Search
You are now searching on the website "Department of Computer and Systems Sciences". Would you like to search the entire Stockholm University website?
Search on su.seShowing 101-120 of 234 hits
- Crisis of trust in schools when AI tools emerged
Department of Computer and Systems Sciences
When generative AI swept across the world, many students felt they had been left to fend for themselves. They received no guidance on how the tools could be used, and the trust between teachers and students took a serious knock.
- Thesis defence: Ali Beikmohammadi
Department of Computer and Systems Sciences
Welcome to a thesis defence at DSV! Ali Beikmohammadi explores autonomous decision-making. The PhD thesis contributes to the research subject AI and Data Science.
- AI has the potential to create greener IoT
Department of Computer and Systems Sciences
The growing number of devices in manufacturing, health care and agriculture are helpful in many ways. But they also consume huge amounts of energy. Praveen Kumar Donta tries to solve this energy problem.
- Predoc seminar: Razan Jaberibraheem
Department of Computer and Systems Sciences
Welcome to a predoc seminar on communication between humans and robots! Razan Jaberibraheem, PhD student at DSV, is the respondent. Seminar Date: Wednesday 5 April 2023 Time: 13.00 – 16.00 Location: Room M10, DSV, Nod building, Borgarfjordsgatan 12, Kista Invitation to PhD student Razan Jaberibraheem’s predoc seminar at the Department for Computer and Systems Sciences (DSV), Stockholm University. She will present her ongoing work on “Towards Designing Better Speech Agent Interaction: Learning from Human-Human Interaction”. Respondent: Razan Jaberibraheem, DSV Opponent: Morten Fjeld, Chalmers Main supervisor: Barry Brown, DSV Supervisor: Donald McMillan, DSV Professor closest to the subject: Teresa Cerratto-Pargman, DSV Contact and more information
- Thesis defence: Mohamed Sapraz
Department of Computer and Systems Sciences
Welcome to Mohamed Sapraz's thesis defence! His dissertation addresses collaboration between government and citizens that can lead to increased environmental sustainability. Sri Lanka is the focus of his studies. Thesis defence Date: Thursday 30 March 2023 Time: 09.00 – 13.00 Location: Room L30, DSV, Borgarfjordsgatan 12, Kista
- Predoc seminar: Georgios Koutsopoulos
Department of Computer and Systems Sciences
Welcome to Georgios Koutsopoulos predoc seminar at the Department of Computer and Systems Sciences! Seminar Date: Wednesday 29 March 2023 Time: 13.00 – 16.00 Location: Rum M20, DSV, Borgarfjordsgatan 12, Kista PhD student Georgios Koutsopoulos presents his ongoing work on “KYKLOS – a modeling method and tool for managing changing capabilities in organizations”. The predoc seminar takes place at the Department for Computer and Systems Sciences (DSV), Stockholm University. Respondent: Georgios Koutsopoulos, DSV Opponent: Mārīte Kirikova, Riga Technical University, Latvia Main supervisor: Janis Stirna, DSV Supervisor: Martin Henkel, DSV Professor closest to the subject: Paul Johannesson, DSV Abstract The environments within which modern organizations are operating, are characterized by a constantly increasing volatility and diversity, and act as a driving force that continuously presents opportunities and threats for the organizations. The result of this situation is that the environments have become so dynamic that the phenomenon of organizational change is slowly becoming the new constant instead of being an exception, as in earlier years. The challenge derived from the need to handle constant change and evolution naturally leads to changing organizational capabilities. Information Systems, as part of an organization, can provide efficient support for the design and analysis of capabilities, and Enterprise Modeling can facilitate the development of specialized methods for this task. Hitherto, a plethora of capability modeling methods exist, and provide a wide spectrum of perspectives. However, the support provided for managing the phenomenon of changing organizational capabilities remains limited, since the existing methods have omitted to capture the necessary elements for depicting organizational capabilities in motion and the motivation triggering such transitions. Additionally, managing change requires more structure approaches, which should be methodologically supported and conceptualized. The goal of this PhD thesis is to develop a capability modeling method, named KYKLOS, which will support organizational change by modeling changing capabilities of organizations, and all the relevant aspects of the phenomenon. Following the Design Science Research framework, it was initiated with explicating the problem via a literature review on the existing capability meta-model literature, followed by the elicitation of method requirements, using both the existing literature and a case study conducted in the domain of public healthcare in Sweden. Using the requirements, an initial meta-model was developed, demonstrated using the same case study, and evaluated by experts using interviews. This activity resulted in extending the initial version of the meta-model, which was then instantiated in a case study conducted in the public arts and culture domain of Greece, while in parallel the compatibility of other modeling approaches was explored. Afterwards, the updated version of the meta-model was converted to language meta-model, suitable for the implementation of the method in a tool, using the ADOxx meta-modeling platform. Simultaneously with the tool development, the syntax, semantics and procedure of the modeling method were completed. The complete method was demonstrated in a case study at a company providing ERP solutions and consulting in Sweden, the result of which was employed during the evaluation of the method by two groups consisting of the method’s stakeholder roles. The thesis contributes towards an improved management of change in organizations, by the development of a method and supporting tool with the capability perspective as the focal point to design and analyze organizational change. Contact Georgios Koutsopoulos
- Predoc seminar: Kazi Masum Sadique
Department of Computer and Systems Sciences
Welcome to a predoc seminar on IoT security! Kazi Masum Sadique, PhD student at DSV, is the respondent. Seminar Date: Friday 1 September 2023 Time: 13.00 – 15.00 Location: Room M20, DSV, Nod building, Borgarfjordsgatan 12, Kista On September 1, 2023, PhD student Kazi Masum Sadique will present his ongoing work on “Securing IoT using Decentralized Trust, Privacy and Identity Management – An artefact for security, trust, privacy and identity management of IoT”. The seminar takes place at the Department of Computer and Systems Sciences (DSV), Stockholm University. Respondent: Kazi Masum Sadique, DSV Opponent: Ding Jianguo, Blekinge University of Technology, Karlskrona Main supervisor: Rahim Rahmani, DSV Supervisor: Paul Johannesson, DSV Professor closest to the subject: Oliver Popov, DSV Contact Kazi Masum Sadique Abstract The Internet of Things (IoT) is a multidisciplinary area where technology meets people to enrich the quality of living with an improved working environment and efficient productivity. As the number of IoT devices increases, many new technology areas are being integrated with IoT. IoT devices mainly connect and collaborate with the central cloud servers for data management. The IoT paradigm is built upon the Internet and accesses different layers of Internet architectures. IoT devices are at the access layer of the Internet, and cloud servers are located at the top layer. The innovative use cases of IoT applications drive the requirement for quick decision-making as near to the source of information. IoT devices need to be authenticated near the source for rapid request processing. Trustworthy interaction and secure communication between different entities of an IoT paradigm are crucial. A centralized cloud-based implementation of IoT solutions can be problematic for ensuring trustworthy and authenticated interactions where quicker decision-making is involved. Also, privacy leakage possibilities increase with cloud-based solutions as it involves multiparty interactions, which introduce more complexity to ensure data privacy. Traditional security models are unsuitable for IoT due to IoT application and service heterogeneity. There is no generic model for IoT data security and user data privacy, with trustworthy collaboration and identity management near the source. This thesis focused on creating a generic state-of-the-art artefact for IoT security by using a decentralized trust, user data privacy, and localized identity management for heterogeneous IoT devices and services. The main contributions of this thesis include: a novel decentralized model for secure and reliable interaction between components of the IoT paradigm, which is complemented by a novel decentralized trust management model, a novel edge gateway-based privacy enhancement scheme, and a new decentralized identity management model with new authentication and authorization mechanism for IoT devices where access to new resources is granted locally, and activities are recorded with context information. The proposed models are generic and can be easily adapted to real-life IoT use cases with minor amendments.
- Thesis defence: Lida Huang
Department of Computer and Systems Sciences
Welcome to a thesis defence at DSV! Lida Huang presents her thesis on how motor-disabled people can express their ideas visually through free-eye drawing. Thesis defence Date: Thursday 14 November 2024 Time: 13.00 – 16.00 Location: L30, DSV, Borgarfjordsgatan 12, Kista On November 14, 2024, Lida Huang will present her PhD thesis at the Department of Computer and Systems Sciences (DSV), Stockholm University. The title of the thesis is “Accessible Artistic Visual Expression Via Eye Tracking”. PhD student: Lida Huang, DSV Opponent: Lina Eklund, Uppsala University Main supervisor: Mirjam Palosaari Eladhari, DSV Supervisors: Sindri Magnusson and Thomas Westin, DSV Contact Lida Huang Download the thesis from Diva The defence takes place at DSV in Kista, starting at 13:00 pm. Find your way to DSV Abstract This thesis explores the use of eye tracking to enable motor-disabled people to express their ideas visually through free-eye drawing. While recent research has yielded state-of-the-art results in unimodal gaze control for free-eye drawing, it was found that there are still limitations such as limited functionality, conflicts between observation and drawing, and brush tailing issues. To address these challenges, this thesis uncovers the theories behind these challenges and proposes novel solutions, including a free-eye drawing application with a smooth-pursuit based gaze control method, gaze-based UI design, a double-blink gaze gesture, and advanced adoption with artificial intelligence. Surveys, experiments, and field studies were conducted to assess the system's performance and usability compared to mainstream gaze-control and hand drawing methods. The results show that the proposed solutions provide efficient interaction and the ability to create graphics at the level of a hand sketch for motor-disabled people. Overall, this thesis contributes first-hand user experience reports from eye tracker users and a robust and professional accessible solution for artistic visual expression, and highlights valuable design considerations for future developments in gaze interaction. Keywords: Eye tracking, gaze control, drawing, accessibility
- Predoc seminar: Beatrice Åkerblom
Department of Computer and Systems Sciences
Welcome to a predoc seminar on programming languages. Beatrice Åkerblom, PhD student at DSV, is the respondent. Seminar Date: Tuesday 18 June 2024 Time: 15.00 – 17.00 Location: Location: M20, DSV, Borgarfjordsgatan 12, Kista On June 18, 2024, PhD student Beatrice Åkerblom will present her ongoing work on “On Typability in Programming Languages”. The seminar takes place at the Department of Computer and Systems Sciences (DSV), Stockholm University.
- Predoc seminar: Kasper Karlgren
Department of Computer and Systems Sciences
Welcome to a predoc seminar on sleep – how new technology keeps track of our sleeping patterns, and how people “hack” their sleep. Kasper Karlgren, PhD student at DSV, is the respondent. Seminar Date: Friday 3 May 2024 Time: 10.00 – 13.00 Location: M10, DSV, Borgarfjordsgatan 12, Kista On May 3, 2024, PhD student Kasper Karlgren will present his ongoing work on “Tracking and hacking sleep: Designing for non-normative sleepers”. The seminar takes place at the Department of Computer and Systems Sciences (DSV), Stockholm University. Respondent: Kasper Karlgren, DSV Opponent: Anna Sigridur Islind, Reykjavik University, Iceland Main supervisor: Barry Brown, DSV Supervisor: Donald McMillan, DSV Professor closest to the subject: Jakob Tholander, DSV Contact information for Kasper Karlgren Find your way to DSV Abstract With sleep-tracking technology being built-in into many available smart systems and wearables, the notion of a user of these systems gets larger. The development still focuses on traditional 8 hours-per-night schedules – overlooking the varied nature of people’s sleep – and focuses on metrics that are difficult to influence such as sleep phases, making the systems unactionable for a large portion of the users. With this work, I investigate what sleep tracking can do for users by studying what actions and challenges already are employed by people. Using a combination of qualitative methods and design approaches I study how people ‘hack’ their sleep, what challenges emerge from sleeping differently, and how sleep technology is used in everyday life. The findings and contributions of this work include (1) rich descriptions of how users share and discuss sleep hacks in online communities (2) insights and reflections on the social factors of sleeping outside norm hours (3) how sleep, and near-sleep technologies, such as sleep tracking, can be designed to be inclusive of those with non-normative sleep patterns. Concluding the contributions, I propose a tentative strong concept called design of the self – a way to design technology that centres on the actions and varying goals the user and their bodies. In discussion, this work describes related to the notions of bodies and users in HCI (human–computer interaction), how changes in modern work arrangements call for new technology to support the arrangement of sleep and life rhythms, and reflections on norms and soft-paternalism of sleep tracking systems.
