Receive up to € 20k to create your proof of concept

WORK WITH REAL DATA FROM LEADING CORPORATIONS

WHAT WE CAN DO FOR START-UPS
Access to leading corporate partners and opportunities for long-term collaboration
Receive up to EUR 20,000 in funding to draft a proof of concept – without giving up equity
Real-world use cases and unique datasets from established corporations

Data Challenges

DataHub Ruhr is a business building program that connects start-ups with established corporations in the Ruhr region. The program tasks start-ups with developing innovative, data-driven ideas to tackle use cases provided by our corporate partners. As part of a three-month collaboration, start-ups will draft a proof of concept, with the opportunity to receive up to EUR 20,000 in funding.
No items found.
Flat-Finder
Result
Finding a cozy and individually attractive place to live can be a hassle. Why do people have to hunt down their personal apartment to rent and not the other way around? Help Vonovia SE to recommend the right apartment to the right person in a quick and easy way, based on machine learning mastery. Develop a smart algorithm that offers the best matching apartments to prospective tenants and makes use of a wide range of available information and a self-improving approach.

Use Case

As Europe's leading private residential real estate company, Vonovia SE currently owns around 416,000 residential units in all attractive cities and regions in Germany, Sweden and Austria. The portfolio of it is worth approximately € 58.9 billion. Additionally, Vonovia manages around 74,000 third-party apartments. As a modern service provider, Vonovia focuses on customer orientation and tenant satisfaction. Thus, the company is committed to help current and prospective tenants and their families in finding a great place to live that best suits their individual needs and boundaries. In order to do so, Vonovia offers to sign up for an email service. This daily newsletter offers a personal selection of available apartments just for you and is based on your individual requirements. In addition, the service provides a certain level of exclusivity with respect to the apartments presented to each subscriber. An initial sign-up to the newsletter service is currently valid for a period 14 days and can be extended. The service generates and sends out about 350 individual newsletter emails every day – increase of volume clearly to be expected in the near future. 


However, the currently used algorithm to match prospective tenants with suitable apartments is rather a static ruleset, relying mainly on basic criteria like location, size and price of the apartment. It does not use additional available information, even if the interested person provided more detail. Furthermore, aspects like user response to recommendations from the past or spatial contexts of apartments are not yet systematically incorporated. We expect better results using all available attributes. We also believe actively enriching information would increase the quality of matches even more. Looking for a 120 m² apartment? Maybe apartments close to schools would be more attractive to you.

The currently used algorithm is quite performance intensive and deeply rooted in the company’s ERP system, therefore updates can only run on a daily schedule during night time. We are sure, there are more elegant and flexible ways to complete this task in order to fully exploit its value within the company – potentially even beyond the current email newsletter service application.


As a first step and in focus of the Data Hub Challenge 2021, the goal is to increase the quality and quantity of successful matches, given search profiles of prospective tenants and information by Vonovia on available apartments. As a benchmark, opening- and click-rates of the current newsletter at hand shall be increased. Based on the outcome of this challenge, new features and use cases will be considered. Therefore, the solution should be easily extensible.


What you will need

  • In-depth knowledge of recommender systems or comparable advanced machine learning algorithms
  • Software engineering skills to implement a high-performance MVP fully usable in an operative IT environment
  • Basic understanding of email marketing is required
  • Experience to work with data in an General Data Protection Regulation (GDPR) sensitive environment

Expected result

  • Improved opening- and click-rates or other response metrics of offered apartments in the current newsletter 
  • Overall increased number of matches available to publish to prospective tenants
  • New approach to match both sides in a fast and efficient way using modern ML
  • Extensible architecture for further improvements and/or additional use cases
  • The final solution shall be able to be hosted on premise inside a data center of Vonovia

MILESTONES

The project can be divided into three milestones:


  1. The first milestone is reached when all relevant data is integrated: requirements of prospective tenants, attributes of available apartments, historical matches, response data, contexts, etc.


  1. To reach the second milestone, the matching algorithm needs to be developed. It shall be considered to enrich matching attributes by further information features.


  1. As a third milestone, A/B testing proves successful in regard to newsletter performance measures and number of offers made.

This is some text inside of a div block.
Website
Startup

Let's talk

Felix Schröder
Program Manager DataHub Ruhr

Join the DataHub Network

We have made it our mission to boost data innovation in the B2B and industrial sectors. Become a member of our DataHub Ruhr network to receive all the latest information about exciting challenges and matchmaking events in the Ruhr region.

Let's talk

Felix Schröder
Program Manager DataHub Ruhr