COM7039M Machine Learning Assessment 1 Brief | YSJU

COM7039M Programme Learning Outcomes (PLO)  

PLOs 7.1-7.7   7.1 

• Evaluate computer science concepts and principles and their application to the effective design, implementation, and usability of computer-based systems.
• Apply the findings of advanced scholarship and/or contemporary research and practice to the solution of computer science problems.
• Critically evaluate computer science problems, including those at the forefront of the field.
• Demonstrate operation within applicable professional, legal, social, and ethical frameworks.
• Demonstrate originality and creativity in the solution of computer science problems.
• Recommend, with detailed justification, the appropriate computer science principles and practices to apply to significant domain-specific activity.
• Apply standards, quality processes, and engineering principles to the solution of computer science problems.

Assignment Description

This coursework aims to highlight the students' comprehensive understanding and knowledge of the module, evaluating their analytical abilities and strengths. It comprises two tasks that will assess and challenge their level of analysis.  The tasks have been carefully designed to test the students' proficiency in applying the concepts learned throughout the module and to demonstrate their capabilities in handling real-world scenarios and problems. The successful completion of these tasks will reflect the student's mastery of the subject matter and their ability to think critically and creatively. 

Task 02 : Programming Exercise - (60 Marks)

a. Develop a Machine Learning model with critical sentiment analysis for Amazon reviews of customer reviews and star ratings to classify each review as either positive, negative, or neutral and employ a predictive analysis to forecast customer behaviour.   

Dataset Description:     

The  "Amazon  Reviews for  Sentiment  Analysis"  dataset is a collection of textual reviews and corresponding sentiment labels from  Amazon's e-commerce platform. This dataset is commonly used for sentiment analysis tasks,  which involve determining the sentiment or emotional tone expressed in the text. The goal is to classify each review as either positive, negative, or neutral based on the sentiment conveyed by the text.

(OR)    

b.  Develop a Time Series forecasting model to predict future sales based on historical patterns and other relevant features with exploratory data analysis of the Sales dataset.  

Dataset Description:    

This dataset contains historical sales data for a certain period. It includes various features that can be used for exploratory data analysis and time series forecasting tasks. The data represents the sales performance of a product or a group of products. The data might have been collected from a company's internal database, a retail store's point-of-sale system, or an e-commerce platform. The dataset covers a specific period, which could be days, weeks, months, or even years, depending on the granularity of the data.  The main objective of analysing this dataset could be to perform exploratory data analysis to understand trends, patterns, and correlations in sales data.

Alternative Source:           

The datasets are available to download from the Machine Learning (ML) module in the Moodle platform.    

Guidelines to Prepare Your Assignment:  

1.  Data Exploration and Pre-processing:  

• Load and explore the dataset to gain insights into the data's characteristics.
• Handle missing values, if any, and perform data cleansing as required.
• Perform data visualisation to understand the distribution of variables and their relationships with them.    

2.  Feature Engineering 

• Identify relevant features that may influence and create new features if necessary.  
• Convert categorical variables into numerical representations using appropriate encoding techniques.   

3.  Model Selection and Training:   

• Split the dataset into training and testing sets.  
• Select appropriate machine learning algorithms  (e.g.,  logistic regression,  decision trees, random forests, support vector machines, etc.) for the predictive modelling task.  
• Train the selected models on the training data and evaluate their performance on the testing data.   

4. Hyperparameter Tuning: 

• Fine-tune the hyperparameters of the chosen algorithm to optimise the model's performance and avoid model overfitting.  

5. Model Evaluation:  

• Compare the performance of different models using appropriate evaluation metrics such as the confusion matrix, accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC-ROC).
• Identify the best-performing model for prediction.   

6. Model Deployment: 

• Deploy the model to generate predictions for new and previously unseen data.   

7.  Conclusion and Recommendations: 

• Summarise the key findings of your analysis, highlighting the model's performance and any insights gained.
• Propose potential improvements or additional steps that could be taken to enhance the system.  

 Submission Guidelines:   

Prepare  a comprehensive report documenting your approach,  methodologies,  results,  and insights gained from the project.   

Include code snippets,  visualisations,  and explanations to  support  your findings.  Your report should be clear, concise, and well-organised.   

Suggested that you divide your work into TWO parts for submission. 

1.  The first part should consist of a combined PDF file encompassing both Task 1 and Task 2 

2.  The second part is expected to comprise code files presented in the  Jupyter Notebook format (.pynb).   

Points to Remember: 

1. Avoid plagiarism and provide proper citations for all academic and industrial sources. 
2. Use clear and concise language in your explanations to facilitate understanding.  
3. Demonstrate your ability to apply theoretical knowledge in practical applications and connect it with real-world scenarios.

Additional Information

Assignment work must be original and not copied directly from existing sources. Quoting others is permissible if proper referencing is provided. If the Turnitin similarity score exceeds 25%  and/or self-plagiarism surpasses  6%,  your work will be investigated for potential academic misconduct.

Assessment Regulations  

Please refer to the York St John University Code of Practice for Assessment and Academic Related Matters 2023-24.     

We ask that you pay particular attention to the academic misconduct policy. Penalties will be applied where a student is found guilty of academic and/or ethical misconduct, including termination of the program (Policy Link).     

You are required to keep to the word limit set for an assessment and to note that you may be subject to a penalty if you exceed that limit.  You are required to provide an accurate word count on the cover sheet for each piece of work you submit (Policy Link).    

For late or non-submission of work by the published  deadline or an approved extended deadline,  a mark of  0NS  will be  recorded. Where a re-assessment opportunity exists,  a student will normally be permitted only one attempt to be re-assessed for a capped mark (Policy Link).     

An extension to the published deadline may be granted to an individual student if they meet the eligibility criteria of the (Policy Link).   

Assessment Criteria