships and patterns across multiple attributes and demonstrates potential for
    innovative visualisation approaches to reveal insights and complex patterns.
    Data Sources: A clear description of data sources and datasets, including justification for which questions you will
    answer with each. The data must be sufficiently large or complex to require exploration and analysis. All datasets must
    be easily available, with URLs provided. For private and proprietary data, evidence of permission and a link to the
    dataset must be provided.
    After submission you will meet with your tutor during the Week 3 Applied Session to discuss your Project
    Proposal, receive feedback and ideally approval to start. If your proposal is rejected, your tutor will specify
    the reasons and suggest areas for improvement. You will need to make these amendments to your proposal
    and get it approved by your tutor prior to commencing your project work.
    Part 2: Project Report (33%)
    Criteria Below 50% Pass (50%+) Credit (60%+) HD (80%+)
    Data Complexity,
    Wrangling, Checking
    and Cleaning (7%)
    Inappropriate checking,
    cleaning, or wrangling.
20. if no demonstration of
    data checking and
    cleaning.
    Appropriate data
    cleaning and checking.
    Demonstrated ability to
    get data into R or
    Tableau;
    Good choices and clear
    justifications for error
    checking, cleaning and
    transforming of
    non-tabular data (e.g.
    spatial, relational,
    textual); large datasets
    (observations or
    dimensions) and/or
    multiple data sets.
    Excellence in data
    processing
    demonstrated and
    documented. Evidence
    of significant complexity
    in the wrangling,
    cleaning,
    transformation, or data
    collection (e.g.
    scrapping).
    Data Visualisation and
    Design Choices (9%)
    No visualisations;
    unsuitable or poor
    choice of visualisations;
    pixelated / poor quality
    images or illegible
    visualisations.
21. if not using Tableau or
    R.
    Suitable visualisations,
    which are well
    presented, described,
    readable and
    interpretable.
    Visualisations are
    appropriate for the
    intended purpose;
    appropriate labeling of
    axes and visualisations;
    clear legends when
    needed; saliency of
    patterns and trends.
    Variety of high-quality,
    complex and/or creative
    visualisations with high
    attention to detail.
    Clearly justified design
    choices incl.
    visualisation idioms,
    choice of visual
    variables, layout and
    labelling.
    Analytical Methods and
    Interpretations of Data
    and Topic Questions
    (9%)
    Unsuitable analysis or
    misinterpretation of the
    data and topics
    questions.
22. if no data analysis is
    demonstrated.
    Demonstrated suitable
    analysis and
    interpretation of the
    data and topic
    questions.
    Analysis that is
    appropriate for the
    intended purpose;
    justification and
    explanation of the
    exploration process and
    use of statistical
    measures; identification
    of trends, patterns, and
    insights.
    High quality of visual
    analysis demonstrated.
    Sophisticated and
    correctly used analytical
    methods such as
    clustering;
    dimensionality
    reduction; sophisticated
    aggregation and/or
    filtering; non-linear
    model fitting; correct
    use of statistical tests;
    or complex time series
    analysis.
    Written Report: Quality
    and Completeness (8%)
    Poor report, or missing
    sections.
    Good report with logical
    structure with all the
    expected sections:
    Introduction, Data
    Wrangling, Data
    Checking, Data
    Exploration, Conclusion,
    Reflection, Bibliography.
    Referencing of sources,
    figures and tables.
    Correct grammar and
    spelling.
    High quality of writing
    and figures/images with
    minimal errors. Correct
    referencing of figures
    and tables within the
    text, and correctly used
    academic referencing of
    sources.
    Professional report with
    excellence of writing
    combined with high
    quality figures/images.
    Clearly articulated
    findings; awareness of
    limitations; deep
    exploration; thorough
    conclusions.
    Originality
    Since this is academic work, it must be original and clearly distinguish between your own contributions and
    those based on other’s work. If you include data, facts, opinions or any other written or graphical
    information from another source, you must cite and reference it according to the APA or IEEE style guide.
    This includes third-party programming code, software used in data exploration and analysis, and any
    definitions or descriptions of concepts or software. Direct quotations or reproductions must adhere to the
    appropriate APA or IEEE style.
    In your report you are encouraged to repeat the questions from your proposal. This is the only
    self-plagiarism that is allowed. If you are retaking this unit from a previous semester, you must choose a
    completely new topic and dataset. The topic and dataset cannot have been used in any other unit. You may
    not reuse any code or written content from previous assessment tasks for any unit. Additionally, content
    from previous assignments or sample reports cannot be used.
    You may use Generative AI tools, such as ChatGPT, to improve writing and expression. However, your writing
    must be logically structured, clear and concise. Repetitive, poorly structured, or vague gibberish as often
    generated by Generative AI tools will result in a low grade. AI is generally unsuitable for data checking,
    cleaning, wrangling, exploration and visualisation of this level and should be avoided. It is important to
    remember that generated content can be biased. Any use of Generative AI in the preparation of your
    assessment must be acknowledged at the end of your submitted document.
    If concerns arise regarding the originality of your work – whether due to plagiarism, collusion, contract
    cheating, or the use of unapproved software – your academic integrity will be reviewed. Confirmed
    breaches of academic integrity may result in penalties affecting your assignment mark, this unit, or even
    your enrolment.
    Submission and Due Dates
    Once you have completed your work, take the following steps to submit your work.
23. Save your proposal or report as a PDF document.
