Repositori ini berisi proyek data mining yang menganalisis perilaku pelanggan di sektor perbankan, dengan fokus pada prediksi churn. Dilengkapi dengan visualisasi data dan implementasi algoritma Naive Bayes untuk analisis lebih lanjut.

Analyzing US crime statistics using hierarchical clustering to uncover patterns in state-level arrest data and Advanced analytics to delineate market segments in retail, optimizing targeted marketing strategies through customer behavior and demographic profiling.

My Final Submission for the 'Santander Customer Transaction Prediction'. I have participated in this very tough and interesting competition on Kaggle a while ago and I finally got the time to put all the work together in this Repo.

Natural language processing for NPS comments. Supervised classification

In this project, the website user conversion has been studied. The dataset has over 20000 observations and 6 variables. The attributes discount, conversion, source and country are characters whereas the attributes total_pages_visited and visit_duration are numerics.

web app to predict insurance bills

Predicting Customer Lifetime Value on a public dataset of e-commerce.

SQL-driven analysis and reporting for strategic insights into Excelsior Mobile's customer usage and billing data to inform business decisions

Room Reservation System focused on optimising profits and mitigating cancellations by prioritising customers

frank's blog

Repositories of analytics workshop

Exploration, visualization and implementation of machine learning models on customer, transnational other such data of Toyota, Universal Bank, Wayfair, etc (R, SAS and Alteryx)

Repository contains the code and data related to the Master Thesis "Exploring the Potential of Open Data for Customer Analytics: A Swiss-Centered Proof-of-Concept"

This is task 2 of 3 from the Power BI PwC Switzerland Virtual Internship organized in partnership with Forage

Building a response model based on the Portuguese bank marketing data set from UCI

A Rust crate for calculating Net Promoter Score (NPS) from survey responses.

A data transformation solution for LendingClub, encompassing data cleansing, feature engineering, and loan scoring to facilitate data-driven decision-making.

Used linear and tree-based models, visualizations techniques to solve commonplace data science problems, including calculating conversion rate, analyzing A/B testing, churn/retention prediction, fraud detection, funnel analysis, pricing testing, marketing campaign optimization, clustering, user referral, loan default prediction, optimization, pe…