wxneo / ga_projects

GA Projects

Capstone Project - CrowdSenseAI: Avert Crowd Crushing with Deep Learning

• Objective: Improve crowd safety during public gatherings and events.
• Problem: Prevent crowd crushing disasters due to overcrowding and mismanagement of spaces.
• Solution: Develop an advanced deep learning model for real-time crowd presence detection.
• Key Features: Estimate crowd capacity, trigger early warnings to authorities in hazardous conditions.
• Impact: Mitigate the risk of crowd incident, enhancing public safety during gatherings.

Project 1 - Data Analysis of Singapore Rainfall

• Conducted weather pattern analysis in Singapore using historical weather data.
• Goal: Identify significant insights on rainfall distribution for informed planning recommendations.

Project 2 - Resale Housing Price Prediction

• Developed a product for HDB resale flat price prediction and determining factors analysis.
• Objective: Provide the general public with predicted pricing and address real estate inquiries.
• Assisted users in exploring options within their budget, identifying affordable flat types and locations.
• Advised sellers on appropriate pricing strategies and offered insights on effective marketing techniques to enhance selling prices.

Project 3 - Analyzing Viewer Preferences and Sentiment towards Popular Sitcoms

• Developed an efficient machine learning solution for a streaming company to optimize their limited budget by selecting and retaining the most popular sitcom for their platform.
• Focused on classifying and analyzing user comments from different platforms, specifically targeting "Big Bang Theory" and "Brooklyn Nine Nine" sitcoms.
• Objective: Create an infrastructure to accurately identify the show referenced in viewers' comments and analyze sentiments expressed towards each show.
• Solution provides valuable insights into viewers' preferences, aiding the streaming company in determining the show elements in the sitcoms that are popular among the viewers and maximizing viewer satisfaction and engagement.

Project 4 - Data-backed solutions for combating West Nile Virus in Chicago

• Collaborated with the CDC as a third-party consulting firm, Data Nine-Nine, to review their West Nile virus (WNV) control efforts.
• Objective: Build a machine learning model to predict the presence of WNV and provide valuable insights and recommendations for enhancing outbreak combat strategies.
• WNV is an infectious viral disease transmitted by mosquitoes, posing significant health risks to humans.
• In 2002, initial human cases were reported in Chicago, leading to the establishment of a comprehensive surveillance and control program by the City of Chicago and the Chicago Department of Public Health (CDPH).
• Our engagement remains focused on reviewing and enhancing the existing program to combat WNV effectively.