- Predoc seminar: Alejandro Kuratomi
Department of Computer and Systems Sciences
Welcome to a predoc seminar on opening the “black box” of AI algorithms! Alejandro Kuratomi, PhD student at DSV, is the respondent. Seminar Date: Wednesday 27 March 2024 Time: 09.00 – 12.00 Location: Room M20, DSV, Borgarfjordsgatan 12, Kista On March 27, 2024, PhD student Alejandro Kuratomi will present his ongoing work on “Orange juice – Enhancing Machine Learning Interpretability”. The seminar takes place at the Department of Computer and Systems Sciences (DSV), Stockholm University. Respondent: Alejandro Kuratomi, DSV Opponent: Professor Toon Calders, University of Antwerp, Belgium Main supervisor: Tony Lindgren, DSV Supervisor: Panagiotis Papapetrou, DSV Professor closest to the subject: Rahim Rahmani, DSV Contact Alejandro Kuratomi Abstract In the current state of AI’s development, it is reasonable to think that AI will continue to expand and be increasingly utilized across different fields, highly impacting every aspect of humanity’s welfare and livelihood. However, not all future scenarios described by AI researchers and institutions are good, i.e., AI has both the potential to be extremely beneficial, and pose existential threats to humanity. Even if such scenarios develop in the longer term, it is necessary to start focusing now on tools that allow humans to open the black-box of the AI algorithms, to increase their understandability and trustworthiness, and in that way try to avoid such conceivably harmful future scenarios. Additionally, the interpretability of the AI algorithms is even now a challenge to the AI development, as it is an obstacle equivalent to those previously seen which triggered past AI winters, such as hardware constraints or the over-expectation of results, in other words, research in interpretability and model understanding will help avoid a third AI winter, which, if triggered, could be devastating for the world economy. This thesis aids in researching the field of AI interpretability, focusing mainly on the algorithms related to local explainability, which are the algorithms that attempt to extract useful explanations for the output of machine learning models on individual instances, specifically targeting areas of improvement in the counterfactual and local surrogate explanations. Since, for some of the datasets, these instances are humans, opening the black-box of the machine learning models may also reveal the existence of biases or fairness issues among different groups of people. This thesis then focuses also on the topic of machine learning fairness assessment through the use of local explainability methods, proposing two novel elements: one single accuracy-based and counterfactual-based bias detection measure and a counterfactual generation method for groups intended for bias detection and actionability-oriented fairness. Finally, the idea behind interpretability is to be able to eventually implement such methods into real-world applications. This thesis presents an application of the conformal prediction framework to a regression problem related to autonomous vehicle localization systems that both generates the predicted positioning error value of a vehicle and its confidence interval with some level of significance.
- Digital Futures for Education: Afterlife of Data
Department of Computer and Systems Sciences
What happens to our online identities when we die – and why should we care? This is the topic for the next conversation in the Digital Futures for Education series. Seminar Date: Wednesday 18 September 2024 Time: 13.00 – 15.00 Location: Digital Futures hub, floor 2, Osquars Backe 5, KTH main campus
- Predoc seminar: Chen Hsi Tsai
Department of Computer and Systems Sciences
Welcome to a predoc seminar on digital business ecosystems. Chen Hsi Tsai, PhD student at DSV, is the respondent. Seminar Date: Tuesday 18 June 2024 Time: 09.00 – 11.00 Location: Room M6:2, DSV, Borgarfjordsgatan 12, Kista, and Zoom On June 18, 2024, PhD student Chen Hsi Tsai will present his ongoing work on “DBEmap – A Modularised Method for Designing Resilient Digital Business Ecosystems”. The seminar takes place at the Department of Computer and Systems Sciences (DSV), Stockholm University. Respondent: Chen Hsi Tsai, DSV Opponent: Estefania Serral Asensio, KU Leuven, Belgium Main supervisor: Janis Stirna, DSV Supervisor: Jelena Zdravkovic, DSV Professor closest to the subject: Ilia Bider, DSV Contact information for Chen Hsi Tsai Chen Hsi Tsai presented his licentiate thesis on February 8, 2023 Find your way to DSV Send an e-mail to Chen Hsi Tsai to get the Zoom link.
