Qiaowen Wang, Intraoperative Surgical Tool Pose Estimation Based on Fluoroscopic Landmark Detection, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
Optical surgical instrument tracking systems have been invented and polished for decades, yet due to a variety of reasons, their popularity is still not reaching a satisfactory state. Among all the obstacles, the high monetary expense of introducing such systems to the operating rooms is considered a significant impediment. However, one type of equipment that is indispensable for arming an operating room is the intraoperative imaging system. Therefore, the idea of providing intraoperative navigation based on intraoperative fluoroscopy has been developed and named X23D. This study aims to explore the potential method to support the realization of X23D, especially the feasibility of integrating an advanced neural network into the pipeline. Learning from existing navigation systems, a prototype of a reference frame for locating instruments in fluoroscopic images is sketched. We focus on the potentiality of locating the reference frame using a single fluoroscopic image and 6 landmarks. We performed error stimulation to build our preliminary expectation of the neural network's performance. We defined criteria that can be used to filter the pose of the reference frame in 2D images, which can separate poses into challenging and unchallenging poses. Based on the criteria, we generated the data needed for model training, validation, and testing. The neural network structure that can fulfil the performance expectation is also designed and trained. Even though the accuracy of the proposed approach still craves improvement before it can be deployed into practice, the value of this project as a stepping stone is not to be neglected. |
|
Dominique Hässig, Visual Analysis of Weather Events Observations based on Crowd-sourced Data by MeteoSwiss, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
Human digital traces (text, image, video) about weather events are a unique and novel resource that could potentially help for early detection and tracking of high-impact weather events.
This thesis proposes the first step toward this ambitious goal by visually analyzing the Online citizens’ reports of the app MeteoSwiss of the Federal Office of Meteorology and Climatology MeteoSwiss.
In October 2021, MeteoSwiss launched in this app a new feature, the Meteo reports. In this, users can report the weather around them and add pictures of the weather if they want. Based on these reports, we created the visualization analysis tool CitizenWeatherVA (Citizen report on Weather Condition Visual Analysis) to analyze them. With the help of the developed tool, based on selected use case events we then analyzed how the crowd reports describe weather events. |
|
Andrea Bergesio, Frictional costs in insurance, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Dissertation)

|
|
Artemis Ioanna Kardara, Detecting Related Stack Overflow Posts for Discord Conversations, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
Programming Question-Answer Communities are at the forefront of modern Developer Work. While existing research has historically focused on finding duplicate posts in Stack Overflow, there is currently little to no research for near-duplicate detection which focuses on Platforms like
Discord. To address this gap, we consider existing state-of-the-art approaches in the field and additionally replicate and apply the most promising to the new domain. A Discord-Stackoverflow dataset in the Java domain is constructed and utilized, while for the evaluation both a classification and retrieval task are considered. The experimental results show that a direct transfer from an existing domain and model is feasible to an extent, while classification and retrieval in the new domain reach up to 77% Precision, 70% F1-Score, and 80% Recall-Rate depending on the length of the examined Input Sequence. |
|
Adrian Lars Benjamin Iten, Classification of Symbols handwritten by Children, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
Classification of handwritten symbols like digits or letters is well-studied. This master thesis focuses on the novel domain of Kinderlabor computer science exercises. It contributes a dataset of symbols handwritten by children in the corresponding domain and evaluates different classification models. This thesis is part of a larger project which aims to implement an automated correction process for those exercises using existing localization and correction algorithms in combination with a symbol classification model developed in this thesis, which classifies each symbol independently.
The dataset is collected using different types of exercise sheets, of which the data collected from productive exercise sheets have a significant drawback of lacking or even entirely missing some symbols. To overcome this limitation, a very time-efficient exercise sheet that contains all symbols is contributed. This thesis starts by inspecting the data, where different characteristics of the handwritten symbols and the prevalence of certain symbols are studied. Then, three different data splits are defined, including a data split to assess performance in the productive application scenario, where the model has to classify symbols from new school classes.
Two important characteristics of the dataset are the label imbalance and that the dataset contains a certain amount of unknown symbols, making the classification problem an open set classification problem. In an open set classification problem, a classification model must not only correctly classify a set of known symbols, but also reject unknown symbols.
Two types of experiments are then performed on the dataset: First, baseline models for correctly classifying all known symbols, including empty fields, are created that are not explicitly trained to reject unknown symbols. Subsequent experiments are performed to evaluate the ability of the models to reject unknown symbols while maintaining good performance on the prediction of known symbols.
As existing work lacks an open set evaluation metric for imbalanced datasets, an adaptation to the existing open set classification rate curve is contributed and used throughout the experiments. |
|
Falko Paetzold, Timo Busch, Sebastian Utz, Anne Kellers, Between impact and returns: Private investors and the sustainable development goals, Business Strategy and the Environment, Vol. 31 (7), 2022. (Journal Article)
 
