Data Engineering: Flights Data Pipeline

In this project, i extract historical flight data (facts) and additional information (dimensions) from various sources, transforming them and loading them into a structured data warehouse in order to make this data available for data analysis.

Predictive Analytics: Loan Eligible Classification

Using LazyPredict, I compared 30+ machine learning algorithms to find the best model that will later be used to predict the eligibility of prospective borrowers at the Dream Housing Finance company. This project is one of my submissions when I became a Machine Learning Developer Trainee at IDCamp.

Recommendation System: KDrama Recommendation

This recommendation system uses the Content Based Filtering method to recommend K-Drama titles based on genre compatibility. Furthermore, the process of calculating item similarity will be carried out with Cosine Similarity.

Predictive Analytics: Medical Cost Personal

Through this project, we are trying to help insurance companies to predict the cost of insurance premiums based on the background of the insurance beneficiary. With the limited availability of features, the resulting model can provide a fairly good performance.

Dashboard: Ads Performance

This dashboard was created to monitor ad performance data sourced from Meta Ads, Google Ads and Marketplace Ads. Previously, the data manipulation process had been carried out to maintain the confidentiality of the original data.

Storyboard: Walmart Data Analysis

Through this storyboard, I try to provide recommendations to the stakeholders regarding efforts that can be made by Walmart to optimize sales in November - December 2012. I also provide the "Dokumen Rencana Karya Olahan Data" as a framework in the storyboard creation process.

Text Analysis: Customer Voice Analysis on Tokopedia

This project aims to improve the quality of RSV Helmet’s products and services based on the results of "the latest 1000 reviews submitted by buyers on their Tokopedia" analysis. Data extracted using Selenium and BeautifulSoup, then processed using Sastrawi.

Dashboard: Spread of Disease

This project aim to analyze how the disease spreads at a certain location and time period. The data source used comes from Klinik Pintar's transaction records for the last 4 years, where previously a data manipulation process was carried out to maintain the confidentiality of company data.