## 2019

• A Consistent Pricing Model for Swaptions and Forward Volatility Contracts on Swaptions

### A Consistent Pricing Model for Swaptions and Forward Volatility Contracts on Swaptions

In the financial markets many different types of contracts are traded in order to control the risks of a portfolio. One of the most traded financial product is the swaption. A risk associated with a swaption, is the so called volatility risk. A tool for hedging this risk is the forward volatility contract, which allows an investor to lock in the forward volatility embedded in the shape of the volatility surface observed today. The objective of this thesis, is to construct a mathematical model, which is able to price these forward volatility contracts consistent with the pricing of vanilla swaptions.

Student: Asbjørn Snitkjær Meinhardt
Education: Mathematics - Economics

• Discovering statistics of family formation using survival analysis

### Discovering statistics of family formation using survival analysis

Given available transaction data and tables of family relations I seek to detect when people met based on co-spending. By use of survival analysis I will then explore statistics between family formation events like moving in together, marriage and getting children. Lastly I will describe the change in spending patterns which occur throughout the different stages of family formation.

Student: Jeppe Skals
Education: Mathematics - Economics

• Data Driven User Experience

### Data Driven User Experience

The goal of this project is to create a way for Danske Bank employees, to gain insight in customer behaviour, and get a better understanding of how they experience the use of Danske Bank’s applications and services. To accomplish this, user interactions in the form of general inputs and mouse movements, are gathered as information. From this information patterns can be detected and recognised as general use. This knowledge can then be used improve the customer experience. Either when we make design choices to an application, or by reactively enhancing the experience by making helpful on the spot suggestions.

Students: Asbjørn T. Gyring and Jonas Buus
Education: Bachelor of Engineering in Information Technology

• Modelling the financial markets using regime switching models

### Modelling the financial markets using regime switching models

The financial markets often seem to move between periods of different behaviour, also called regimes. A classic example is periods of low volatility and periods of high volatility. Attempting to model  such a market without incorporating the possibility of a regime switch can lead to a model that does not give a clear view of the markets behaviour. The purpose of my thesis is to investigate the class of regimes switching models known as Hidden Markov models and their capability of modelling the financial markets for practical purposes.

Student: Bo Sørensen
Education:
Education: Mathematics - Economics
• Working with UX in agile IT development

### Working with UX in agile IT development

UX and agile are two popular concepts within IT development these years. In order to better understand the intersection between the two disciplines and how they work in practice, the thesis will look into how the processes, activities and organization of user experience (UX) design compare to those of agile IT development discussing differences and similarities and how these influence development processes in practice. Could understanding of these possible challenges and opportunities compliment current UX design processes?

Student: Lena Buhl
Education: Information Science

• Offline payments

### Offline payments

We aim to develop a system that allows third parties to integrate with MobilePay, by exposing an open interface. Through this integration, third party systems can implement their own payment solution. Prioritizing the properties they find important - be it offline connectivity, easy authentication or some other feature. MobilePay would simply put, handle the transaction while leaving the customer interface to the integrators.  To demonstrate this open interface, we would like to create a system focusing on offline payments.

Students: Jonas Junker Zarling, Mikkel Kastrup and Peter Runge Christensen
Education: Bachelor of Engineering in Information Technology
• Predict Customer Behavior

### Predict Customer Behavior

The purpose of this project is to try and predict, what information would be relevant to show to the customer in different Danske Bank applications. Feeding user data to a machine learning algorithm, we hope to recognise a behavioural pattern, and thereby create a more personal application experience.

Students: Tobias Kjær Henriksen, Mikkel Bitsch Overgaard and Flemming Lundahl
Education: Bachelor of Engineering in Information Technology
• Automated static reference datanormalization system

### Automated static reference datanormalization system

The purpose of the project is a proof of concept of a working automatic instrument data collection system, allowing the current workflow of data acquisition to be simplified and sped-up. The resulting system should be providing real value to end users, namely the developers behind the Static Data department and investment bankers inside of Danske Bank. The value will lie in being a user-friendly way of specifying rules and weights, used while deciding the vendors of instrument data, at the same time building a framework for further development and extension of features.

