Contributions published at Information Management (Gerhard Schwabe)

Blockchain technology is on the rise, and many organizations are banding together in order to form so-called consortia. In order to create a case repository of blockchain consortia that can be queried, a definition for blockchain consortia was established, followed by a struc-tured internet research in order to identify as many blockchain consortia around the world as possible. The identified consortia were characterized according to a set of predefined attributes, imported into a MySQL database and quantified in order to pinpoint similarities and differences among them. In a further step, 15 blockchain consortia were analyzed in a more detailed manner.

Research into different types of ecosystems has long been at the forefront of many economist. While there are different types of ecosystems, especially interesting are entrepreneurial and innovation ecosystems, as the world is ever changing. There have been a variety of theoretical analyses and case studies done in this field of research. What comes with the territory of innovation ecosystems is innovation, one of the latest being blockchain. The technology with a distributed ledger system has garnered much attention over the past years, mainly due to crypto currencies. In this thesis, the two fields will be connected, applying the study of entrepreneurial and innovation ecosystems onto a blockchain innovation system located in Zug, Switzerland. Through the introduction of related work, the development of a framework and a qualitative analysis, the ecosystem’s genesis and development is studied and presented.

Up to now in Sindelfingen city hall mystery calls have been executed, which recorded the telephone reachability of the employees. The greetings spoken by the employees on the telephone were also checked for accuracy by means of an official welcome formula. The manual effort to do this Mystery calls is very high. In addition, the employees receive no feedback on their performance, which is why they do not improve further. In this work a digital trainer is developed, which handles the execution of the calls to the employees. The trainer should not only save the results of a call, but also give the employees a direct feedback. Repeatedly making these calls to the employees will create a training effect for them, what will lead to a further improvement of their performance. The solution developed in this work provides this functionality and can be further developed.

The provision of targeted and comprehensible health information plays an important role in the achievement of patients' disease management goals. Information therapy aims to inform patients extensively so that they can make informed health decisions through a deep understanding. The aim of this thesis is to explore the challenges of information therapy and to investigate to what extent IT can support information therapy. A thorough literature review and 13 interviews with dentists and dental students were conducted to identify the challenges of information therapy and validate designed prototypes. A solution was developed based on the findings, tested in staged consultations and evaluated in a focus group. The results show that information therapy is associated with a number of challenges. The challenges can be patient-related, but also occur at the communicative and organizational level. By using IT during the consultation, the identified challenges can be counteracted.

The economy states a need for data markets in order to solve inefficiencies and the blockchain-based cardossier project desires to design such a data market, thus creating an opportunity to use current data market research concepts in a practical setting. In this thesis, we design a data market model for the blockchain-based cardossier project and simulate it. Additionally, answering how the data should be priced and how the revenue should be distributed. We use experimental simulations and interviews to validate the design. The results indicate that the designed hybrid market mechanism, with a pricing function type particular to a blockchain-based setting, outperforms the additionally designed mechanism and thus builds a foundation for future blockchain-based data market designs.

Social Media is an increasingly popular way of communication within professional communities. In blockchain communities, Twitter is used as the prevalent tool. Researchers are focusing on analysing the key players, interaction patterns, and themes of such communities. Formalised Twitter data offers an ideal mapping of these communities and can be used as a base for further in-depth analysis. For this thesis, a design science research approach has been chosen in order to construct a workbench that aims to facilitate the collection and analysis of Twitter data. The developed artefact offers an enormous efficiency gain by automating the preparation and pre-processing of the data. Keywords: Blockchain, Communities, Social Network Analysis, Twitter.

Blockchain technology is said to have the potential to revolutionize entire industries and systems. One area in particular is considered the best fit for blockchain: land registries. Land registries have several issues around the world, such as faulty registers, complex registration systems and various corruption cases. Blockchain-based land registries try to tackle these issues by creating a secure and transparent registration system through the integration of blockchain. This thesis deals with four projects that are being pursued in Sweden, Georgia, Ghana and Honduras. Within these four case studies, the benefits of a blockchain implementation for land registries are outlined and the aspects are shown, under which such a system can work. By analysing the projects, specific advantages have emerged and three aspects were defined, which must be taken into account for a successful blockchain implementation. The findings form the basis for research on further blockchain-based land registries to identify additional benefits and aspects.

Cardossier is a project throughout Switzerland to enter all possible information of vehicles during the whole life cycle on a blockchain platform. Various advantages for many stakeholders result when introducing such a dossier. In this master thesis, a market simulation is used to test an integration proposal of this cardossier into Autoscout24, a used car platform. For this purpose, a conceptual prototype was proposed, which was then developed within a replicated game platform. The experiment finally was held with real Autoscout24 users. Resulting, exciting conclusions about the values of cardossier, for the market, as well as for users, were affirmed. Also, the results confirm that the integration proposed in this work presents a marketable solution that offers a variety of new possibilities.

During cooperative interaction, participants introduce materials, artefacts, and other individuals into the ongoing interaction. Depending on how this introduction unfolds, the participants may embrace the new element in an easy way or not. If the new element is a collaborative application of interactive software designed to support the interaction, it may or may not improve the collaboration because of how it was introduced. Therefore, understanding and designing the initial interaction is key for unleashing the positive impact of collaborative systems. The literature has identified the fact that humans employ a specific range of behaviors when introducing an element into an ongoing interaction. Those introduction rituals are determined by whether the new element is a human or a material artefact. Introduction rituals involving interactive elements are still underexplored: How do participants introduce and initiate interaction with them? This manuscript explores the introduction behaviors emerging when an augmented-reality collaborative application is being introduced into a financial advisory service. It shows that the participants employ a wider range of introduction rituals during the introduction of this application than they do when they introduce a brochure. Notably, many of the observed behaviors resemble familiar opening rituals typically used when introducing and greeting humans. This supports the computers-are-social-actors argument and provides evidence that introducing a collaborative application has a social rather than a material character.

The cardossier project, launched by the University of Zurich, deals with the use of blockchain technology to reduce information asymmetry in the used car market. The cardossier is a data package that stores the life cycle of a car, based on blockchain technology. For the first time, a variable pricing method, the so called dynamic pricing, was applied to the cardossier data. In addition, a value determination of the data was carried out for the first time. To study the effects of dynamic pricing and the value of the data, an experiment was carried out with students from the University of Zurich. The experiment is a simulation game of the used car market. The investigations provided new insights to reduce the information asymmetry between sellers and buyers on the used car market. Thus, by using the dynamic pricing the market’s fairness on the used car market could be increased. As a result, more used cars, which are characterized by good quality, could enter the market. Furthermore, trading of data from the cardossier could be increased by dynamic pricing. The evaluation of the individual data elements of the cardossier confirms, the value assessments made so far and shows, that the qualitative value of the data for the individual market players is strongly influenced by individual prerequisites and preferences.

As a result of the digitalization, the fields of application of conversational agents (CA) are increasingly spreading into our lives. With its usage in financial advisory services, new spheres in the area of human-computer interaction are now being explored where pioneering work needs to be done. Since the research question in the design science research, such as how the initial interaction with a conversational agent in a financial advisory encounter is to be designed, is still unclear, this research work is dedicated to this issue. Using the process model of Peffers et al. (2007), design-based introductory proposals for conversational agents are developed based on a literature research and a Wizard of Oz experiment. Different types of the visual representation of the natural language interface are distinguished, that either (1) have no visible representation component, (2) are represented by a human avatar or (3) have a non-human avatar. Under these conditions and beside many other results, the present discussion has unequivocally shown that the design of the initial interaction is influenced to a significant extent by the visual representation of the conversational agents natural language interface and its role to be taken in the advisory encounter.

The technological progress drives automation and allows for advanced machine learning applications in many industries. However, digital transformation in the still strongly paper-based healthcare industry is rather slow, as studies have shown. Digital, web-based patient feedback solutions at the point of experience transform the currently paper-based patient surveys and allow the collection and analysis of patient feedback in real-time. These solutions are available around the clock and increasingly used. Therefore, the amount of data can become hard to process and important comments by patients can get lost between irrelevant feedback. With a practical use case of a company providing a real-time patient feedback solution, this thesis explores different approaches on how to present stakeholders from healthcare institutions with the most valuable and relevant textual patient feedback by leveraging methods from machine learning and natural language processing. Based on the current feedback solution and preliminary interviews, requirements are derived and different prototypes implemented and evaluated. The findings from the development and evaluation of the prototypes show that new technologies can be leveraged to effectively indicate relevant comments and support stakeholders of healthcare institutions with the evaluation of textual patient feedback.

Buying a used car can be a stressful situation. The buyer does not act with the same information as the seller, and therefore has to trust data presented by the seller. To prevent this the cardossier, a blockchain data market governed by a consortium, intents to offer a used car’s lifecycle history. The data for this information product is provided by different consortium members who offer it on the blockchain market. With the data available on the market, a variety of other information products can be created. To find such new products, different industries are analyzed to understand best practice. These products are then categorized with the help of a framework, to allow for comparison. With the help of this fundamental knowledge, products specifically tailored for stakeholders of the data market are created. But product creation alone does not create profit. The market needs a structure and products need allocation mechanisms. This thesis presents the optimal market structure for a blockchain based market and highlights the best allocation mechanisms for the offered information products. Furthermore, it has found that the blockchain technology does not influence the form of an information product, although it has a significant impact on the choice of market structure and coordination mechanisms.

