Background
In later 2019 and early 2020, Australia faced devastating bushfires started in late 2019, which swiftly got worse
efore rains helped contain many of the worst fires in Fe
uary 2020. Bush fires are major environmental issues,
creating economic and ecological damages. It is reported that Australia's catastrophic bushfire crisis has
destroyed thousands of homes, burned millions hectares of forest, and taken an enormous toll on wildlife.
Therefore, fast and automatic detection of bushfires at an early stage is crucial for a successful firefighting.
Traditional human surveillance is expensive and inefficient, which can also be affected by subject factors. With
the advances in information technologies, a variety of data about the forest can be collected, such as remote
images collected by satellites and meteorological data collected by local sensors. The collected data contains
ich information about the status of the forest, the analysis of which can help us detect potential bushfires so as
to make effective and efficient firefighting plan and then minimize the damage caused by the bushfires.
In this project you are given a bushfire dataset and an article that uses this dataset. The authors have developed
several ML models for predicting the burned area of a bushfire and compared their performance. You must read
the article to understand the problem, the dataset, and the methodology to complete the following tasks.
Dataset
The dataset contains 517 fire instances, each of which have 13 columns: the first 12 columns co
esponding to
the attributes (e.g., spatial coordinates, month, day, four fire indices, and other meteorological data) and the last
column containing the burned area, i.e., the variable that we will predict. The details of the dataset can be found
in the original research paper. The dataset files are stored in UCI's website below (click the hyper-link to
download the data) Forest fires data : There are two files on the website. One called “forest-fires.csv” contains
the data needed for the analysis, and another called “forestfires.names" contains the information about the
dataset.
Tasks:
1. Read the article and reproduce the RMSE results presented in Table 3 using Python modules and packages
(including your own script or customised codes). Write a report summarising the dataset, used ML methods,
experiment protocol and results including variations, if any. During reproducing the results: XXXXXXXXXXMarks)
i) you should use the same set of features used by the authors.
ii) you should use the same classifier with exact parameter values.
iii) you should use the same training/test splitting approach as used by the authors.
iv) you should use the same pre/post processing, if any, used by the authors.
v) you should report the same performance metric (RMSE) as shown in Table 3.
N.B.
(i) If some of the ML methods are not covered in the cu
ent unit. Consider them as HD tasks i.e., based on the
knowledge gained in the unit you should be able to find necessary packages and modules to reproduce the results.
(ii) If you find any issue in reproducing results or some subtle variations are found due to implementation
differences of packages and modules in Python then appropriate explanation of them will be considered during
evaluation of your submission.
(iii) Similarly, variation in results due to randomness of data splitting will also be considered during evaluation
ased on your explanation.
(iii) Obtained marks will be proportional to the number of ML methods that you will report in your submission
with co
ectly reproduced results.
http:
www3.dsi.uminho.pt/pcortez/fires.pdf
https:
archive.ics.uci.edu/ml/machine-learning-databases/forest-fires
(iv) Make sure your Python/R code segment generates the reported results, otherwise you will receive zero marks
for this task.
Marking criteria:
i) Unsatisfactory (x<6): tried to implement the methods but unable to follow the approach presented in
the article. Variation of marks in this group will depend on the quality of report.
ii) Fair (6<=x<9): appropriately implemented 50% of the methods presented in the article. Variation of
marks in this group will depend on the quality of report.
iii) Good (9<=x<12): appropriately implemented 70% of the methods presented in the article. Variation
of marks in this group will depend on the quality of report.
iv) Excellent(x>=12): appropriately implemented >=90% of the methods presented in the article.
Variation of marks in this group will depend on the quality of report.
2. Design and develop your own ML solution for this problem. The proposed solution should be different from
all approaches mentioned in the provided article. This does not mean that you must have to choose a new ML
algorithm. You can develop a novel solution by changing the feature selection approach or parameter
optimisations process of used ML methods or using different ML methods or adding regularization or different
combinations of them. This means, the proposed system should be substantially different from the methods
presented in the article but not limited to only change of ML methods. Compare the RMSE result with reported
methods in the article. Write in your report summarising your solution design and outcomes. The report should
include: XXXXXXXXXX20 Marks)
i) Motivation behind the proposed solution.
ii) How the proposed solution is different from existing ones.
iii) Detail description of the model including all parameters so that any reader can implement your model.
iv) Description of experimental protocol.
v) Evaluation metrics.
vi) Present results using tables and graphs.
vii) Compare and discuss results with respect to existing literatures.
viii) Appropriate references (IEEE numbered).
N.B. This is a HD (High Distinction) level question. Those students who target HD grade should answer this
question (including answering all the above questions). For others, this question is an option. This question aims
to demonstrate your expertise in the subject area and the ability to do your own research in the related area.
Marking criteria:
(i) Unsatisfactory (<10): an appropriate solution presented whose performance is lower than the reported
performances in the article (Table 3). The variation in the marking in this group will depend on the quality of the
eport.
(i) Fair (10 - <14): an appropriate solution presented whose performance is at least equal with the lowest
performance reported in the article (Table 3). The variation in the marking in this group will depend on the
quality of the report.
(ii) Good (>=14): an appropriate solution presented whose performance is better than the best reported
performances in the article (Table 3). The variation in the marking in this group will depend on the quality of the
eport.
.
https:
www.bath.ac.uk/publications/li
ary-guides-to-citing-referencing/attachments/ieee-style-guide.pdf