Students: Grzegorz Jacek Kot, Sherzad Abdulaziz Ali & Aayush Dhakal,
Education: Bachelor of Engineering in Information and Communication Technology

• User Outcomes in Danske Bank

### User Outcomes in Danske Bank

As the biggest bank in Denmark, Danske Bank is a cooperation which includes thousands of internal systems. Through the years the bank has mastered the ability to gather user needs and implement the necessary features. But one thing they lack is tying up the process by looking into how their users interact and behave in their systems. In other words, they do not know if what they have invested time and money on implementing is being used as intended.
This project focuses on answering the question: How might Danske Bank measure/forecast how their users use and interact with their systems and analyze if the users use it as intended?

Student: Iman Chelhi,
Education: Bachelor of Engineering in Software Technology

• RELIABLE, EFFICIENT, AND SCALABLE PUB/SUB MESSAGE STREAMING SYSTEM

### RELIABLE, EFFICIENT, AND SCALABLE PUB/SUB MESSAGE STREAMING SYSTEM

Our goal is to develop a distributed message streaming system that is dynamically scalable and is performance comparable to Apache Kafka. The system stores messages to support slow and faulty consumers; thus, allowing the consumers to re-consume messages. The system should scale based on the workload to minimize resource consumption and energy usage.

Students: Anders Meidahl Münsberg & Nicklas Nielsen,
Education: Master of Computer Technology

• Consumer driven contracttesting tool for the mainframe endpoints

### Consumer driven contracttesting tool for the mainframe endpoints

In Danske Bank a lot of business logic runs on a mainframe system that has been running for a long time. This means that some technology is of older date. For instance testing on the banks mainframe is not present or poor around lot of the old but important logic. The logic is very important for the banks daily business. So every change needs to be verified. Verification is at the moment done manual for the most. Therefore this project will make a consumer driven contract testing tool for the mainframe endpoints. A standard test tool don’t work out of the box since the banks setup is special.

Student: Søren Katborg,
Education: Bachelor of Engineering in Information and Communication Technology

• MobilePayMarketplace

### MobilePayMarketplace

The goal of the project is to add a marketplace functionality to the MobilePay app. This will give users of the app the ability to buy products, such as food, to be picked up from small vendors, digital tickets for any type of event and physical products through simple webshops.
MobilePay Marketplace will give merchants new opportunities to sell their products in front of a very large userbase and configure their shops through an easy to use website.

Students: Daniel Pat Hansen, Søren Bech, Martin Haugaard Andersen & Fatima Kodro
Education: Bachelor of Engineering in Information and Communication Technology
• Automated Test of anAuthentication and Signing System.

### Automated Test of anAuthentication and Signing System.

This project aims to produce an automated test suite for an existing system of high criticality within Danske Bank - an authentication and signature system, which enables other systems, both internal and external, to enrol through a standard protocol (OpenID Connect/OAuth2). The product should build confidence in the developers, that the software changes do not break the existing system, thus assisting in maintaining a high success rate when deploying new software.

Students: Nickolai Lisberg Lundby & Rune Aagaard Keena,
Education: Master of Computer Technology

## 2018

• Detecting fraudulent behaviour on mobile devices using machine learning

### Detecting fraudulent behaviour on mobile devices using machine learning

The purpose of this project is to see if it’s possible to detect fraudulent behaviour using data collected from smartphone sensors. Feeding this data to a machine learning algorithm, we hope to recognise a behavioural pattern, and thereby detect suspicious behaviour.

Students: Halfdan V. Bjerre, Ahmad Sabah, and Jacob  M. Hansen
Education: Informations- og  kommunikationsteknologi

• Receipt manager

### Receipt manager

The goal of the project is to create an application that will help Danske Bank’s employees to minimize the amount of time spent on receipts. This is accomplished by making a mobile app so it is going to be easy for the employee to take care of every receipt on the spot.

Students: Theis  E. Hansen, Nina Mølgaard and Frederik Andersen
Education: Informations- og  kommunikationsteknologi

• Corporate Passport

### Corporate Passport

The project will essentially be an application that aims to reduce the resources needed, when banks get new company customers. Our application will be a web application on which a company can get a digital stamp of approval when for example Danske Bank has scrutinized their economy and trusts the customer.

