Reference no: EM133543414 , Length: word count:2000
Assessment: Project and presentation (Solving a real-world analytics problem)
Assessment details
Purpose: This assessment requires students, based on the topics covered in during the term, to perform advanced business analytics on a chosen data set and write a structured report using the format specified below..
Description: Students will work in groups to prepare a group report demonstrating the competent application of business analytic techniques. Groups will be formed in class early in the term.
Question: Business analytics is the process of performing autonomous or semi- autonomous insight extractions from real-world data using sophisticated techniques and tools. Through this assessment, students are expected to work collaboratively in utilising sophisticated techniques and tools to derive facts for driving decision-making based on complex, real world data that will create value for business stakeholders.
In addition, students will be required to respond individually to questions posed after the group presentation in week 11. Assessment 3 is designed to test students' ability to apply concepts, tools and techniques covered during the term's learning sessions to ensure they achieve the required learning outcomes.
Step 1: Explore the available datasets in the following websites, and select one that seems interesting to you and your group members:
• Datasets (Kaggle n.d.).
• UCI Machine learning repository (UCI n.d.).
• Datasets for data mining, data science, and machine learning (KDNuggets 2022)
Step 2: Send the name of the selected dataset and the names of students in the group to the lecturer for approval. (Time will be spent in class in week three for the formation of groups.
Step 2A: Working collaboratively with your group members, discuss the choices involved in Step 4. Devise your timeline for the completion of the project tasks and written report and assign tasks to each group member. Document the assigned tasks and their completion during semester and attach this document as an appendix to your written report.
Step 3: Go through the stages of data pre-processing to make your dataset ready to be analysed.
Step 4: Choose between the following models:
a. Build a simple classifier and apply to your dataset (decision tree)
b.Cluster analysis (K-means)
c. Sentiment analysis
d.Topic detection analysis
Note: Your report should explain why you chose your particular model.
Step 5: Define any built-in assumptions made by the technique about the data (e.g., quality, format, distribution). Compare these assumptions with those in the Data Description Report. Make sure that these assumptions hold and step back to the Data Preparation Phase if necessary.
Step 6: Task: Build a model. Run the modelling tool on the prepared dataset to create one or more models. (Using Knime Tool as demonstrated in class in the lab sessions).
Step 7: State your conclusions regarding patterns in the data (if any); sometimes the model reveals important facts about the data without a separate Assessment process,
(e.g., that the output or conclusion is duplicated in one of the inputs).
Step 8: Assess the extent to which the model meets the business objectives and seek to determine if there is some business reason why the model is deficient.
Compare results with the evaluation criteria defined at the start of the project.
Step 9: In collaboration with your group members, assign responsibility and deadlines for writing sections of the written report.
Your report must include the following sections:
a. Cover page and table of contents
b.Executive Summary
c. Introduction
d.Data pre-processing and feature extraction
e. Experiment
f. Result analysis
g. Conclusion
Step 10: Submit your report before the due date (20%), present your group report in week 11 (10%) and participate in the structured Q & A session that follows your presentation (10%).
Research expectation
• The submission needs to be supported with information obtained from credible sources.
• Credible sources should be varied and include, but not limited to, the Textbook, Government reports, Industry reports, Newspaper articles, Books, and Journal articles
• Use Harvard referencing, including the reference list