Model Comparison

We need to create another model, other than the basic logistic regression, and compare how they perform. A minimum example of this issue would be to run the model through a Neural Network and output a confusion matrix for the logistic regression and the Neural Network and see how the matrices compare (note it does not need to be a NN). Does the NN do better with false positives? There are many additional ways we can compare model results, you are free to do some research.

Here is a great resource to understand and implement confusion matrices.

Here is a resource to help you with the concepts around a Git workflow. Branching to committing to merging. It is a good overview of the whole process but doesn't contain a lot of detail on how to actually do it.
Here is a resource to help you with branching and merging in Git. It is quite long so don't spend too much time on it. I find it useful to really understand Git.

The assignee must:

1. Create a branch off master (this way you can take the existing working code and build off it)
2. Edit the Jupyer-Notebook named "Train.ipynb"
3. When finished create a pull request to merge your code into master
4. At this stage the ticket should be moved to In Review column in the (Task Board)[https://github.com/awebsters/Phishing-Detection/projects/1]
5. Assign awebsters as a reviewer and discuss make changes for comments he has about the code
6. When the pull request is approved merge, delete the branch, and move this issue to "done".

More features

Some features discussed and included in the Google Document shared within the discord are not implemented. Implement the remaining of these features and evaluate if they should be included in the model.

The two features not implemented in the Google Doc:

• Look for known words within the data, or looking for slightly misspelled words (Difficult but has a large potential for impact)
• Number of numerical digits in an email address (Medium difficulty)

Here is a resource to help you with the concepts around a Git workflow. Branching to committing to merging. It is a good overview of the whole process but doesn't contain a lot of detail on how to actually do it.
Here is a resource to help you with branching and merging in Git. It is quite long so don't spend too much time on it. I find it useful to really understand Git.

The assignee must:

1. Create a branch off master (this way you can take the existing working code and build off it)
2. Edit the Jupyer-Notebook named "features.ipynb"
3. When finished create a pull request to merge your code into master
4. At this stage the ticket should be moved to In Review column in the (Task Board)[https://github.com/awebsters/Phishing-Detection/projects/1]
5. Assign awebsters as a reviewer and discuss make changes for comments he has about the code
6. When the pull request is approved merge, delete the branch, and move this issue to "done".

HTML features

To improve our model's accuracy we need more data and we should be able to get all the data we need straight from a URL! A complex feature to work on is how to safely get the HTML code from a website and do research on the types of things we can use in this HTML code!

Here is a resource to help you with the concepts around a Git workflow. Branching to committing to merging. It is a good overview of the whole process but doesn't contain a lot of detail on how to actually do it.
Here is a resource to help you with branching and merging in Git. It is quite long so don't spend too much time on it. I find it useful to really understand Git.

The assignee must:

1. Create a branch off master (this way you can take the existing working code and build off it)
2. Create a new Jupyer-Notebook named "html_features.ipynb"
3. When finished create a pull request to merge your code into master
4. At this stage the ticket should be moved to In Review column in the (Task Board)[https://github.com/awebsters/Phishing-Detection/projects/1]
5. Assign awebsters as a reviewer and discuss make changes for comments he has about the code
6. When the pull request is approved merge, delete the branch, and move this issue to "done".

Feature Visualization

Our project needs to visualize the dataset we have created.

Here is a resource to read to start you off with some information about visualizing data!

Here is a resource to help you with the concepts around a Git workflow. Branching to committing to merging. It is a good overview of the whole process but doesn't contain a lot of detail on how to actually do it.
Here is a resource to help you with branching and merging in Git. It is quite long so don't spend too much time on it. I find it useful to really understand Git.

The assignee must:

1. Create a branch off master (this way you can take the existing working code and build off it)
2. Edit this Jupyer-Notebook named "features.ipynb" too add a good visualization for each feature created
3. When finished create a pull request to merge your code into master
4. At this stage the ticket should be moved to In Review column in the (Task Board)[https://github.com/awebsters/Phishing-Detection/projects/1]
5. Assign awebsters as a reviewer and discuss make changes for comments he has about the code
6. When the pull request is approved merge, delete the branch, and move this issue to "done".