Big Data Analytics Applications CSIP5203

Module name:

Big Data Analytics Applications

Module code:

CSIP5203

Title of the Assessment:

Assignment 2: Big Data Analytics

This coursework item is:

Summative

This summative coursework will be marked anonymously:

Yes

The learning outcomes that are assessed by this coursework are:

1. Demonstrate self-direction and originality in analysing vast amounts of unstructured data using massively parallel and scalable cloud computation systems.

2. Critically evaluate data mining and machine learning algorithms using a large-scale data analytics platform.

3. Research into an area of big data ethics, integrate complex and sometimes conflicting ideas into a coherent analysis that demonstrates integrative, synoptic, and analytical skills.

This coursework is:

Individual

If other or mixed ... explain here:

This coursework constitutes 60 % of the overall module mark.

Date Set:

Date & Time Due (the deadline):

31^st January 2025 at 12.00 noon

In accordance with the University Assessment and Feedback Policy, your marked coursework and feedback will be available to you on:

21^st February 2024

You should normally receive feedback on your coursework no later than 15

University working days after the formal hand-in date, provided that you have met the submission deadline

If for any reason this is not forthcoming by the due date your module leader will let you know why and when it can be expected. The Associate Professor Student should be informed of any issues relating to the return of marked coursework and feedback.

When completed you are required to submit your coursework via:

1. LearningZone Assignment Link

Late submission of coursework policy:

Late submissions will be processed in accordance with current Please check the regulations carefully to determine what late submission period is allowed for your programme.

Academic Offences and Bad Academic Practices:

Deliverables to be submitted for assessment:

One technical report file containing two tasks work.

One Juptyer Python notebook code file as a separate file for both tasks through the LearningZone platform. This file must include the full implementation of the code used to generate the outputs presented in the report.

How the work will be marked:

Marks will be awarded according to the marking grids below, considering the distribution of marks outlined in the body of the coursework brief.

Marking will be based on the report alone.

Failure to submit the code file will result in a zero mark for this assessment, as the evidence of outputs cannot be validated.

Criteria

Fail 0-9%

Fail

10-19%

Fail 20-29%

Fail 30-39%

Fail 40-49%

Pass 50-59%

Merit 60-69%

Distinction 70-79%

Distinction 80-89%

Distinction 90-100%

Presentation (5%):

- Appropriate structure, writing style, etc.

- Code file submitted separately to validate outputs in the report.

Code file missing, unstructured report, and unclear or inconsistent academic style.

Barely recognisable structure; significant

deficiencies.

Minimal structure and very poor formatting.

Disorganised structure with some effort toward

formatting.

Weak structure and unclear academic style. Formatting errors evident.

Basic structure and style;

noticeable formatting inconsistencies.

Code file submitted but with issues or partially aligns with the report.

Clear presentation;

minor formatting or stylistic weaknesses.

Code file submitted but with issues or partially aligns with the report

Well-structured and coherent, with strong adherence to

academic norms.

Code file submitted, wellstructured, and aligns

seamlessly with the outputs and discussions in the report.

Highly

professional and clear presentation with minor inconsistencies.

Code file submitted, wellstructured, and aligns

seamlessly with the outputs and discussions in

the report.

Exceptional clarity, structure, and academic style.

Impeccable adherence to formatting

guidelines. Code file submitted, wellstructured, and aligns seamlessly with the outputs and discussions in the report.

Introduction of dataset (5%): - Introduce your dataset e.g. mention its source, what it is, the period it covers, what it represents and so on

No attempt to describe the dataset.

No meaningful dataset description; irrelevant or erroneous.

Inadequate or incorrect dataset description; lacks relevance.

Minimal dataset information;

limited relevance or accuracy.

Weak description with errors or missing critical details.

Basic description; lacks depth or detailed

connection to the task.

Clear dataset description with adequate relevance.

Accurate dataset description with appropriate

detail and relevance.

Detailed description of dataset with strong

connection to the task.

Comprehensive and insightful description of dataset, including origin, coverage, and significance.

Appropriate annotation of code(5%):

- Annotate your program, and if relevant, you can use snippets from it in your report.

No annotations provided.

Barely any annotations; hinders code comprehension.

Very weak annotations, impeding understanding.

Minimal or inconsistent annotations.

Limited annotations reducing clarity.

Basic annotations; some parts lack explanations.

Adequate annotations, though some areas lack depth or clarity.

Code is wellannotated, enhancing clarity.

Clear and consistent annotations with minimal omissions.

Comprehensive and insightful annotations enhancing understanding and usability.

Depth, breadth and variety of data analytics (15%):

- Data analysed appropriately, in accordance with the chosen method, demonstrating depth and breath

No analytics performed.

Negligible or incoherent analytics.

Very weak analytics; results are superficial and poorly developed.

Minimal analytics performed; lacks critical thinking.

Limited analytical approaches, lacking depth or critical insights.

Basic analytics presented but lacks depth or variety.

Adequate depth with minor gaps or limited variety.

Good analytical depth, demonstrating critical reflection and some originality.

Strong analytical depth with minor omissions; demonstrates

critical thinking.

