Dutch   |   English

Boompjes 40
3011XB Rotterdam

Last Vacancies

16-01-2020 Zwitserland na na

Welcome to !#CUSTOMER_NAME#!

Asset Management AI/HCI

Vacancy information

Work area

South Holland
Rotterdam of Amsterdam


Full-time and part-time

Level of education





Share |

At the moment we are looking for a:

Asset Management AI/HCI

From03-07-2018 LocationRotterdam of Amsterdam HoursFull-time and part-time Educationna



Thesis topic to develop an artificial intelligence (AI) / human-computer interface (HCI) related application capable of performing or supporting asset management tasks for housing associations.

Ortec Finance Tech Labs
Recently, Ortec Finance Tech Labs emerged in order to lock in innovation through a dedicated Research and Development center in the field of IT. At Tech Labs we investigate the most recent trends and techniques ranging from Blockchain, Virtual/Augmented reality, Machine Learning to the latest front and backend techniques.

Asset management
For the real estate market we develop applications that allow housing associations to determine what policy they should develop and implement for their real estate portfolio. During this process asset management is a crucial investment decision making process that results in a policy of continue exploitation, renovation, sell or demolition for every real estate complex the association has. Asset managers need information about the current state of the complex, such as the current yield, the state of maintenance, the happiness of tenants, whether it is achieving social goals and the sustainability of the complex to help them make investment decisions. The end goal is to create an AI assistant using innovations in both machine learning and human-computer interaction to help our users achieve their goals.

There are several interesting areas for AI and HCI. We could use machine learning to help make investment decisions or even to make investment decisions on its own. We could use new human-computer-interaction techniques to better support interaction with complex applications and lastly using smart algorithms we could monitor progress and provide smart hints and feedback during the Asset management process.

Machine Learning
• We know the characteristics of each complex, and we know whether a housing association currently wants to continue exploitation, renovate, sell or demolish the complex. We could try to create an AI through machine learning capable of automatically determining whether a certain complex should be exploited, renovated, sold or demolished.
• Asset managers also need to know the ‘soft’ effects of their investment decisions, if an asset manager decides to renovate a certain complex they want to know what the effect will be on for instance the happiness of tenants. Right now we are unable to determine the expected ‘soft’ effects the investments, but through machine learning we might be able to determine what that effect will be. We have data about current complexes and current tenant happiness, so we might be able to create a machine learning model that learns which characteristics of complexes make tenants happy. The asset manager can then use this to transform their current complexes into the complexes that the machine learning model indicates as making tenants happy.

Human-computer interaction
• Currently most interactions with our systems work through point & click, yet most humans interact with each other through natural language, which is quicker and allows for much more complex interactions. Supporting interaction through speech has several technical challenges that would need to be solved. Allowing users to interact with complex software through speech (either spoken or written) is difficult but might allow for a much smoother experience. We want to research how we can best use this innovation and how much impact it would have on the user experience. Hardware such as the Hololens (Microsoft, 2018) and Amazon Echo (Amazon, 2018) are available.
• Currently our software is available to our users when requested, but our software does not pro-actively work with our users to achieve common goals. We want to research whether it is possible to make our software monitor how well our users are performing and automatically give advice and recommendations on what actions should be taken at what time. The software should know what goals have to be achieved by what date and monitor those goals automatically, giving notifications and updates when necessary to make sure our users achieve their goals. This might require some of the latest innovations in web development, such as progressive web apps. (Google, 2018)

Previous work
Research has already been done in several areas, mainly in machine learning and human-computer-interaction, both topics provide valuable work to build upon.
Machine Learning
• The machine learning model will have to support or operate in the asset management process of a housing association, the process translates the high-level strategy of a housing association to a concrete policy that can be acted upon. (Gruis, Nieboer, & Brown, 2003)
• Supervised learning techniques such as regression, classification, decision trees and neural networks can all be used for the assignment. For an overview of techniques and how they work see for instance Pattern Recognition and Machine Learning by Christopher Bishop. (Christopher, 2016)
• One of the challenges relevant to the assignment will have to deal with the black-box problem posed by machine learning methods: can we explain why the machine learning model made a certain decision. Our clients will have to be able to explain why certain decisions were made, so for the models to be useful they need to be able to explain somehow in a financial way why they made a certain decision. (Castelvecchi, 2016)

Human-computer interaction
• Cassell talks about some of the requirements and challenges for conversational interface agents, all of which are relevant to our assignment, though solving a subset of the challenges is already sufficient. The challenges include: Understanding verbal and non-verbal input, generating verbal and non-verbal output, using conversational functions such as feedback (nodding) or turn-taking, and providing input to the user on the current state of the conversational and pointers for the continuation of the conversation. (Cassell, 2000)
• There are different research methods available for human-computer interaction, which include both qualitative and quantitative analysis. For the assignment statistical analysis, surveys, usability testing and measuring users are all possible. (Lazar, Feng, & Hochheiser, 2017)
Required Knowledge
For the data-driven assignments understanding of machine learning techniques, both linear and techniques such as neural networks is required. For the UI/UX assignment strong web-development skills are required.

What we offer
We offer a challenging and inspiring work environment with excellent career opportunities in both specialized positions and management. The exact scope of the thesis will be discussed before the thesis starts, we are flexible when it comes to the exact scope and topic. Ortec Finance gives you the opportunity to combine IT, mathematical models and specific market knowledge in your work. In Addition, we will provide:
• Required development tools
• Office-space in Rotterdam or Amsterdam
• Guidance from Ortec Finance Tech Labs

What we are looking for
We are looking for students in the field of IT, that love to do research on innovative topics. Because guidance for these thesis subjects is done through Tech Labs we will only be able to provide guidance for IT students.

Amazon. (2018). Develop for Amazon Alexa. Retrieved from Amazon Developer: https://developer.amazon.com/alexa?cid=a
Cassell, J. (2000, April). Embodied conversational interface agents. Communications of the ACM, pp. 70-78.
Castelvecchi, D. (2016, October). Can we open the black box of AI? Nature News, pp. 20-23.
Christopher, B. (2016). Pattern Recognition and Machine Learning. New York: Springer-Verlag.
Google. (2018). Progressive Web Apps. Retrieved from Google Developers: https://developers.google.com/web/progressive-web-apps/
Gruis, V., Nieboer, N., & Brown, T. (2003). What determines asset management approaches in the social rented sector? ENHR Conference, (pp. 26-28). Tirana.
Lazar, J., Feng, J. H., & Hochheiser, H. (2017). Research Methods in Human-Computer Interaction. Morgan Kaufmann.
Microsoft. (2018). Mixed reality: Your world is the canvas. Retrieved from Microsoft Hololens: https://www.microsoft.com/microsoft-hololens/en-us

Share vacancy    
16-01-2020 Zwitserland na na