Proactive Quality Control: Predicting Component Failure in Manufacturing using Anomaly Detection and Explainable AI

This project aims to develop an innovative anomaly detection system using advanced data mining and deep learning techniques to accurately identify and localize defects in manufacturing components, thereby enhancing quality control processes and reducing production losses.

Inbox Alchemy: Maximizing Enron’s Goldmine with Topic Modeling & Summarization

This project utilizes natural language processing to automatically identify email topics, summarize their content, and alleviate the burden of email overload by facilitating efficient navigation and extraction of relevant information from inboxes.

The Power of Reinforcement Learning: Reviving Nostalgia Through Super Mario Bros

This project aims to utilize reinforcement learning (RL) techniques to train an artificial intelligence agent capable of playing the iconic Super Mario game.

Two-stage Facial Recognition and Classification Model

This project implements a sophisticated two-stage facial recognition and classification model, leveraging Multi-task Cascaded Convolutional Neural Networks (MTCNN) for face detection and Inception-ResNet v1 for accurate facial classification.

The Decoder Transformer: Implementing a nanoGPT with Andrej Karpathy

In this notebook, we embark on a coding journey alongside Andrej Karpathy, following his instructional video titled "Let's build GPT: from scratch, in code, spelled out." Our objective is to create a Decoder Transformer, also known as nanoGPT.

French to English Translation Using Sequence-to-Sequence RNNs with Attention

In this notebook, we implemented Sean Robertson's "NLP FROM SCRATCH: TRANSLATION WITH A SEQUENCE TO SEQUENCE NETWORK AND ATTENTION" using a Jupyter notebook. The purpose of this project is to deepen our understanding of the original implementation.

Project OPTIMA: A Predictive and Prescriptive Quality System for [Redacted]

Project Optima presents advanced predictive modeling methods and a novel prescriptive tool to address challenges within a manufacturing quality process, resulting in substantial improvements in production efficiency and cost savings for the organization.

Friendly Skies or Stormy Weather? Clearing the Fog Around Airline Passenger

In this study, we use Explainable Boosting Machine (EBM) to identify relevant features and extract key insights from US airline passenger data. In addition to the global/local explanations provided by EBM, we also use Diverse Counterfactual Explanations (DiCE) to provide passenger-level explanations

Why Transform Data? Examining Feature Transformation and How it Boosts Model Performance

Getting accurate results is the goal of every machine learning project. But feeding data as it is to a model does not always give the desired outcome. And doing so will introduce certain biases to your machine learning model as it learns. One way to mitigate this problem is by feature transformation

From Publication to Citation: Understanding the Factors Driving Research Recognition

This project aims to address the significant challenge of low citation rates and limited readership for journal articles, which has far-reaching implications for the scholarly community and the broader research landscape. What are the factors that contribute to higher citation rates and readership?

Analyzing Temperature and Precipitation Patterns for Climate Assessment

This study aims to examine the potential presence of climate change to understand its impact and identify potential indicators. Specifically, to answer the question: Do precipitation and temperature patterns exhibit notable variations over time in certain regions, potentially indicating the

Unveiling the Shifting Landscape of Airbnb: Understanding Supply, Demand, and Pricing Dynamics

This study aims to analyze the changes in Airbnb prices, availability, and demand from 2019 to 2022 and explore variations in these metrics across different countries. How did Airbnb prices, availability, and demand change over the recent years? How do prices, availability, and demand vary?

ECLAIRS: Explainable Clustering and Item Recommender System

This project aims to leverage clustering-based categorization and a recommender system to optimize dessert listings on an e-commerce platform, capitalizing on the "first-mover advantage" to boost revenue and profitability.

Hey Weeb, Watch This: An Anime Recommender System

As the library of Anime titles from different platforms grows, how do we create a recommender system that can suggest relatively unknown titles that are highly similar to the viewing history of a specific user and the recommendations being personalized to each viewer?

A Study on Professional Athletes in the European Football League

As scouts and managers for an international team, how can we cluster players based on skills and attributes to match team expectations and aid in player selection for creating the ultimate pool of players for the Upcoming Seasons of the UEFA Europa League.

Birit Ka Ghorl? Your Next Karaoke Song Choice

About This project aims to build a recommender engine that determines the top five song titles based on word query. It is a usual occurence during karaoke sessions that you probably know the song, even the entirity of its lyrics, but you just can't seem to remember its title.

Getting Over the Hangover: A Book Recommender System

A reader does not have to feel so empty after finishing a good book after all. Hence, this Book Recommender system aims to present the next top 5 books based on a previously read book - to ultimately fill the void of a book hangover.

London Bridge is Burning Down: A Fire Brigade Analysis

The London Fire Brigade (LFB) is one of the largest firefighting and rescue organizations in the world. In the UK, they are also the busiest fire and rescue service. Using the incident and the mobilisation datasets of London Fire Brigade (LFB), the study aims to describe the characteristics of LFB's

Data Science's Next Chapter in the Philippines: Boom or Bust?

This study endeavors to address the pressing question of how to comprehensively understand the evolving appeal of data science in the Philippines, with a clear objective of informing strategic growth and promotion strategies for the upcoming quarter.

Delay Propagation Dynamics and Disruptive Elements: An Analysis of the United Airlines Network System

In the complex airline industry, disruptions often trigger chain reactions across transportation systems, leading to delay propagation. This study applies network science, experimenting with Chauhan et al.’s methodologies using the United Airlines network, to address this challenge.

Unraveling Philippine Aviation: A Comprehensive Analysis through Network Science and Scenario Exploration

What if a major airport like the Ninoy Aquino International Airport (NAIA) suddenly ceased operations? While flight disruptions would undoubtedly follow, I can't help but wonder about the broader implications for the other airports across the country.

Bridging the Gap: A Study on Access to Basic Necessities in the Philippines

This is a comprehensive study examining access to basic necessities in the Philippines, with the goal of identifying disparities and proposing actionable solutions to enhance accessibility and equity across communities.