Students: Laura Aggerbo Johansen, Frederik Rosenørn Jensen and Alexander Thulstrup Bruhn

Education: Informations- og  kommunikationsteknologi

• VIA Wallet

### VIA Wallet

These days our wallets are full of access cards, credit cards, club cards, etc. Taking what MobilePay started by managing to remove the need for credit cards, the project will prove that it is possible to start moving away from all of the cards that clutter our everyday lives and turn them into a simple mobile application. The project will use VIA UC Horsens as a testing ground, since their cards are used for several things such as printing, opening doors, etc.
etc.

Students: Konstantin Vasilev, Lyubomir V. Zhivkov and Kristiyan  Kyukov
Education: Informations- og  kommunikationsteknologi
• Analysis and Forecasting of Finacial Time Series Using Machine Learning

### Analysis and Forecasting of Finacial Time Series Using Machine Learning

Time series can be analysed and predicted using machine learning in a supervised learning framework. The time series of interest are price processes and historical trades obtained from the department of rates, options and inflation trading. The key object for us is to establish whether or not these price processes have a certain pattern after trading with certain customers.

Student: Kirstine Liedecke
Education: Mathematics Economics

• BudgetBud

### BudgetBud

We are developing an app that helps the user pay off loans and distribute expenses in a manner best suited for him/her. We utilize the user’s e-banking data to create a personalized budget. When the loans should be paid off, the amount the user wants to save each month and expense distribution are all control parameters granted by this app. In other words, we strive to grant the user financial control and stability.

Students: Victor Busk & Andreas Lüdemann
Education: Bachelor of Engineering in Information Technology

• Fraud Detection at MobilePay

### Fraud Detection at MobilePay

Today it is free for private users to use Mobilepay and the transactions fees are paid by the businesses. Thus there is an incentive for small business owners to use their private Mobilepay account for business purposes and thus circumvent the fees. My thesis uses machine learning and transaction history of private users and businesses to find the fraudsters that use private accounts for business purposes.  Ideally, these individuals can then be converted into business accounts and pay transactions fees to Danske Bank.

Student: Anders Damsgaard
Education: Mathematics Economics

• How the Banking Industry Can Leverage from Cooperating with FinTechs to Strengthen Their Digital Transformation

### How the Banking Industry Can Leverage from Cooperating with FinTechs to Strengthen Their Digital Transformation

The financial service sector is facing a massive change undergoing a digital transformation due to external factors such as an increased demand of customers and changing regulations - e.g. PSD2, increased focus on security, new technology driven innovations, and a heightened competition also with the emergence of FinTechs. This is indeed the case for Danske Bank, where digital transformations is becoming more and more vital to ensure their survival in a digital world. Our focus in this thesis will be on how banks can cooperate with these FinTechs proactively in response to the external factors to strengthen their own position in the digital transformation.

Student: Nina Madsen & Kaja Skjerbek Hemmeløff Andersen
Education: Information Technology

• MobilePay Interactive

### MobilePay Interactive

AR is already a fast growing branch due to its positive perception on immersive interaction, which would definitely influence and change the marketing industry.

The project "MobilePay Interactive" aims to build a framework module for developers and facilitate a different type of Augmented Reality environment setup. This would help create different marketing materials mainly aiming for merchant users. The combination between the AR interaction processes with MobilePay's capabilities of rapid transfers, can lead to another type of advertising embedded with a transaction mechanism.

Students: Alexandru Diaconu & Nicolai Sand
Education: Bachelor of Engineering in Information and Communication Technology

• Post-Quantum Cryptography

### Post-Quantum Cryptography

Quantum computers would make it possible to solve previously unsolved computational problems. While they have great potential, they also represent a great security threat Already in the 90’s it was shown that they would make it possible to do prime factorization much faster than on normal computers. This would break modern cryptography.
Therefore, I will be looking at how we can create digital signatures that are secure against quantum computers.

Student: Morten Zdrenka Christensen
Education: Computer Science

• Virtual Reality Telepresence Technology

### Virtual Reality Telepresence Technology

The importance of productive and effective interaction with colleagues in remote locations is increasing with outsourcing and companies becoming more international in their daily operations. The 3D tool being developed is a new proposal for immersive interaction with colleagues in remote locations. It is a proof of concept of how mixed reality technologies can enhance the telepresence experience already being used in the bank.

