There is a business opportunity in providing a more efficient and effective way for potential property buyers to search for, view and make an offer for the property they are interested to purchase. The potential annual revenue is estimated at between $2.5 Mil and $5 Mil for the first two years of operation in local market. Three of the six activities in the Property Search and Make Offer process which are more tedious and repetitive can be automated by the solution, leaving only the decision-making task to the buyer.
The solution comprises (1) User Fronting Chat-bot, (2) Property Search Agent, (3) Appointment Scheduling Agent, (4) Market Price Data Retrieval Robot, (5) Image Clustering Model and (6) Current & Future Price Prediction Model. The User Fronting Chat-bot is the conversational user interface that gets request from user to search for a new property and to view selected properties in the list of shortlisted properties. It also answers anticipated user queries about the solution. The Property Search Agent helps user to search for ideal properties from property listing website. It feeds property images into the Image Clustering Model to further down select the properties according to the user’s latent needs. The Appointment Scheduling Agent checks user’s calendar and arranges with the seller/seller agent the earliest opportunity to view a selected unit. It also passed the details of the selected unit into the Current & Future Price Prediction Model and provide the Estimated Fair Market Value for the user’s reference in case the user is keen to make an offer after viewing the unit.
Team Name: A22G
Members : Alfred Tay Wenjie, Wang Zilong, Wong Yoke Keong
Please refer to Annex D of report for further details.
You can download the video under the Report Folder.
Please download the project report from the link below. The user guide is in Annex A.