Group Assignment: Big Data Analytics $800$

Course:

Level:

Group Size:

Big Data Analytics $800$

NQF Level $8$

As assigned by the administrator

Objective:

To apply the concepts of Big Data Analytics, including data acquisition, data processing,

analysis, and visualisation, using a real$-$world dataset from Kaggle.

Students are expected to work through the entire Data Analytics Life Cycle and provide

actionable insights based on their analysis.

Assignment Instructions:

$1.$ Dataset Selection:

o Visit Kaggle and select a dataset relevant to a specific industry $($e$.$g$,$ healthcare,

finance, retail, social media$).$

o Ensure the dataset is comprehensive and includes various data types $-$ structured,

unstructured, and semi$-$structured $($IF APPLICABLE$).$

$2.$ Data Analytics Life Cycle:

o Follow the stages of the Data Analytics Life Cycle to analyse the selected dataset:

Business Case Evaluation:

$$ Define the business problem or opportunity you aim to address with

your analysis.

$$ Outline the objectives and potential benefits of the analysis.

Data Identification:

$$ Identify all relevant data sources within the selected dataset.

Data Acquisition & Filtering:

$$ Collect the necessary data and filter out any irrelevant information.

Data Extraction:

$$ Convert the data into a format suitable for analysis.

$3.$ Data Validation & Cleansing:

o Ensure the data is accurate and clean by detecting and correcting errors.

$4.$ Data Aggregation & Representation:

o Combine data from different sources to create a unified dataset.

$5.$ Data Analysis:

o Apply statistical and machine learning techniques to derive insights from the

data.

$6.$ Data Visualisation:

o Present your findings using visual tools like charts and graphs.

$7.$ Utilisation of Analysis Results:

o Discuss how the insights can be applied to solve the business problem or

improve processes.

$8.$ Concepts Application:

o Utilize some of the following concepts in your analysis and provide detailed

explanation:

$$ Big Data Characteristics: Volume, Velocity, Variety, Veracity, Value.

$$ Types of Analytics: Descriptive, Predictive, Prescriptive, Diagnostic.

$$ Big Data Technologies: Use tools like R$,$ Apache Spark, MongoDB,

etc.

$$ Machine Learning Algorithms: Apply algorithms like regression

analysis, clustering, collaborative filtering, and association rule mining.

$$ Data Visualisation and Interpretation: Visualize your results using

tools like Tableau, Power BI$,$ or D$3.$js$.$

$9.$ Report

o Prepare a detailed report $($minimum $20$ pages$)$ covering all the stages of your

analysis, methodologies used, and insights gained.

Your report should include:

$$ Executive Summary

$$ Introduction to the Business Problem

$$ Data Description and Pre$-$processing Steps

$$ Analytical Methods and Techniques Used

$$ Key Findings and Insights

$$ Visualisations

$$ Conclusion and Recommendations

o Create a presentation $(15-20$ slides$)$ summarizing your findings and

recommendations. Convert the presentation to PDF and submit it together with

your report.

$10.$ Submission:

o Submit the report, presentation, link to the data files, and scripts used for

analysis on one PDF$.$

Evaluation Criteria:

$$ Problem Definition and Business Case $(15\%)$

$$ Data Preprocessing and Cleaning $(20\%)$

$$ Application of Analytical Techniques $(20\%)$

$$ Insights and Recommendations $(25\%)$

$$ Visualization and Interpretation $(10\%)$

$$ Report Quality and Presentation $(10\%)$

Additional Notes:

$$ Collaboration: All group members are expected to contribute equally. Keep a log of

individual contributions to be submitted along with the final report.

$$ Originality: Ensure all work is original and cite any external sources or references

properly.

$$ Tools and Resources: You are encouraged to use various data analytics tools and

programming languages such as Python, R$,$ SQL$,$ and visualization tools.

Example Datasets on Kaggle:

$$ Healthcare: Heart Disease Dataset

$$ Finance: Credit Card Fraud Detection

$$ Retail: Online Retail Dataset

$$ Social Media: Twitter Sentiment Analysis