Exceptional analytical depth and variety, demonstrating innovative and insightful approaches.

Data analytics (types and representations of analytics) (15%):

- Ensure that the analytics done are clearly outlined and discussed both in text form and visual representations

No representations or relevant text provided.

Barely any visualisation or meaningful explanation.

Very weak or irrelevant representations.

Minimal effort in visualisation or text explanation.

Poor

representation; visuals/text are unclear or lack insight.

Basic

representations provided but lack clarity or

integration.

Adequate representation, with some gaps in clarity or integration.

Clear representation using appropriate

visuals and text.

Excellent visuals and textual explanation with minimal weaknesses.

Exceptional

integration of visuals and text, conveying insights with high originality.

Discussion/Findings (5%): - Data analysis and literature are critically discussed with clear arguments.

- Discuss how the analytics produces useful insights from the data.

No meaningful discussion or findings presented.

Barely any discussion or connection to findings.

Very weak discussion; lacks coherence and depth.

Minimal discussion; lacks critical insights or literature.

Limited discussion; findings weakly connected to arguments or

literature.

Findings are presented but lack critical evaluation or literature connection.

Adequate discussion; lacks depth or minor gaps in literature use.

Clear discussion of findings with evidence of

critical thinking and literature support.

Findings are well- discussed, with strong arguments and relevant

literature.

Findings are critically evaluated with exceptional insight, supported by literature and clear arguments.

Criteria

Fail 0-9%

Fail 10-19%

Fail 20-29%

Fail 30-39%

Fail 40-49%

Pass 50-59%

Merit 60-69%

Distinction 70-

79%

Distinction 80-

89%

Distinction 90-100%

Introduction(5%):

- Understanding of the given problems

- Problem decomposition into subtasks - Literature reviews (The most relevant ones)

- Include your innovative thinking

after the review

No introduction or problem understanding presented.

Barely recognisable problem understanding or review.

Very weak introduction;

lacks coherence and relevance.

Minimal effort in decomposition or literature review.

Weak problem decomposition; irrelevant or minimal

literature review.

Basic decomposition with minimal or generic literature review.

Adequate problem decomposition; minor weaknesses in

literature review.

Clear problem decomposition and appropriate literature review.

Detailed problem decomposition and thorough, relevant literature review.

Exceptional problem decomposition and insightful literature review, showcasing advanced understanding.

Approach(15%):

- A clear description of the chosen approach for each step

- The justification behind the chosen solutions/tools that they are the most appropriate choice to the given scenario

- (Compare with different models/functions that you may choose as alternatives before experiments) - Use a flowchart/diagram to

demonstrate the logical data flow of your finalised approach intuitively. - Code file submitted separately to

validate outputs in the report.

No approach described or justified. Code file not

submitted.

Barely any description or justification.

Very weak or irrelevant approach description.

Minimal effort in explaining or justifying the approach.

Weak or unclear approach; no alternative

considerations.

Basic approach; minimal

justification or alternative

considerations. C ode file submitted but with issues

(e.g., not executable or mismatched outputs).

Adequate approach with basic justification; limited consideration of alternatives.

Logical and welldescribed approach, with justification and minor gaps. Code file submitted, complete, and aligns perfectly

with the outputs presented in the report.

Clear and welljustified approach

with consideration of alternatives.

Code file submitted, complete, and aligns perfectly with the outputs

presented in the report.

Highly innovative approach, clearly described and justified with robust comparisons to

alternatives. Code file submitted, complete, and aligns perfectly with the outputs presented in the report.

Results analysis(20%):

- A clear description of the processed data after each step

- Visualise the results after each step with a description.

- Draw your conclusion about the data during processing.

- Clear explanation of the experimental results

No results analysis or evaluation was provided.

Barely any results analysis or meaningful evaluation.

Very weak analysis; needs coherence or

detail.

Minimal results analysis with superficial visualisation or evaluation.

Limited analysis, visualisation, and statistical evaluation could be clearer and stronger.

Basic analysis provided; lacks depth in visualisation or statistical evaluation.

Adequate analysis; minor gaps in clarity or statistical evaluation.

Clear results analysis with appropriate visualisation and statistical evaluation.

Results are thoroughly analysed and visualised with strong statistical evaluation.

Results were critically analysed with advanced statistical evaluation and impactful visualisation.

Discussion and conclusions(10%):

- Evaluation of the performance of the chosen approach

- Compare the different models that you have chosen with a detailed discussion, and explain how you have improved the models/results for multiple trials.

- Issues encountered are described, and solutions/actions taken are justified. - Alternative solutions and

suggestions for future work

No meaningful conclusions were presented.

Barely recognisable conclusions or suggestions.

Very weak or irrelevant conclusions.

Minimal effort in providing conclusions or evaluation.

Weak conclusions; minimal evaluation or irrelevant suggestions.

Basic conclusions;

limited evaluation or future suggestions.

Adequate conclusions; lacks depth or minor gaps in evaluation or suggestions.

Clear conclusions with a good evaluation of

performance and some future suggestions.

Strong conclusions with critical

performance evaluation and clear suggestions for future work.

Conclusions are highly insightful, critically evaluate performance, and propose innovative future solutions.