Rheyhan's portofolio

Our project exemplifies the synergy of expertise cultivated through the Bangkit Program. We combine technical proficiency with a deep understanding of agricultural dynamics to deliver a solution that bridges innovation and practical application.

Developed collaboratively by a diverse team of machine learning, mobile development, and cloud computing cohors. Our project exemplifies the synergy of expertise cultivated through the Bangkit Program. We combine technical proficiency with a deep understanding of agricultural dynamics to deliver a solution that bridges innovation and practical application.

Looking ahead, we envision expanding our platform's capabilities to include predictive analytics and personalized agronomic recommendations. By integrating advanced data analytics and feedback mechanisms from users, we aim to continuously enhance the platform's efficacy and relevance in supporting sustainable agricultural practices globally.

Repository link: Private

Project Plan: HERE

Presentation and demo: HERE

Our current focus is on developing an NLP tool for sentiment analysis in Indonesian politics, specifically for the 2024 election. This project, part of the Satria Data competition, aims to provide insightful analysis and understanding of public sentiment towards political candidates and issues. We are leveraging advanced natural language processing techniques to enhance the accuracy and relevance of our analysis, empowering stakeholders with valuable insights during this pivotal electoral period.

Repository link: HERE

Our current endeavor focuses on harnessing YOLOv8, an advanced deep learning framework, to develop a robust system for real-time detection of vehicle types, license plates, and helmet usage among motorcyclists. This initiative is nearing completion and undergoing final refinements before its official release.

Once polished and released, our YOLOv8-based system promises to revolutionize traffic management and safety enforcement practices. Providing authorities with real-time insights into vehicle types, license plate data, and helmet compliance, it facilitates proactive measures to enhance road safety and traffic efficiency.

At this stage, the project is undergoing final polishing to refine algorithms, optimize performance, and validate results through rigorous testing. We are committed to delivering a robust and reliable solution that meets the demands of real-world traffic management applications.

Looking forward, we aim to expand our system's capabilities to include additional features such as pedestrian detection and integration with smart city initiatives.

Repository link: Private

Research paper: HERE

Poster: HERE

Our team:

The project was inspired by my passion for deep learning and interactive gaming. I aimed to develop an automated fruit slicer that mirrors the gameplay mechanics of Fruit Ninja, where virtual fruits are sliced with precision and speed.

This project serves as a testament to the application of deep learning in recreating interactive gaming experiences. Moving forward, I plan to further refine the system's capabilities, explore additional gaming dynamics, and potentially contribute the project to open-source communities to inspire further innovation in AI-driven gaming interfaces.

In conclusion, the automated fruit slicer using YOLOv8 represents a fusion of technological innovation and gaming nostalgia, showcasing the potential of deep learning in transforming interactive entertainment experiences.

Repository link: HERE

This team-based project, undertaken as the final assignment for the Time Series Forecasting Methods course, focuses on comparing the effectiveness of ARCH/GARCH and LSTM models in forecasting sea levels at Marina Ancol for the period 2022-2023. Our goal is to determine which model offers the most accurate and reliable predictions, thereby providing valuable insights into the application of time series analysis for environmental forecasting.

The findings of this project have significant implications for environmental forecasting and management. Accurate sea level predictions are crucial for coastal planning, flood prevention, and climate change mitigation efforts. By identifying the most effective forecasting method, this project contributes to enhancing the accuracy and reliability of environmental predictions.

This project not only fulfills the requirements of the Time Series Forecasting Methods course but also provides a comprehensive analysis of the comparative effectiveness of ARCH/GARCH and LSTM models in sea level forecasting. Our findings will be valuable for researchers, environmentalists, and policymakers involved in coastal management and climate change adaptation strategies.

In summary, this project demonstrates the power of combining traditional statistical methods with modern deep learning techniques to tackle complex environmental forecasting challenges, thereby contributing to the advancement of time series analysis and its applications in real-world scenarios.

Repository link: HERE

Research Paper: HERE

developed a DNN model as part of the Introduction to Data Science course, focuses on revolutionizing heart health monitoring through advanced Deep Neural Network (DNN) techniques. By leveraging sophisticated algorithms, our project aims to classify individuals' health statuses, detecting potential diseases early and providing proactive health management insights. Through meticulous data analysis and model refinement, we strive to contribute to enhanced healthcare practices and personalized patient care strategies. This initiative underscores our commitment to harnessing cutting-edge technology for improving public health outcomes and advancing data-driven approaches in medical science.

Repository link: HERE

Poster link: HERE

Developed a robust web scraping tool using Python to gather detailed specifications of various devices from online sources. The goal was to create a comprehensive database that included information such as processor type, RAM, storage capacity, display resolution, camera specifications, and more.

This project aimed to provide tech enthusiasts, consumers, and developers with a valuable resource for making informed decisions when purchasing or researching devices. By automating the process of gathering and updating device specifications, it facilitated easier comparison and analysis, ultimately enhancing the efficiency of information retrieval in the tech industry.

Repository link: HERE