5 Data Science Projects that You Can Complete Over the Weekend



Original Source Here

1. Predicting credit card spend

This project is based on a regression problem. The task is to train a machine learning model that can predict credit card spending based on historical data. This model can help banking industries to decide credit card limit based on user’s past experience.

This project has 15+ columns to find the best features out of them. In this way, you will also learn different features elimination and selection techniques.

Finally, we will use bagging and boosting mechanisms to find a better model for our data.

Data Screenshot

Project Source: You can find the solved Jupyter Notebook on my GitHub page here.
Project Data: Dataset for the project can be found here.

2. Walmart sales forecasting

This project is another real-time industrial task. The main goal is to forecast the future sales of different items. This project can help companies to maintain their supply depends on future sales.

Here, we have the interval level data for sales. We can try ARIMA, SARIMA, Holts, and other prediction models to forecast sales for the upcoming months.

Data Screenshot

Project Source: You can find the solved Jupyter Notebook on my GitHub page here.
Project Data: Dataset for the project can be found here.

3. Network intrusion detection

This project is another exciting classification task with 100+ features to train a classification model to classify the network intrusion type.

The main task here is to try different features elimination and features selection techniques to reduce the number of features. And, finally, we have to come up with some important features that affect the target variable.

In the end, we can try a set of classification models to find the best suitable model for our data.

Data Screenshot

Project Source: You can find the solved Jupyter Notebook on my GitHub page here.
Project Data: Dataset for the project can be found here.

4. Segmenting credit card users

This project is from the segmentation area. The task is to make segments of different users who are having the same characteristics. Industries use these segments to target their users with a set of campaigns.

This project will use the KMeans clustering to define different clusters based on the silhouette scores of different clusters.

Data Screenshot

In the end, we will get an excel file with different segments of users to target our marketing strategies.

Demo Output

Project Source: You can find the solved Jupyter Notebook on my GitHub page here.
Project Data: Dataset for the project can be found here.

5. Analyzing online job posts

This project is another exciting text analysis task. Here, you will learn different ways to deal with data cleaning steps. You will learn about different ways to convert text data into numeric vectors that help our machine to understand the text data.

We will also work with the text classification model followed by word cloud representation of the text data for summarization.

Data Screenshot

Project Source: You can find the solved Jupyter Notebook on my GitHub page here.
Project Data: Dataset for the project can be found here.

AI/ML

Trending AI/ML Article Identified & Digested via Granola by Ramsey Elbasheer; a Machine-Driven RSS Bot

%d bloggers like this: