BRM5420 Livestock Production Science Coursework Brief 2025/26 | Aberystwyth University

BRM5420 Coursework Brief

Learning Outcomes:

Learning Outcomes to be addressed by the coursework:

1.Review and evaluate scientific literature and apply the results to the development of livestock production systems. 
Explain how research can be used to increase the efficiency of farm animal production

Clarification of coursework brief

It is your responsibility to ensure that you understand the requirements of the coursework that has been set. AU staff policy is to reply to emails within three working days so you should not expect a response in time to help if you ask for guidance less than three days before the deadline.   

University policy on late/non-submission

Work submitted after the deadline without an extension will be given a mark of zero.

Extension to coursework deadlines
 
Extensions may be awarded in exceptional circumstances, e.g. on medical grounds, or for genuine, unforeseen personal circumstances such as illness, family problems or death of a relative or close personal friend.  You may apply for an extension using the Extension Request Form available to download from the ‘Undergraduate Information’ or ‘Information for Taught PG Students’ module on Blackboard. This should be submitted together with any supporting documentary evidence such as a medical certificate to fsestaff@aber.ac.uk . In order to be awarded an extension, you should submit the Extension Request Form (with appropriate evidence) at least three working days before the deadline. If insufficient evidence is submitted, then the extension may be awarded subject to further evidence being provided by a given date. 

Failure of computers, printers and backups etc. will not be accepted as reasons for late submission of assessed work. 
If an extension is granted you should follow the advice on submission provided by FELS Extension team. 
Please refer to your handbook for further information on Coursework Extensions.
 
If an extension has not been confirmed, then you should always submit an attempt of the work by the deadline.

Do You Need Help with BRM5420 Livestock Production Science Coursework?

Chat With Experts

Machine learning modelling of cattle behaviour

AIMS

The aims of this assignment are to: 

1.Develop your skills in using a programming language.
2.Develop your skills in producing machine learning models of livestock behaviour data.
3.Develop your skills in academic manuscript writing.

DATASET

You have been provided with a dataset that has been collected and processed using standard data preparation techniques such as those discussed in your lectures. The data is ready for modelling with machine learning (ML) algorithms.

BACKGROUND TO THE DATA

• Data collected on 5 dairy cows while grazing in a single group (each cow contributed one day of data to the whole dataset). 
• One 3-axis accelerometer (Axivity AX3) set to record at 12Hz was used to collect the accelerometer data (g-force values) from each cow (accelerometer fitted to a neck collar).
• A video camera was used to record the behaviours of the cows during the experiment.

DATA PROCESSING

• All data were downloaded from the accelerometers and labelled with three behaviour classes: 
  oGrazing (G) – Cow has its head down, is either walking or stationary and is actively ingesting grass.
  oResting (R) – Cow is either standing (may be ruminating) or lying down (may be ruminating) or sleeping while lying down.
  oWalking (W) – Cow is moving purposefully forward without its muzzle on the ground.
• Raw data (X, Y and Z axes) were used to create an additional two new variables (Movement Intensity (MI) and Signal Magnitude Area (SMA)
• Total variables are: AccX, AccY, AccZ, MI, SMA
• Data were aggregated into 5 second windows (epochs) and 20 features were generated based on the mean, minimum, maximum and standard error of the features across every 5s window.
• A feature importance exercise was undertaken leaving you with the 10 most important features for cow behaviour classification.

YOUR TASK

• To analyse the data using ML algorithms
• To write a manuscript in preparation for the Journal Computers and Electronics in Agriculture (https://www.sciencedirect.com/journal/computers-and-electronics-in-agriculture). Your manuscript will be based on the analyses that you undertake on the dataset provided to you. Your manuscript must be prepared in accordance with REVIEWED SUBMISSIONS format of the journal that can be found here: https://www.sciencedirect.com/journal/computers-and-electronics-in-agriculture/publish/guide-for-authors. 
• For ease of preparation, a journal manuscript template is provided to you on Blackboard with the assignment materials.

NOTES

1.You must undertake your analysis using R Studio software.
2.You should report on the results of THREE ML algorithms in your manuscript.
3.You should report results using tables and figures where appropriate.
4.You should include the code you used in your work in an Appendix.
5.Some of the information on the methods have been provided to you but you must include additional information where necessary based on your understanding of how such methods would be undertaken (plug the gaps where necessary).
6.The word limit for the manuscript is 2,000 words. Please note that although standard manuscripts are between 6,000-8,000 words, you must adhere to the word count provided to you for this task.
7.The paper should be written according to the conventions of the journal Computers and Electronics in Agriculture using the template provided on Blackboard.
8.The following skills will be assessed: 
a.Layout of the manuscript according to the conventions
b.Written communication 
c.Understanding of the background research 
d.Delivery of the research results
e.Discussion of the results within the context of the work undertaken
f.Limitations of the work
g.Code used in preparation of the models
9.The marking criteria is provided overleaf.

BRM5420 Assessment Criteria: