Task

You are required to analyse a dataset of your choice from the given list. You can find the dataset in any open source repository such as Kaggle $($the links are provided for your convenience$).$ Based on this dataset, build a Machine Learning $/$ Deep Learning model that can classify the data into associated classes.

Classification of X$-$Ray images to identify patients$$ lungs affected by Covid$-19.$ Classify X$-$Ray images into covid$-$infected or normal X$-$Ray images.

Links to an external site.

Build a machine learning Model for a recommender system to recommend games $($A game recommender system$)($Data source: Steam Video Games $|$ Kaggle$)$

Links to an external site.

Classify credit card transactions to be fraud or normal. Fraud detection in credit card transactions $($Data source: Credit Card Fraud Detection $|$ Kaggle$)$

Links to an external site.

Classification of computer network data for anomaly detection. $($Anomaly detection in network traffic$)($Data source: ADCG $2016$ : Network Anomaly Detection $|$ Kaggle$)$

Links to an external site.

Your report should contain some background study and methodology followed by results and discussion. Further, a reflection on the efficacy of the technique$($s$)$ used and an appendix with code $($or a link to online repository $/$ Jupyter Notebook$)$ should also be included. This will be equivalent to a $3000-$word report.

Follow the steps to complete this assessment

Setting the context

In this section, you will describe the dataset by explaining its variables, and the business context. You can develop the context relevant to the dataset that you are analyzing. Further, you will briefly describe the analytics that you are focusing on$.$ Apply some data visualization tools to describe the data characteristics such as its distribution over classes $($number of instances belonging to one class$).$ You can use graphs to present the data distribution. It will also help you describe the bias in the data and its reflection in the results. Please note that it is not expected that you will remove bias from data, but reporting it properly is crucial.

Select Machine Learning$/$Deep Learning Model

Based on your data you can now select a machine learning or deep learning model to do the classification task. There should be sufficient justification for this selection. Describe the selected model briefly and explain different parameter values.

Results

This section will present the classification results. You can document some or all of the possible results such as Confusion Matrix, Accuracy, Root Mean Squared Error, Precision, Recall, etc. You can also use graphs to present some of the results.