24. Name your file using the following structure: Proposal_Surname_StudentID.pdf or
    DEP_Surname_StudentID.pdf
25. Submit and upload your document.
    ● Project Proposal: Submit a one-page PDF in Week 3.
    ● Project Report: Submit a 10-page PDF (excluding cover page and appendix) in Week 7.
    See Moodle for dates and times.
    Your assignment must show a status of ”Submitted for grading” before it can be marked. Any submission in
    “Draft” mode will not be marked.
    Late Submissions
    ● There will be zero marks for late Project Proposal submissions. Everyone must submit the Project
    Proposal. Even if the deadline has passed, you must still submit a proposal (with a grade of 0) as
    your project must be approved before you can continue working on the Data Exploration Project.
    The proposal is a hurdle requirement. If it is not submitted and approved by your tutor, the mark for
    the Data Exploration Project is 0.
    下面这一部分全在说原创性
    ● For the Project Report, submissions received after the deadline (or after an extended deadline for
    those with an extension or special consideration) will be penalised at 5% of the total available
    mark [33%] per calendar day up to a maximum of 7 days. If submitted after 7 days, it will receive
    zero marks and no feedback will be provided.
    ● For further information on eligibility for Extensions or Special Consideration, see:
    https://www.monash.edu/students/admin/assessments/extensions-...
    Example Data Sources
    The following is a list of data sources to get started. Feel free to use these as a source of inspiration and
    ideas for your project. You are not limited to the data sources listed below.
    ● Data search tools and repositories, e.g.:
    ○ Google dataset search: https://toolbox.google.com/datasetsearch
    ○ Google Trends: https://www.google.com/trends/explore
    ○ Google Ngram Viewer: https://books.google.com/ngrams
    ○ Registry of Open Data on AWS: https://registry.opendata.aws/
    ○ Kaggle: https://www.kaggle.com Note that using data from Kaggle exclusively is not
    acceptable, you must use at least one additional data source.
    ○ Science Hack Day: http://sciencehackday.pbworks.com/w/page/24500475/Datasets
    ● Open local and national government data portals, e.g.:
    ○ Victorian Government Data: http://data.vic.gov.au/
    ○ Australian Government Data: http://data.gov.au/
    ○ National Map: https://nationalmap.gov.au/ (Australian data)
    ○ Australian Bureau of Statistics: https://www.abs.gov.au/statistics
    ○ Atlas of Living Australia https://ala.org.au/
    ○ European Union Open Data: https://data.europa.eu/en
    ○ UK Government Open Data: https://data.gov.uk/
    ○ U.S. Government Open Data: https://www.data.gov/
    ● Humanitarian data sources, e.g.:
    ○ UNdata: http://data.un.org/
    ○ The World Bank Data Catalog: https://datacatalog.worldbank.org/
    ○ Our World in Data: https://ourworldindata.org/
    ○ Berkeley Library Health Statistics:
    http://guides.lib.berkeley.edu/publichealth/healthstatistics/...
    ● Open corporate/industry data, e.g.:
    ○ Uber: https://movement.uber.com/?lang=en-AU
    ○ Inside Airbnb: http://insideairbnb.com/get-the-data.html
    Example Project Proposal
    Please note this mock example is relatively old now. We expect your data to ideally include recent data, i.e.,
    data from 2022, 2023 or even 2024. It is possible to complete this example project with only Data Source A
    and B, but C provides different opportunities and additional difficulty when doing the exploration and
    visualisations. If done well, this added depth and difficulty can gain extra marks but might take longer to
    complete. The student could use both datasets A and B to identify temporal aspects in the data, such as
    accidents near to sunset and sunrise across the whole dataset, but dataset C allows them to identify areas
    which are poorly lit and see if this correlates with the spatial pattern of pre-sunrise and post-sunset
    accidents. Furthermore, whilst Data Sources A and C are currently tabular data, they can be converted to
    spatial features and spatial analysis can be carried out.
    Name: Jesse van Dijk, Student ID: 12345678, Teaching Associate: Jo Bloggs & Alex Smith, Applied 01.
    Project Title: Causes of Serious Bicycle Accidents in Canberra
    Introduction
    Recent media and industry reports indicate that Australian roads are becoming even more dangerous for cyclists
    [1,2]. I believe this is an important topic for many audiences such as cyclists, road safety officers, and public
    health policy makers. Therefore I want to find out more about the factors that affect bicycle accidents in
    Canberra.
    Motivation
    I am a keen cyclist and am concerned about cycling in Australia. I have recently moved to Canberra from the
    Netherlands where cycling is very safe and accidents linked with road vehicles is unusual. I have noticed it is
    difficult to see during sunset on a number of roads and would like to see if this pattern is evident in the data.
    Questions
26. What are the most common kinds of serious bicycle accidents in Canberra, and how do these vary over
    different time periods (e.g. hour of day/day of week/month/season)?
27. How do lighting conditions affect these accidents?
    Data sources
    A. ACT Road Cyclist Crashes 2012 to 2021, which have been reported by the Police or the Public through
    the AFP Crash Report Form. This data is tabular data: ~1K rows × 11 columns. It has both spatial and
    temporal attributes including the geographical (latitude and longitude) location and a datetime stamp
    for the time of accident. Some numerical and simple text attributes relating to the incident. i.e. number
    of casualties, description of accident, including direction of traffic.

B. Canberra’s sunrise and sunset times, 2012 to 2021. Tabular data in HTML: ~365 rows × 4 columns for
each year to be scrapped from sunrise website. Columns are simply date, time of sunrise, time of sunset
and hours of daylight.

C. ACT Streetlights, 2021. Tabular data in CSV format with ~80K rows × 10 columns. These include latitude
and longitude for the streetlight location and various text columns including lamp type, Luminaire,
height and street and suburb name. There is no date column for the age of the lamp, but the source of
the data is dated from 2017 and was last updated in Nov 2021.

Data Source A will be used to address Question 1, whilst A to C will allow me to answer Question 2.

WX：codinghelp