..
/ cps / home /
 

Julia Nießner

Julia Nießner
Email: julia.niessner(at)uni-siegen.de 

 

Raum: US-D 117 (Ludwig-Wittgenstein-Haus, Campus Unteres Schloss)

Telefon:  +49 (0) 271 / 740 5247

Sprechstunde: Nach Vereinbarung

Vita

Julia Nießner ist wissenschaftliche Mitarbeiterin am Lehrstuhl für Cyber-Physische Systeme. Sie studierte an der niederländischen Fontys University of applied Science den Bachelor-Studiengang „Business Informatics“ (B.Sc.). Im Zuge dessen absolvierte sie ein Erasmus-Auslandssemester an der „INSEEC“ in Chambéry (Frankreich), mit Schwerpunkt im Bereich „International Trade“. Parallel zu ihrem Studium sammelte Julia Erfahrungen in der Praxis unter anderem bei einem führenden Automobilhersteller und einem internationalen IT-Dienstleister. In Kooperation mit diesem schrieb sie zudem ihre Bachelorarbeit zum Thema „Digital Twins – Powerprediction for Wind Turbines“.

Im Anschluss an das erfolgreiche Bachelorstudium begann Julia mit ihrem Masterstudium der Wirtschaftsinformatik (M.Sc.) an der Universität Siegen. Ihr Masterstudium schloss sie im Februar 2021 erfolgreich ab mit ihrer Masterarbeit zum Thema „Entwurf, Entwicklung und Evaluation eines Empfehlungssystems für Kochrezepte“.

Seit März 2021 ist Julia als wissenschaftliche Mitarbeiterin tätig. Ihr aktueller Aufgabenschwerpunkt liegt in der Unterstützung eines Projekts zur Erschließung eines regionalen Zukunftszentrums für künstliche Intelligenz.

Publikationen

2021


  • Nießner, J. & Ludwig, T. (2021)Design of a Knowledge-Based Recommender System for Recipes From an End-User Perspective

    Mensch und Computer 2021. New York, NY, USA, Publisher: Association for Computing Machinery, Pages: 512–519 doi:10.1145/3473856.3473888
    [BibTeX] [Abstract] [Download PDF]
    Nowadays, recommender systems are a fundamental part of several online services. However, most of these systems rely on collective user data and ratings or a preselection of parameters to derive appropriate recommendations. Within this paper, we examine recommendations without previous user data. We therefore designed and evaluated a knowledge-based recommender system by turning to recipe recommendations that offer alternatives for favorite recipes. We introduce and compare three versions of a given algorithm. Our evaluation shows that the knowledge-based approach may serve as a good start for deriving appropriate recommendations without prior user data. Moreover, we show that end-users’ assumptions about decisive criteria of a recommender system do not necessarily match the later actual decisive criteria.
    @inproceedings{niesner_design_2021,
    address = {New York, NY, USA},
    series = {{MuC} '21},
    title = {Design of a {Knowledge}-{Based} {Recommender} {System} for {Recipes} {From} an {End}-{User} {Perspective}},
    isbn = {978-1-4503-8645-6},
    url = {https://doi.org/10.1145/3473856.3473888},
    doi = {10.1145/3473856.3473888},
    abstract = {Nowadays, recommender systems are a fundamental part of several online services. However, most of these systems rely on collective user data and ratings or a preselection of parameters to derive appropriate recommendations. Within this paper, we examine recommendations without previous user data. We therefore designed and evaluated a knowledge-based recommender system by turning to recipe recommendations that offer alternatives for favorite recipes. We introduce and compare three versions of a given algorithm. Our evaluation shows that the knowledge-based approach may serve as a good start for deriving appropriate recommendations without prior user data. Moreover, we show that end-users’ assumptions about decisive criteria of a recommender system do not necessarily match the later actual decisive criteria.},
    urldate = {2021-09-14},
    booktitle = {Mensch und {Computer} 2021},
    publisher = {Association for Computing Machinery},
    author = {Nießner, Julia and Ludwig, Thomas},
    month = sep,
    year = {2021},
    keywords = {Knowledge-based Filtering, Recipes, Recommender System, Similarity Metrics, User Study},
    pages = {512--519},
    file = {Full Text PDF:C\:\\Users\\Nathanael Klein\\Zotero\\storage\\L28HGLP3\\Nießner und Ludwig - 2021 - Design of a Knowledge-Based Recommender System for.pdf:application/pdf},
    }

  • Weber, P., Krings, K., Nießner, J., Brodesser, S. & Ludwig, T. (2021)FoodChattAR: Exploring the Design Space of Edible Virtual Agents for Human-Food Interaction

    Designing Interactive Systems Conference 2021. New York, NY, USA, Publisher: Association for Computing Machinery, Pages: 638–650 doi:10.1145/3461778.3461998
    [BibTeX] [Abstract] [Download PDF]
    There has been recent criticism from researchers towards simple replication of traditional role models in the design of virtual agents and robots, and a call for new forms of interaction and communication with technology. By exploring the field of Human-Food interaction (HFI) – a sub-area of Human-Computer Interaction (HCI) which aims to investigate the diversity of ways people interact with food – we therefore specifically examine the design space of edible anthropomorphic virtual agents (EAVAs). To understand human-to-food interactive communication, we conducted an interview study with 19 participants, followed by a co-design workshop on the design of conversational agents for personified food. Based on the results, we implemented a prototype called FoodChattAR that employs augmented reality and chatbots to interact and communicate with food. Our evaluation with 21 participants shows that FoodChattAR turns eating into fun, while at the same time the food conveys relevant societal facts about itself. We contribute to the field of HCI by introducing EAVAs as a novel human-to-food interaction.
    @inproceedings{weber_foodchattar_2021,
    address = {New York, NY, USA},
    series = {{DIS} '21},
    title = {{FoodChattAR}: {Exploring} the {Design} {Space} of {Edible} {Virtual} {Agents} for {Human}-{Food} {Interaction}},
    isbn = {978-1-4503-8476-6},
    shorttitle = {{FoodChattAR}},
    url = {https://doi.org/10.1145/3461778.3461998},
    doi = {10.1145/3461778.3461998},
    abstract = {There has been recent criticism from researchers towards simple replication of traditional role models in the design of virtual agents and robots, and a call for new forms of interaction and communication with technology. By exploring the field of Human-Food interaction (HFI) – a sub-area of Human-Computer Interaction (HCI) which aims to investigate the diversity of ways people interact with food – we therefore specifically examine the design space of edible anthropomorphic virtual agents (EAVAs). To understand human-to-food interactive communication, we conducted an interview study with 19 participants, followed by a co-design workshop on the design of conversational agents for personified food. Based on the results, we implemented a prototype called FoodChattAR that employs augmented reality and chatbots to interact and communicate with food. Our evaluation with 21 participants shows that FoodChattAR turns eating into fun, while at the same time the food conveys relevant societal facts about itself. We contribute to the field of HCI by introducing EAVAs as a novel human-to-food interaction.},
    urldate = {2021-07-05},
    booktitle = {Designing {Interactive} {Systems} {Conference} 2021},
    publisher = {Association for Computing Machinery},
    author = {Weber, Philip and Krings, Kevin and Nießner, Julia and Brodesser, Sabrina and Ludwig, Thomas},
    month = jun,
    year = {2021},
    keywords = {Human-Food Interaction, rendezfood, Anthropomorphism, Augmented Food, Conversational Agents, Edible Anthropomorphic Virtual Agents, Virtual Agents},
    pages = {638--650},
    file = {Full Text PDF:C\:\\Users\\Nathanael Klein\\Zotero\\storage\\JCW56FUW\\Weber et al. - 2021 - FoodChattAR Exploring the Design Space of Edible .pdf:application/pdf},
    }