Inspiration - To make a one stop shop for an investor looking to invest in commercial real estate.
What it does - Provides customized site recommendations and projected revenue for a given set of user inputs such as Location, Business Sector and Budget. It also considers area demographics, labor related costs, competitors, demand and affordability as a part of the ML prediction model
How we built it -
1)Data Accumulation from sources like Demographic data, Historical Data, Labour data, Industry trends from CBRE box and external goverment websites 2)Used Python for Extensive Data-Preprocessing - Removal of missing/null/duplicate values, Removing irrelevant features 3)Used Google Studio for Market analysis dashboarding and visualization of metrics such as Sector-wise revenue, Zone-wise land value and Labour costs etc 4)We built an interactive Frontend that takes user input using JavaScript, HTML5, CSS3, VanillaJS and Flask 4)The backend model was based on Neural Network ML model to predict Revenue - using parameters like location , industry, competitiors. Predicted top 3 most viable sites based on said input parameters
Challenges we ran into
Data Cleaning - Extremely messy data with unnecessary features Cors - We had trouble integrating Frontend and Backend with flask but eventually figured it out
Accomplishments that we're proud of
Accomplishing our Minimum viable product within the given timeframe Learning on the go
What we learned
Data preprocessing One hot encoding Neural networks
What's next for Sitation
- Incorporating more robust models for revenue estimation consisting of more determining features
- Estimation of area wise traffic flows to determine busiest sites/locations
- Using real-time data containing sites currently available
Log in or sign up for Devpost to join the conversation.