🔭 Exploration

Netflix Recommender Bot

Dec. 2023, Python

GitHub: https://github.com/keanteng/chatbot

  • A chatbot capable of recommending Netflix movies of shows created using the Streamlit framework using PaLM-2 API.
  • The program allows large language model to read input data and make recommendation from the data. The data can be of jobs, movies, songs or texts related.

🏆 Competitions

Geo-based Sustainable Job Solution

Nov. - Dec. 2023, Python

GitHub: https://github.com/keanteng/datathon

  • A geo-based sustainable job solution prototype using Streamlit build using LLM, language model and time series forecasting models.
  • The objective of the project is to understand Malaysia’s labor market structural issues for resilient and sustainable economic growth. Malaysia remains stuck in a low-wage and low-skill economy because of the work offered domestically, and not because of the talent available. There are also significant skills mismatches between graduates and industry needs.

🏢 Internship Projects

Flood Analysis App

Jul. - Sept. 2023, Python

GitHub: https://github.com/keanteng/streamlit_floodv2

  • A flood analysis app that is able to visualize flood statistics such as flood incidents on map and charts. It is also capable of providing estimation of the extent of a flood based on user input.
  • The source of data is extracted from the annual report published by the Department of Irrigation and Drainage, Malaysia.
  • The flood estimation tool is build using source code from Mapaction

Univariate Flood Risk Model

Jul. - Sept. 2023, Python

GitHub: https://github.com/keanteng/flood_risk_model

  • A flood risk model that is able to predict flood risk for any location in Malaysia
  • An OLS or ordinary least square method was used in the study
  • The model is trained using Logistic Regression on 4500 labelled random geo-coordinates in Malaysia.

Multivariate Flood Risk Model

Jul. - Sept. 2023, Python, Google Earth Engine

GitHub: https://github.com/keanteng/flood_risk_model_2

  • A flood risk model that is able to predict flood risk for any location in Malaysia
  • The model is trained using Logistic Regression on 19000 labelled random geo-coordinates in Malaysia.
  • The study is referenced from the paper published on IOP Science

Python Package For Reporting Automation

Jul. - Sept. 2023, Python

GitHub: https://github.com/keanteng/blueautomata

  • A Python package that provides a framework for reporting workflow automation
  • The package was published on Pypi for global installation
  • A documentation site for the package was created using Materials for Mkdocs

Climate Data Catalogue Website

Jul. - Sept. 2023, Next.js, CSS, HTML

GitHub: https://github.com/keanteng/nextjs-directory

  • A website to allow users browse through data item containing in the Climate Data Catalogue with design inspired by the NGFS Directory