Reference no: EM133113724 , Length: word count:250

1. Application Project: This is the most common type of project students pick. Here, you will pick a cybersecurity problem that interest you and explore how best to apply AI or ML to solve it.

2. Algorithmic Project: Pick a problem or family of problems, and develop a new AI or ML algorithm, or a novel variant of an existing ML algorithm, to solve it.

3. Theoretical & Application Project: Prove some interesting/non-trivial properties of a new or an existing cybersecurity problem and then apply AI/ML algorithm to solve it. (This is often quite challenging, and so very few, if any, projects will be purely theoretical.)

4. Others: Some project will also combine elements of applications, algorithms, and theory.

Short descriptions on the proposal on following :

• Background
• Dataset
• Feature Creation and Selection
• Method
• Evaluation

Preprocessing datasets: While I don't want you to spend much time collecting raw data, the process of inspecting and visualizing the data, trying out different types of preprocessing, and doing error analysis is often an important part of machine learning. Hence if you choose to use pre-prepared datasets (e.g. from Kaggle, the Canadian Institute for Cybersecurity encourage you to do some data exploration and analysis to get familiar with the problem.

Background: This section should include the problem statement and brief discussion of existing approaches that have been proposed in the past to solve the problem that you have picked to work on.

Dataset: Describe the dataset that you plan to mine or utilize for your proposed project. Provide details regarding the source of the dataset, how many samples your dataset has?, how will you extract the data from the source if the data is being collected from the sources other than Kaggle or Canadian Institute for Cybersecurity? and is there any pre-existing category or class label?

Feature Creation and Selection: Discuss whether your project will involve feature extraction, feature transformation or feature selection strategies? If yes, list what features do you plan to extract, what transformation techniques will you employ, what standardization or normalization technique will you apply and what algorithms will you leverage for feature selection.

Method: Briefly discuss what learning algorithms will you be implementing. For example, you might be implementing classical machine learning algorithms such as Logistic Regression, SVM, Bagging, Random Forest etc. and few deep learning techniques such as CNN, ResNetXt, ResNet- 50, etc. Discuss what tools and languages will you be likely utilizing to implement these algorithms?

Evaluation: How do you plan to evaluate your proposed approach? Discuss whether you will be employing X-Fold Cross Validation, Jackknife Validation, or Independent Testing? And, what evaluations measures such as Overall Accuracy, F1-score, Mathew's Correlation Coefficient (MCC), Receiver Operating Characteristic (ROC) curve, Area Under the Curve (AUC) etc. will you be utilizing. Additionally, discuss if you plan on comparing your final model with existing approaches.

Authors and Title of the Reference Paper: Provide the authors name and title of the paper that you are using as the guideline for your research.