Preference adjustment
Scoring algorithm (normalized and weighted)

Find Your District

Context

In today’s rental market, it’s increasingly difficult to find affordable housing without sacrificing access to essential amenities. Renters are often forced to manually compare neighborhoods, prices, and nearby necessities such as transit, grocery stores, and healthcare — a process that is time-consuming and overwhelming. We asked ourselves: why not build a tool that helps people find the best places to live within the same price range, based on what actually matters to them?
That question became Find Your District.

What it does

Find Your District is a map-based application that helps users identify and compare neighborhoods based on nearby amenities and overall livability.

Users explore areas on an interactive map where neighborhoods are scored and visualized using color-coded heatmaps
The system emphasizes popular and highly rated amenities (based on Google ratings and activity)
Users can quickly visualize areas that match their interests
Supports real-time district-level comparisons

How we built it

Data Fetching

We used the Google Maps API and Google Places API to collect amenity and commodity data.

Visualization

We used Deck.gl to render heatmaps that highlight high-value living areas.

Caching System

API calls are expensive, so we implemented a geohash-based grid caching system.
Small geographic grids are cached and reused to minimize redundant API requests.

Scoring Algorithm

We evaluate each data point relative to the mean and standard deviation of the dataset. Each area is scored based on:

User preferences
Amenity popularity and rating
Distance to preferred amenities

Challenges we ran into

Accurate area representation required large amounts of data, which increased operational costs (approximately $32 CAD per 1,000 API calls). Additionally, each grid was evaluated against every amenity data point, significantly increasing time complexity and causing delays in user experience.

Designing a fair scoring system was also challenging. Not all amenities have equal importance, and balancing their weights required multiple iterations. We also faced difficulties with map rendering, API rate limits, and efficiently displaying large volumes of location data without overwhelming users or hurting performance.

Accomplishments we’re proud of

Turning raw geographic and amenities data into an intuitive visual experience.
Building a working heatmap that meaningfully reflects neighborhood value within a hackathon timeframe.