- ECSO-CMS 2024
Department of Computer and Systems Sciences
The Department of Computer and Systems Sciences (DSV) will be the host of the international European Conference on Stochastic Optimization and Computational Management Science (ECSO-CMS) in 2024. Conference Start date: Thursday 4 July 2024 Time: 08.00 End date: Friday 5 July 2024 Time: 20.00 Location: Aula Nod, Borgarfjordsgatan 12, Kista Campus Decision making under uncertainty and competition is the main focus of ECSO-CMS which will be organised July 4–5, 2024. The conference brings together researchers and practitioners, interested in stochastic programming, robust optimisation and game theory. Professor Afzal Siddiqui, DSV, is the contact person More information on the conference website
- Thesis defence: Piyumi Udeshinee
Department of Computer and Systems Sciences
Welcome to a thesis defence at DSV! Piyumi Udeshinee presents her thesis on how text chats can be used in language learning. Thesis defence Date: Wednesday 4 December 2024 Time: 13.00 – 16.00 Location: Aula Nod, DSV, Borgarfjordsgatan 12, Kista
- Thesis defence: Yongchao Wu
Department of Computer and Systems Sciences
Welcome to a thesis defence at DSV! In his PhD thesis, Yongchao Wu explores how pretrained language models and AI can be used in education. Thesis defence Date: Monday 16 December 2024 Time: 09.00 – 12.00 Location: L30, DSV, Borgarfjordsgatan 12, Kista On December 16, 2024, Yongchao Wu will present his PhD thesis at the Department of Computer and Systems Sciences (DSV), Stockholm University. The title of the thesis is “Exploring the Educational Utility of Pretrained Language Models”. PhD student: Yongchao Wu, DSV Opponent: Filip Ginter, University of Turku, Finland Main supervisor: Aron Henriksson, DSV Supervisors: Jalal Nouri and Martin Duneld, DSV Download the PhD thesis from Diva Contact Yongchao Wu The defence takes place at DSV in Kista, starting at 09:00. Find your way to DSV Abstract The emergence of pretrained language models has profoundly reshaped natural language processing, serving as foundation models for a wide range of tasks. Over the past decade, pretrained language models have evolved significantly, leading to the development of different types of models and approaches for utilising them. This progression spans from static to contextual models and from smaller models to more powerful, generative large language models. The increasing capabilities of these models have, in turn, led to growing interest in exploring new use cases and applications across various domains, including education, where digitalisation has created opportunities for AI applications that leverage pretrained language models, particularly due to the abundance of text data in educational contexts. This thesis explores the educational utility of pretrained language models, specifically by investigating how different paradigms of these models can be applied to address tasks in education. These paradigms include various methodologies for leveraging the knowledge embedded in pretrained language models, such as embeddings, fine-tuning, prompt-based learning, and in-context learning. For collaborative learning group formation, a clustering approach based on pretrained embeddings is proposed, enabling the creation of either homogeneous or heterogeneous groups depending on the specific learning situation. For automated essay scoring, a pretrained language model is fine-tuned using both the essay instructions and the essay text as input; the proposed method also highlights key topical sentences that contribute to the predicted essay score. For educational question generation, a method based on prompt-based learning is introduced and shown to be more data-efficient than existing methods. Finally, for educational question answering, certain limitations of the in-context learning (or prompting) paradigm, such as a tendency of large language models to hallucinate or miscalculate, are addressed. Specifically, workflows and prompting strategies based on retrieval-augmented generation and tool-augmented generation are proposed, allowing large language models to ground answers in specific learning materials and to leverage external tools, such as calculators and knowledge bases, within chain-of-thought reasoning processes. These strategies are shown to produce more reliable and transparent answers to complex questions. Through five empirical studies, methodological innovations within each paradigm of pretrained language models are proposed and evaluated for specific educational use cases. In addition to contributing methodologically to natural language processing, the results demonstrate the potential utility of pretrained language models in educational AI applications, thereby advancing the field of technology enhanced learning. The proposed methods not only improve predictive performance on specific tasks but also aim to enhance the transparency of pretrained language models, which is essential for building reliable and trustworthy educational AI applications. Keywords Natural Language Processing, Technology Enhanced Learning, Pretrained Language Models, Large Language Models, Generative AI, Collaborative Learning, Automated Essay Scoring, Educational Question Generation, Educational Question Answering
- ExtraliTalks: Creating a Digital Twin of the Universe: A New Frontier in Tracing Cosmic Origins
Department of Computer and Systems Sciences
Welcome to an open seminar series at Extrality Lab, Department of Computer and Systems Sciences (DSV). We invite representatives from both academia and industry to join and share their knowledge on current topics. Seminar Date: Wednesday 11 December 2024 Time: 16.00 – 17.00 Location: Extrality Lab, DSV, Campus Kista
- Hybrid seminar: AI in schools is a mess… is this a problem?
