Two Sigma Connect: Rental Listing Inquiries

Predicting the New York Tenants interests on new rental listing with Rent


Rent Hop makes apartment search smarter by using data to sort rental listings by quality. Two Sigma and Rent Hop, a portfolio company of Two Sigma Ventures wanted to predict the number of inquiries a new listing receives based on the listing’s creation date and other features. Doing so will help Rent Hop better handle fraud control, identify potential listing quality issues, and allow owners and agents to better understand renters’ needs and preferences.

Data Introduction:

We will predict how popular an apartment rental listing is based on the listing content like text description, photos, number of bedrooms, price, etc. The data comes from an apartment listing website. These apartments are located in New York City.The target variable, interest level, is defined by the number of inquiries a listing has in the duration that the listing was live on the site

Numeric Predictive Solution:

We have chosen R language for Prediction on interest level, so in our analysis we considered, Test and Train tables which are of JSON format. Train dataset is useful for our model prediction where we train our model for prediction, Here model suggests algorithm used for prediction. Test is the table where we have to find out interest level by individual fields HIGH, MEDIUM, LOW. To go further we have to consider fields which are in our Test dataset, by default both Test and Train datasets are having same fields. From the data fields section of our Train data we will look at mostly on features field which will decide the customers interest so we split the sentence in to individual words for our model convenient and for our visualization to. We use an algorithm called XGB where it is useful for performance (Running in low time) and another algorithm called Prediction for predicting our model. After having done with our XGB & Prediction algorithms it gives the predicted data based on listing id with in less time of around 0.55982 seconds. Based on listing id we are going to have our model generated which is shown below

Based on listing id we are going to make merge with Train data set and have our visualization.


For this we are going to use SAP Lumira tool and create a dashboard, below is the dashboard with data we predicted using R Language.

Dash Board:


Numeric Technologies Inc has openings for the positions

Numeric Technologies Inc has openings for the positions

Senior Software Engineer/Linux (NUM1631)with Bachelor's degree in ComputerScience, Engineering(any),Technology, Any Analytical Science or related and5 yrs of exp. to design, develop, implement ,troubleshoot application andsystem level software in a variety of programming languages. Deploy andmanage VMware ESX Server Farms. Create Windows, OEL Linux OStemplates and images with necessary pre-requisite packages and manage Linux servers.

Design, document programs and system implementation for all Informaticaprograms. Interacting with all application owners to understand the risk,dependencies in migrate the application from shared infrastructure todedicate.

Software Engineer (NUM1632)with Master's degree in Computer Science,Engineering(any),Technology, Any Analytical Science or related to providethe top management and end users, a fully functional data warehouse for eachof the business segments separately and as per the user requirements.Develop the reporting system using Hyperion Planning, Essbase and Analyzer.Design, develop, implement, troubleshoot application and system levelsoftware using programming languages including Java, Visual Basic, Essbaseand analyze user needs and develop software solutions with aim of optimizingoperational efficiency.

Work location is Warrenville, IL with required travel to client locationsthroughout the USA. Please mail resumes to 4200 Cantera Drive, Suite 100,Warrenville, IL 60555 (OR) e-mail to