Student: Esteban Javier Struve Taguaruco
Education: Computer Engineering

• The Behavioral Economics of Saving and Investing on the June Platform

### The Behavioral Economics of Saving and Investing on the June Platform

Younger people are notoriously bad at saving, let alone investing, their money. While the most common barriers to doing so involve lack of capital and knowledge about investing, many new financial products such as ETFs and index-funds have been established to make the process both easier and less capital intensive — Danske Bank’s new June investment platform requires as little as 100 DKK. With such products coming to the market, the main challenge is how to change the behavior of young people to start saving and investing their money in Danske Bank’s June, and what strategies Danske Bank can employ to onboard more young people.

Student: Moma Vujisic
Education: MSc. International Economic Consulting

• Scale Development for Evaluating Innovation Capabilities on Team Level

### Scale Development for Evaluating Innovation Capabilities on Team Level

Based on innovation management theory the intention of the thesis is to develop a scale for measuring the innovative capabilities of teams. Through a mixed methods triangulation design the thesis aims at analysing different factors influencing team performance to determine which factors are the most important for team performance, in relation to innovation. If the thesis is succesful in describing the most important factors for building innovation capabilities it can aid the bank in developing high performance innovative teams.

Student: Carsten Brinck
Education: Innovation Management

• Proposal for a cross-Nordic blockchain collaboration

### Proposal for a cross-Nordic blockchain collaboration

The Thesis will seek to explain the opportunities for a succesful cross-Nordic blockchain collaboration seen from a business perspective. This will be done by creating a business model containing possible suggestions on how responsibilities and roles in a Nordic cooperation should be distributed, in order to generate value for all partners. The findings will then be used to recommend a strategic proposal for MobilePay.

Students: Mads Rohrdanz Jepsen & Nicolaj Solvang Christensen

• Analysing clickstream data from Danskebank.dk

### Analysing clickstream data from Danskebank.dk

Using weblog data, we present a pipeline for distributed processing of large quantities of customer behavioural data for predictive analysis. We show that using state of the art methodologies from natural language processing, together with recent trends in deep learning, results in a more successful modelling of customer behaviour, compared to hand engineered features and traditional machine learning methods.

Students : Xuwen Zhang Andreas Borgstad, Christoffer Thrysøe
Education : Computer Science.

2017

• Reliable and Lightweight Message - Oriented Middleware Based on UDP

### Reliable and Lightweight Message - Oriented Middleware Based on UDP

Our goal is to implement a reliable and lightweight message oriented middleware based on the User Datagram Protocol. This should lead to a middleware that may achieve the same reliability and synchronisation capabilities as a message oriented middleware using the Transmission Control Protocol.

Student: Peter Kragelund & Réne Rotvig Jensen
Education: Computer Enginnering

• Mobile App for Corporate Card

### Mobile App for Corporate Card

The project develops a mobile app for customers with a corporate card. The purpose of the app is to make it easy to get an overview of your corporate card and administrate it. Some of the desired functionality is desired is to get an overview of your transactions made on the card, to block the card, see your value-added services, etc. Besides the base functionality, we may include the option to automate the process of documenting your transactions made on the card.

Students: Peter Nethan Henriksen og Amir Kodro

Education:  Informations and communication technology

• Learning approach for Fraud detection

### Learning approach for Fraud detection

In fraud detection, fraudulent transactions are far fewer than genuine ones. This type of data set is also known as an imbalanced data set. Implementing machine-learning algorithms for imbalanced data sets introduces different complications. This project will look into these complications and discuss how to improve performance for such data sets. Finally, we will design, implement and test different machine-learning algorithms for fast and reliable fraud detection.

Students:  Simon Østergaard Kristensen & Torben Werenberg Vogt

Education: Computer engineering

The project explores if and how academic input from talented students inspires innovation at Danske bank. The overall goal is to optimise the Talent Box for the benefit of both students and Danske Bank. This is achieved using an empirical design process consisting of three primary tools: Critical observation, interviews and being a user.