Department of Computer and Systems Sciences
Welcome to an important discussion in the Digital Futures for Education seminar series: How does AI transform education, and what are the long-term effects? Seminar Date: Wednesday 19 March 2025 Time: 11.30 – 13.00 Location: Digital Futures, Osquars Backe 5, KTH or Zoom Does Sweden need a coordinated national plan around AI and education? Do teachers need to have AI competence? And what aspects are we missing when we’re introducing AI tools in education? Questions like these will be discussed by Neil Selwyn , professor at Monash University in Australia, and Annika Agélii Genlott , Swedish Association for Local Authorities and Regions. More information and registration The host of this Digital Futures seminar is professor Teresa Cerratto Pargman , Department of Computer and Systems Sciences (DSV). Digital Futures is a cooperation between Stockholm University, KTH and Rise.
- Hybrid seminar: Dark Machines
Department of Computer and Systems Sciences
Welcome to a book conversation with Victor Galaz in the Digital Futures for Education seminar series. Seminar Date: Wednesday 12 March 2025 Time: 15.00 – 16.30 Location: Digital Futures, Osquars Backe 5, KTH or Zoom The host of this Digital Futures seminar is professor Teresa Cerratto-Pargman , Department of Computer and Systems Sciences (DSV). The speaker is Dr. Victor Galaz , Stockholm Resilience Centre at Stockholm University. He will present his new book “Dark Machines: How Artificial Intelligence, Digitalization and Automation is Changing our Living Planet”. More information and registration Digital Futures is a cooperation between Stockholm University, KTH and Rise.
- Open access to research data and automatic pseudonymization: Two years with the Mormor Karl project
Department of Computer and Systems Sciences
Welcome to a research seminar at the Department of Computer and Systems Sciences (DSV). We are happy to have professor Elena Volodina as our guest, presenting her work on automatic pseudonymization. Seminar Date: Tuesday 3 June 2025 Time: 14.00 – 15.00 Location: Lilla Hörsalen, DSV, Borgarfjordsgatan 12, Kista Elena Volodina is a professor at the Department of Swedish, multilingualism, and language technology, University of Gothenburg. She says the following about her talk at DSV on June 3: “ This talk will be devoted to the challenges of working with data that contains personal information. I will describe a set of experiments with automatic pseudonymization that we have performed within the Mormor Karl project. Among others, experiments with detection and labeling of personal categories using BERT models (Szawerna et al. 2024, 2025), attempts at using LLMs to “fill in the blanks” when substituting personal information with pseudonyms (yet unpublished) and a study on whether pseudonyms can provoke biased automated classifications (Muñoz Sánchez et al. 2024). The choice of models for our experiments is currently dictated by the sensitive nature of our data. To extend the choice from open source to proprietary models, we are currently collecting a “pseudo-corpus” with fictitious personal information that we will be able to share freely for future research. You are welcome to contribute to the pseudo-corpus collection as well Finally, in this talk I will name several strategies to unify the research on automatic pseudonymization, and outline further challenges, needs for standardization and a proposal of a shared task. ” References Maria Irena Szawerna, Simon Dobnik, Ricardo Muñoz Sánchez, and Elena Volodina, 2025: The Devil’s in the Details: the Detailedness of Classes Influences Personal Information Detection and Labeling Maria Irena Szawerna, Simon Dobnik, Ricardo Muñoz Sánchez, Therese Lindström Tiedemann and Elena Volodina, 2024: Detecting Personal Identifiable Information in Swedish Learner Essays Ricardo Muñoz Sánchez, Simon Dobnik, Maria Irena Szawerna, Therese Lindström Tiedemann and Elena Volodina, 2024: Did the Names I Used within My Essay Affect My Score? Diagnosing Name Biases in Automated Essay Scoring More information about Elena Volodina More information about the Mormor Karl project This seminar is organised at DSV, Campus Kista. If you are a researcher at another Stockholm University department or centre and wish to participate, please contact Hercules Dalianis Find your way to DSV