Embedded Files
Tags: Classification Model, Machine Learning
Scikit Learn
Matplotlib
Pandas
Python
Jupyter Notebook
● CONTEXT
● CONTEXT
Megaline, a leading mobile carrier, strives to enhance customer satisfaction by ensuring subscribers are on the most suitable plan for their needs. Despite introducing newer plans, Smart and Ultra, many customers continue to use legacy plans. To address this, Megaline has launched a program to transition customers to the newer, optimized plans. The Customer Insights Department aims to develop a predictive model to recommend either the Smart or Ultra plan based on user behavior.
● OVERVIEW
● OVERVIEW
We developed a classification model to predict the most appropriate plan for each user, using monthly behavior data from subscribers who have already switched to the new plans. The dataset includes metrics such as the number of calls made, call duration, number of text messages sent, internet traffic used, and the plan for the current month.
The project involved splitting the data into training, validation, and test sets, experimenting with various models and hyperparameters, and evaluating the models' performance. We set a minimum accuracy threshold of 75% to ensure the model's effectiveness. Finally, a sanity check was conducted to verify the model's robustness. The ultimate goal is to deliver a high-accuracy model that facilitates personalized plan recommendations, improving customer satisfaction and loyalty.
● LIBRARIES / PACKAGES USED
● LIBRARIES / PACKAGES USED
The following libraries and packages were used in this project:
Pandas: Data manipulation and analysis (pandas)
Matplotlib: Plotting library (matplotlib)
NumPy: Numerical computing (numpy)
Scikit-learn: Model building and testing (sklearn)
Train Test Split
Decision Tree Classifier
Random Forest
Logistic Regression
Metrics
Dummy Classifier
To see the code, please visit my GitHub:
Page updated
Google Sites
Report abuse