Student: Carsten Johannsen Brinck
Education: Cand. Merc. Innovation management

We strive to create an intelligent budgeting and advisory system that enables private customers to manage their finances, using big data analysis tools.
Utilizing the banks vast amount of customer data, empowers our application to help customers implement best practice financial behaviour in their everyday life.

Students:  Martin Boje Carpentier,  Joachim Dam Andersen & Bjørn Nørgaard Sørensen

Education:  Informations and Communication engineering

• PocketPay

### PocketPay

In collaboration with MobilePay we are developing an app for children between 8-13 years. The app can transfer money to the cafeteria or similar, while the parents still have control on the amount that can be spent. With this app, the child has a safe environment to learn about money, while the parents no longer has the need for physical money.

Students: Anders Meidahl Münsberg & Jakob Laursen

Education: Informations and communication technology

• Payments in Connected Cars

### Payments in Connected Cars

Developing business models for different use-cases of in-vehicle payments such as McDonald´s drive-in and gas refuelling at petrol stations.

Student: Yousef Mustapha
• Machine Learning for Card Fraud Detection

### Machine Learning for Card Fraud Detection

Machine Learning is a subset of Artificial Intelligence and encompasses many algorithms, which are able to learn from data and then make decisions on new data. As a lot of data is collected about card usage, Machine Learning is possibly a solution to the problem. In this thesis, several Machine Learning algorithms will be analysed, tailored, tested and compared to see which - if any - are good at detecting card fraud.

Student: Niclas Gleesborg
Education: Computer Engineering
• Understanding Customer Determinants of Adopting Mobile Payment Services

### Understanding Customer Determinants of Adopting Mobile Payment Services

The thesis will seek to explain how MobilePay, from a customer value perspective, will strengthen its position against future substitutional services on the market for mobile payment. The found and defined value determinants of market actors (customers/merchants), will be analysed in order to make strategic recommendations.

Students: Patrick Schacksen and Frederik Arve
Education: Cand.Merc.Mark
• Increasing Economic Awareness by Visualising Financial Interactions in a Home Context

### Increasing Economic Awareness by Visualising Financial Interactions in a Home Context

Over the last couple of decades the phenomenon of money has been detached from its physical form (coins, bills, checks, etc.) and moved into the digital world. Seeing how this change has altered the relationship between humans and money, our project seeks to create a system for a domestic context where multi-user interaction can take place creating an improved personal banking experience – and ultimately increase the users' economic awareness.

Students: Lasse Legaard and Christian Lorentzen
Education: IT Product Development
• Distributed Enterprise Applications

### Distributed Enterprise Applications

The requirements for enterprise application availability, performance and maintainability are constantly rising due to high amounts of information with unpredictable and varying load, as well as a demand for failure free and feature rich applications.

Student: Jørgen Heinsen
Education: Computer Engineering
• Manual, User and Design

### Manual, User and Design

My thesis is about manuals. Simply put: Great manuals make even complex tasks and designs easy. How do you make them? When are they needed? And why? I have teamed up with Jeppe M. Thastum as my supporter and will be working with ATM’s.

Student: Martin Thorhauge
Education: Information Science
• Designing Conversational Artificial Intelligence Interfaces

### Designing Conversational Artificial Intelligence Interfaces

The thesis explores the challenges in designing text-based conversational interfaces between artificial intelligence agents (e.g. Chatbots) and users.

Student: Jørgen Hovind
Education: M.Sc IT in Digital Design
• Fraud Detection

### Fraud Detection

Automation of fraud detection using machine learning. We are making a prediction model which utilises the available transaction data to infer if a transaction is fraudulent or not, along with a likelihood of fraud. A substantial part of the project is the selection of the best performing model measured by a certain error measure.

Students: Mathias Kvist Aarup, Troels Lægsgaard and Christian Bomholt
Education: MA Math-Econ
• Employer Branding

### Employer Branding

The focus for this thesis was to explore the perception of and experience with an employer brand and how this balance influences employees' role as employer brand ambassadors. The research was conducted as an empirical case study at Development Center Aarhus.

Student: Line Søborg Rasmussen

Education: MA Corporate Communication