We investigate the expectations of wealthy private investors regarding the impact and financial return of sustainable investments. Our paper focuses on the sustainable development goals (SDGs) as a framework for investors' attempts to create impact. We analyze the behavior of 60 high-net-worth individuals (HNWIs), a powerful yet overlooked investor segment. Our results show large allocations in line with the SDGs, which demonstrates these investors' aim of achieving real-word changes. Furthermore, we show that these “impact investors” have a clear preference for SDGs that are associated with high financial returns. As such, we confirm that both impact and attractive financial returns are expected. Our findings provide rich, deep insights into how HNWIs practice impact investing and their underlying motivations. We outline practical implications for different stakeholders, notably regarding the fact that financially attractive SDGs are likely to attract substantial amounts of capital, with other SDGs remaining underfunded. |
|
Andrea Meier, IVIE-Docs: A Visual-Interactive Tool for Information Extraction from Documents through Clustering and Data Labeling, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
Information Extraction (IE) deals with the task of extracting targeted information from documents, such as the invoice amount in an invoice. In order to apply IE in practice, a corresponding machine-learning model must first be trained, for example a Named Entity Recognition (NER)
model. This poses several challenges: First, the more specifically the models are trained on a concrete document template, the better they are, which requires that the documents be sorted before training. Second, the documents must be annotated by a human. Both of these are time-consuming and repetitive tasks that do not utilize the human’s potential. To address these issues, I have developed IVIE-Docs, an Interactive Visual Information Extraction tool for Documents that includes a clustering component and a NER component to complete the process of training NER models. The clustering component allows users to quickly group their documents. In the NER component, active learning principles are used to identify those documents that can train the NER model the fastest. Users can decide which document they consider most useful based on multiple information sources for active learning. A particular challenge here is that clustering occurs at the document-level, while NER is trained at the word-level. Moreover, in classical active learning, one instance of the same granularity as the prediction is proposed at a time. This was not practical in my approach, since not
only a single word should be labeled, but a complete document. In IVIE-Docs, two measures help to close the granularity gap. A new document layout vector based on layout information of the individual words created a consistent basis between the clustering and the NER model. Second,
the individual word predictions are aggregated at the document level to enable cross-granularity active learning. IVIE-Docs was tested in two studies with a total of 6 subjects. The results show that users were able to cluster their documents based on the document layout vector and that they achieved better results using the active learning components with fewer labeled documents than with a random selection. |
|
Carlos Kirchdorfer, The influence of a conversational agent on the efficiency and effectiveness of financial advisory meetings: An analysis of the conversational agent “Heinzelmännli” and ist different development stages, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Bachelor's Thesis)
 
