1. Please refer to ′Assessment Brief′ (page 5 onwards) for the marking rubric. This needs to be followed. 2. A2 HR Employee Data needs to be used (does not need to be cleaned). 3. A2 Template is the format the report needs to be in. All graphs can be in the Appendix and referred to. Please use the report template to address the questions in the marking rubric. Report can be up to 12 pages. Appendix not included in the count but references are. 4. The types of Modelling to be used are KMeans and DBScan. 5. Week 4 webinar slides contain hints from the lecturer about what to consider (please review) 6. Week 3 & 4 lecture slides give an overview of the level of detail we have gone into 7. Please provide a word copy export as well as python code for PC (not MAc) so I can replicate in my own workbook. 8. From the lecturer: Hints for modelling part: – Should we include EmployID when modelling? –Should we include Resigned variable when modelling ? –Converting variables to numerical values ? –Data transformation prior to clustering is needed ? –All variables in modelling and/or just a subset ? –Parameter selection for K-Means and DBSCAN –Discuss any findings from the resulted clusters ….what is the cluster of interest, e.g., cluster with high ratio of turn-over ….any interesting findings e.g., based on the guiding question in case study 9. Also from the lecturer: In this course, weeks 3 and 4 only show the students to evaluate the clustering results based on different metrics like silhouette score etc. on the whole data set. Hence, it is fine to skip train test split in this Assessment 2 10. We do need to show visualizations of the modelling.
