This assignment challenges you to apply similarity$-$based learning and probability$-$based learning concepts to a dataset of your choice. You will then use Power BI to visualize your findings and insights. The goal is to creatively express your understanding of these machine learning concepts while exploring the capabilities of Power BI$.$

Instructions:

Choose a Dataset

$$ Option $1$: Select a dataset from sklearn.datasets $($or a similar imported dataset$),$ such as the Iris dataset we worked with $$ but NOT the Iris dataset specifically, please.

$$ Option $2$: Choose a dataset from public sources that interests you, possibly one from your final project.

Note: Take time to understand the dataset you choose and how its features relate to the machine learning tasks!

Data Preparation

$$ Load your dataset into a Python environment.

$$ Perform any necessary preprocessing steps, such as handling missing values, encoding categorical variables, or normalizing features if needed.

Implement Machine Learning Models

$$ Apply a similarity$-$based learning algorithm to your dataset, such as k$-$Nearest Neighbors.

$$ Make predictions and evaluate the model$$s performance.

$$ Now, apply a probability$-$based learning algorithm to your dataset. An example of this would be the Naive Bayes algorithm.

$$ Again, make predictions and evaluate the model$$s performance.

$$ Note: Experiment with different settings for your models, such as the number of neighbors in k$-$NN$.$

Export your Results

$$ Combine your predictions with the actual values into a new dataset.

$$ Export this new dataset to a CSV file for use in Power BI$.$

$$ Create Visualizations in Power BI

$$ Import your CSV file into Power BI$.$

$$ Create visualizations that compare the performance of the two models.

$$ Highlight and take note of any interesting findings from the models$$ predictions.

$$ Save your results in a $.$pbix for submission $($a Power BI dashboard$).$

$$ Note: Use Power BI to explore your data from multiple angles, leveraging different types of charts and features for interactivity.

Write a Reflection

Document your process, findings, and any insights gained.

Discuss the performance of the similarity$-$based vs$.$ probability$-$based models.

Reflect on how your Power BI visualizations helped in analyzing the models$$ outcomes.

Note: Reflect on the entire process in your report, considering not just what you did, but also why you did it and what you learned.