In recent years, the research field of conversational agents has made more and more progress. We have now reached a point where conversational agents are so advanced that they can be used in financial advisory encounters to support advisors in their work. The research project around the conversational agent Heinzelmännli has the goal of doing just that, but it has not been evaluated yet, what influence Heinzelmännli has on conversational efficiency and effectiveness. This is precisely where this work closes an important research gap. First, efficiency and effectiveness factors were defined, which can be used to measure conversational efficiency and effectiveness. Then, these factors were used to measure the conversational efficiency and effectiveness of traditional financial advisory encounters (without Heinzelmännli), financial advisory encounters with an optimal version of Heinzelmännli, and financial advisory encounters with a non-optimal version of Heinzelmännli. Finally, the data from these three settings were compared. This thesis concludes that a differentiated picture emerges concerning conversational efficiency and effectiveness: While the settings with the different versions of Heinzelmännli seem less efficient than the traditional setting when looking at the factors, Heinzelmännli seems to promote conversational effectiveness because the common understanding among the participants seems to increase. These observations are stronger for the non-optimal setting than for the optimal one. This means that although the non-optimal setting is less efficient than the optimal one, it increases conversational effectiveness more. From the results, it can be interpreted that it is important to minimise the error rate and processing time of input for digital agents and that it is important to allow clients and advisors to share knowledge by using the artefacts. |
|
Severin Kunz, Novel Artificial Intelligence Techniques and System Calls to Detect Heterogeneous Malware Affecting IoT Spectrum Sensors, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
he spreading of IoT devices yields new attack vectors for hackers. In addition, the connectivity of IoT devices increases the potential damage to IoT systems. Therefore, detecting malware on such systems is crucial to limit the damage. Some years ago, Machine Learning combined with behavioral fingerprinting which takes information from the devices’ state has superseded file-based malware detection. This thesis concentrates on system call based malware detection and entails the following main contributions: Firstly, it extends malware detection by enabling the classification of specific attack phases of malware. Secondly, it evaluates the potential of Deep Learning models in the area of system call based attack phase detection in IoT devices and compares it with a Neural Network serving as a baseline model. Finally, the thesis assesses a TF-IDF based adapted preprocessing technique (TF-DF) for system calls, that seeks an enhanced representation of the most expressive system calls. For these purposes, a dataset consisting of system calls coming from a Raspberry Pi connected to a radio frequency network has been created. From the system calls of this dataset, eleven different attack phases stemming from four malware types (backdoor, botnet, ransomware, and rootkit) and one benign phase have been deducted. The classification results of the Neural Network model have significantly outscored the results of the implemented DL models. In combination with the proposed preprocessing technique TF-DF, an F1-score of 99.2% has been achieved then applying it on system call sequences with differing lengths. In a final step, the models have been evaluated with receiving equal length system call sequences where TF-IDF outperformed TF-DF and yielded an F1-score of 78.42%. |
|
Ning Xie, Quantifying the Trustworthiness Level of Federated Learning Models, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
In the last decade, the rise of Deep Learning (DL) in the development of Artificial Intelligence (AI) has greatly improved the performance of AI models which are becoming increasingly relevant as a support to the human decision-making process. With the ever widening spread of AI applications powered on Big Data, centralized machine learning became challenging due to the existing data silos in many industries where data contain sensitive information. The rising concern for data privacy in AI is promoting the development of privacy-preserving Machine and Deep Learning (ML/DL) techniques such as Federated Learning (FL) where model training is performed collaboratively by distributed data contributors in a decentralized manner. FL enables data privacy by design since local data are not exposed.
The increasing interest and adoption of FL systems prompt the need to investigate the ability to trust the decisions made by FL models as compared to centralized machine learning. There is a large body of existing literature on the topic of Trustworthy AI where the requirements are drawn out for an AI system under the five pillars of trust: i) robustness, ii) privacy, iii) fairness, iv) explainability and v) accountability. These pillars were developed in the context of traditional ML/DL systems. As the attention of AI shifts to FL, more efforts are needed to identify trustworthiness pillars and evaluation metrics relevant for FL models. This work analyzed the existing requirements for trustworthiness evaluation in AI and adapted the pillars and metrics for state-of-the-art FL models.
A comprehensive taxonomy for Trustworthy FL is proposed as a result of the analysis. Based on the taxonomy, an evaluation algorithm, FederatedTrust, was designed and implemented as a third-party Python library which can be imported as a plugin to an FL development framework to evaluate the trustworthiness level of FL models. The FederatedTrust library harnesses the meta data and configuration settings of FL models gathered from the development framework and generates inputs and outputs for trustworthiness analysis based on the metrics identified in the taxonomy. At the end of an FL training, a report containing the trust scores of each metric and pillar that make up the aggregated trustworthiness level is generated for the FL model created.
The report helps to identify the areas impacting trust within the model configuration and execution so that improvements can be made to make the model more trustworthy. Validation of the algorithm was conducted in the form of experiments to test the usefulness of the trustworthiness report generated by FederatedTrust under different FL settings. Observations and discussions were made on the experiment results to analyze what can be improved in the future development of this evaluation framework for Trustworthy FL. |
|
Alexandru-Gabriel Petrescu, The Information Content of Currency Options, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)