The vehicle registration process in Switzerland is characterised by media disruptions, manual work and inefficient data storage. This thesis investigates how this process can be optimized with blockchain technology within the project Cardossier . After the identification of the current process, an optimized process is developed. In this process, physical documents are digitized and process steps automated. The implementation of a prototype shows that the process can be optimized as part of the Cardossier by introducing blockchain technology. The knowledge gained from this work supports the further development of the Cardossier in the area of vehicle registration as well as its goal of bringing trust into the used car market.

Together with Ergon and Recapp, the Information Management Research Group of the University of Zurich is involved in a proof-of-concept project called Heinzelmännli, evaluating the benefit of integrating a conversational agent into bank consultation sessions. This thesis explores and systematically evaluates different approaches for automatic error correction in the output of an automatic speech recognition system. Also, it proposes concepts and architectural changes to further expand the existing framework in functionality, so that all the specified user scenarios can be realized efficiently. Finally, following an exploratory research approach, interesting topics and fields for further research are identified and presented.

In a mortgage advisory meeting paper is still the most used medium, although there exist software applications to help. So far the interaction with these applications disrupt the flow of conversation. Therefore, it is important to understand how the participants behave in an advisory meeting when the technical support is introduced. In order to understand what behaviours are shown and how they affect the interactivity of the conversation, an analysis of recorded advisory meetings was carried out. It was shown that writing and operating the technical artifact often triggers a pause that disrupts the flow of conversation. Therefore, a speech recognition assistant was developed, making handwriting obsolete in order to find out whether it increases the interactivity of the meeting. The speech recognition assistant was tested in simulated advisory meetings and it has been shown that the opening of the meeting becomes more interactive with the use the assistant.

The ongoing utilization of blockchain-based systems can be comprehended as a contemporary para-digm shift that brings lots of innovations to applications that rely on transactions. In recent years, interest in research on the blockchain technology strongly increased and steadily improved its performance, cryptographic security, and reliability. However, its development has opened up new organi-zational aspects as well; decentralized autonomous organizations describe a new type of organization that is grounded on consensus-based, distributed autonomy. The governance of DAOs is fairly unexplored, and this is where this thesis commences. In a case study consisting of Aragon, Tezos, and DFINITY, the principles of governance of DAOs are worked out and discussed focusing on case-specific disparities, basic commonalities, and explorative meta-considerations. With the incorporation of grey and academic literature as well as practically relevant assessments through semi-structured interviews with experts, a meaningful overview over the internal and external governance structures of DAOs is obtained by continuously integrating the sources with qualitative codings. The findings of this study contain statements about actors, their interests, incentive structures, control and coordination mechanisms, technical considerations, interdependencies among layers, the organizational structure of the investigated cases, as well as external influences from off-chain entities. In the context of an on- and off-chain continuum, it appears that DAOs currently provide many mechanisms that might enable autonomous decision-making but at the same time find themselves in a legal dilemma and are strongly influenced by the interests of external stakeholders. As soon as DAOs have overcome the hurdles and challenges of a disruptive innovation, they can develop their full potential and change the way the society thinks about organizational constructs in the long-term.

The aim of this thesis is to investigate how a technical implementation of a networking artifact for a business lunch app can look like. Furthermore, it is investigated how such an artifact can support networking within an organization. To achieve this goal, a networking artifact is created in the form of a web application. The user-friendliness of the application is tested by means of user tests, and the findings are incorporated into the application in a further iteration. The possible effects of using such an application to promote networking are examined through interviews.

In collaboration with the University of Applied Sciences and Arts Northwestern Switzerland and the Hypothekarbank Lenzburg, the University of Zurich developed two digital solutions to support financial consulting. These products are called LivePaper and BankNotes and should combine the qualities of the traditional consulting approach with the opportunities that modern technology offers. The two products were already used as part of simulated consulting cases. The data that was recorded during these sessions is available for this thesis and is the basis for the research about learning in financial consulting. The focus lies on the difference between the learning success that really happened and the one from a selfassessment. These findings are then interpreted and compared to the knowledge of the already existing literature.

Crypto Valley is one of the most important blockchain innovation ecosystems. As in other innovation ecosystems, the Crypto Valley has people that play a central role. These people, also called leaders, are an important part of the innovation ecosystem because they have an influence on the decisions and opinions of people in the ecosystem. Therefore, this paper examines how these leaders differentiate and represent each other. For this purpose, Twitter profiles and publications on fifteen leaders from the Crypto Valley are analyzed. For the analysis, a framework, derived from leadership theories, is presented, which should allow a comparison between the individual actors. The goal of this work is to present a typing that not only describes the leaders of the Crypto Valley, but also applies to other innovation ecosystems in the world.