The information content of equity options has been a topic of much research, evidence showing that there is indeed forward looking information that can be obtained about the underlying securities’returns by analyzing options data. While in the case of equities data is easier to obtain, less is available in the case of currency options and therefore there is significantly less research covering this area. This thesis attempts to find what information can be obtained about future currency returns by analyzing relevant and available foreign exchange options. My goal is to find what variables influence these returns and in what way, and then build a model that predicts upcoming positive and negative price jumps. Lastly, I try to incorporate everything into a model for the currency returns themselves with the use of Machine Learning techniques. In the end backtests for strategies using the derived information will be presented and discussed. |
|
Tobias Frauenfelder, GoDDSSy: A DDoS Signaling and Control Network based on GossipSub, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Bachelor's Thesis)
 
Distributed Denial-of-Service (DDoS) attacks occur daily and are growing in size due to the Internet of Things (IoT) popularity. Since many of these devices are designed to be insecure, botnets can use them to launch large-scale attacks. DDoS attacks are highly distributed; thus, the best counter plan also includes a distributed defense to reduce attacking traffic at several locations that may be closer to their origins. This thesis tries to develop a DDoS signaling network called GossipSub DDoS Signaling System (GoDDSSy), which operates in a trusted environment and is agile and resilient.
The reader is first provided with a theoretical overview covering the fundamentals of DDoS attacks and publish-and-subscribe systems. Furthermore, a summary of related work is given. Additionally, the design of GoDDSSy is proposed, which is then later implemented, deployed, and evaluated. Results show that GoDDSSy operates in an agile and resilient way. In addition, we show the limits of the implemented system and how it performs compared to related systems. |
|
Tim Graber, Nachhaltigkeitsvergleich bedeutender Zentralbanken, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Bachelor's Thesis)

|
|
Fabio Pandolfo, Kann mit einer ESG-Trend Strategie eine Outperformance erlangt werden?, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
Die Forschungsfrage dieser Arbeit ist, ob man mit einer ESG-Trend Strategie eine Outperformance
erreichen kann. Diese Arbeit erg¨anzt bisherige Studien mit einer spezifischen Auslegung der ESGTrend
Anlagestrategie f¨ur die Industrie- und Schwellenl¨ander. Mithilfe von Faktormodellen wurde
die Strategien getestet. Die empirische Analyse ergibt, dass in allen Schwellenl¨andern ein Alpha
erzielt wird, w¨ahrend dies in Industriel¨andern lediglich f¨ur das DM Konstant Portfolio gilt. Hingegen ist nur das Alpha des DM Konstant Portfolio statistisch signifikant. Basierend auf den Resultaten
dieser Arbeit kann abschliessend gesagt werden, dass der Erfolg einer Implementation der ESG-Trend
Strategie in einem Portfolio stark von der Region abh¨angig ist. |
|
Adrian Schmidli, Empirische Analyse des Momentum-Effekts in Aktienmärkten, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Bachelor's Thesis)
 
Hanauer und Kaserer (2017) zeigen in ihrer Studie, dass durch die Anwendung einer Momen-tum Strategie nachhaltig eine Überrendite gegenüber des Marktes erzielt werden kann. Die vorliegende Untersuchung bestätigt dies für die drei grossen Regionen der Welt (Asien/Aust-ralien, Europa und Amerika), während sich der asiatische Raum als am geeignetsten heraus-stellt. Zudem setzt sich diese Studie mit einer Vielzahl der benötigten Parameter zur Bildung von Momentum Portfolios auseinander. Optimal wäre es, Winner Portfolios in der Region Ame-rika gemäss einer Periodizität von drei Monaten neu zu bilden, während für Winner Portfolios der anderen zwei Regionen eine Periodizität von 6-12 Monaten ideal wäre. |
|
Sarah Keist, Communicative Risks on Social Media Platforms and Resulting Responsibilities of Their Providers, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Bachelor's Thesis)

|
|
Thierry Bessire, The Mispricing in the Structured Products Market - An Empirical Analysis for Switzerland, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
Based on last best bid prices this thesis examines the fairness of quoted prices in the secondary market
for Swiss structured products. The theoretical fair price is derived from a model price replicating the
payoff structure of the analysed structured product. Most of the analysed product types are observed
to be priced with an implicit premium. Competition among issuers, time to maturity, complexity of
the product and the underlying type are found to impact the degree of mispricing on the secondary
market for structured products in Switzerland. |
|
Jean-Luc Scheiber, Overnight Anomaly in the Swiss Bond Market, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Bachelor's Thesis)
 
This paper uses secondary market quotes of bonds part of the Swiss Bond Index in order to study bond intraday (open to close) and overnight (close to open) returns. The study shows that overnight returns over the last decade were positive in almost all cases and a lot more robust to common market factors than intraday returns. This overnight effect also holds for most sectors, maturities and ratings. |
|
Christian Laely, ESG Certification as a Value Driver for Companies on the Swiss Stock Exchange, University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)
 
This thesis analyzes ESG-certified SPI companies’ stock performance on a portfolio level and stock
level, operating profitability, and WACC from 2016-2021. The companies were analyzed, controlling
for systemic risk factors and other performance factors. The panel-data regressions were conducted
both with OLS and fixed effects to control for unobserved heterogeneity. No significant impact of the
ESG rating on the stock performance was found. Furthermore, the results showed a slightly negative
effect on the operating profitability and no apparent effect on the WACC. However, methodological
concerns and the short sampling period make it difficult to draw a causal link. |
|
Romina Sieni Saliba, Is Corruption part of the cost of doing business. When informal payments are necessary for trade., University of Zurich, Faculty of Business, Economics and Informatics, 2022. (Master's Thesis)

This study uses firm level survey data from 33 countries and 18’887 enterprises, to analyse
the magnitude of, and factors contributing to, corruption, here defined as the practice of firms
giving informal payments or gifts to public officials. The data was obtained from the EBRDEIB-
WB Enterprise Survey (ES) (World Bank, the European Bank for Reconstrcution and
Development (EBRD) and the European Investment Bank (EIB), 2022). To empirically
evaluate the connection between bribery and the observed factors, a series of multiple
regressions were performed using the Ordinary least squared method (OLS).
In this study it was found that corruption is more likely to occur in specific countries, while
the predominant religion of the country has been shown to have no impact upon corruption.
However, the industry within which a firm operates can have an impact upon corruption.
Lastly, the Corruption Perception Index was found to have no correlation with the type of
bribery investigated in this study.
These results could help us understand how much corruption exits, and under which
conditions corruption is more likely to occur. Such findings could help assist in the
mitigation process of corruption. |
|