<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>3265</id>
    <completedYear>2024</completedYear>
    <publishedYear/>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>21</pageFirst>
    <pageLast>22</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <type>conferenceobject</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2025-02-07</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Utilizing Chatbots for Automating Examinations in Higher Education: Perceived Fairness, Trust and Learning Outcomes</title>
    <abstract language="eng">Large Language Models (LLM) are increasingly used to support educational assessment (González-Calatayud et al.,2021), offering a promising approach to addressing challenges associated with scalability, consistency, andpersonalized feedback that manual assessment implies (Fagbohun et al., 2024).Pre-configured interfaces for building custom chatbots based on common AI models provide a low-level approachfor educators to design their own AI-based examination tools that ask learners questions and provide feedback onthe answers. An example of a prototypical application - albeit based on a self-programmed interface - is Nitze's(2024) StudyBuddy, which simulates oral exams. Yet, the extent to which these AI-based exam assistants canautonomously conduct complete exam assessments—creating questions and evaluating and grading responseswithout manual oversight—is still largely unexplored. Despite the clear advantages, issues related to the quality ofthe generated content, the precision of assessments, ethical implications, and acceptance among learners persist assignificant concerns.This paper is a proof-of-concept study combined with a field experiment for an AI-based exam assistant based onthe StudyBuddy suggested by Nitze (2024). The exam assistant provides students with a set of quiz questions andcase studies based on an 80-page reader on topics in organizational behavior, offering personalized feedback andquestion-by-question scoring ranging from 0 to 100%, along with an overall grade and personalized feedback. Toevaluate learning behavior and outcome (in the practice phase), perceived fairness (in the exam phase) and overalllearner acceptance, we run a field experiment with N=35 students enrolled in a business management program at atechnical university in Germany. They take part in a quiz assignment provided and graded by the exam assistant andsubsequently answer a survey with questions based on the Perceived Fairness scale (Sonnleitner &amp; Kovacs, 2020)and the Trust in Automated Systems Test (Wojton et al., 2020). Additionally, the learning outcomes are assessed ina retrospective pre-post design according to Drennan and Hyde (2008). Results will be available by the End of April2024.Insights gained from this study will shed light on the question whether custom chatbots can be effectively used forscalable, efficient assessment processes in higher education. As the ready-made interfaces already available requireminimal prerequisites and technical skills, they present an accessible opportunity for instructors to develop theirown AI-based examination tools, and to ease manual corrections, to guarantee consistent grading and to providestudents with individualized, constructive feedback on their solutions even in large courses where personalizedfeedback would not be possible when manual grading is used.The results of this study will shed light on whether customised chatbots can be effectively used for scalable, efficientassessment processes in higher education. The readily available interfaces, requiring minimal prerequisites andHow do we promote fairness in digital learning futures?22technical skills, enable educators to develop their own AI-based assessment tools. This would open opportunitiesfor educators to build their own AI-based assessment tools and facilitate manual corrections, ensure consistentgrading and provide students with individualised, constructive feedback, even in large courses where personalisedfeedback would not be possible with manual grading.</abstract>
    <parentTitle language="eng">EDEN 2024 Annual Conference, University of Graz, Graz, Austria, 2024</parentTitle>
    <identifier type="url">https://repositorio-aberto.up.pt/bitstream/10216/159440/2/678463.pdf#page=104</identifier>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="PeerReviewNachweis">nein</enrichment>
    <licence>Keine Nutzungslizenz vergeben - es gilt das deutsche Urheberrecht</licence>
    <author>Katharina Frosch</author>
    <author>Friederike Lindauer</author>
    <author>Carmen Winkel</author>
    <collection role="institutes" number="">Fachbereich Wirtschaft</collection>
    <collection role="Hochschulbibliografie" number="1">Hochschulbibliografie</collection>
  </doc>
  <doc>
    <id>3373</id>
    <completedYear>2024</completedYear>
    <publishedYear/>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>59</pageFirst>
    <pageLast>68</pageLast>
    <pageNumber/>
    <edition/>
    <issue>3&amp;4</issue>
    <volume>17</volume>
    <type>article</type>
    <publisherName>IARIA</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2025-09-18</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Lessons learned from building sustainable municipal LoRaWan infrastructure</title>
    <abstract language="eng">Abstract—This paper introduces the LoRaWAN Collaboration Framework (LCF), a strategic blueprint for deploying and managing LoRaWAN infrastructures in smart cities with an emphasis on rural and small municipalities. LoRaWAN technology distinguishes itself by its capability to support long-range, low-power IoT applications, making it ideal for extensive and sparsely populated areas. The LCF aims to address common challenges in these settings, such as limited technical expertise, financial constraints, and the need for cross-municipal cooperation. It outlines roles and responsibilities across various stakeholders including municipal authorities, IT service providers, application developers, and end-users. The framework emphasizes the balance of technological, economic, social and ecological sustainability in line with the United Nations' Sustainable Development Goals. In this paper, we describe the experiences from several LoRaWAN projects in small towns and municipalities in Germany and give some insights to these use cases, the derived collaboration framework, and other arguments to consider before implementing LoRaWAN infrastructures.</abstract>
    <parentTitle language="eng">International Journal On Advances in Networks and Services</parentTitle>
    <enrichment key="PeerReviewNachweis">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <licence>Keine Nutzungslizenz vergeben - es gilt das deutsche Urheberrecht</licence>
    <author>André Nitze</author>
    <author>Tingting Wang</author>
    <author>Josephine Jahn</author>
    <author>Sabah Ali</author>
    <author>Timon Miesner</author>
    <collection role="institutes" number="">Fachbereich Wirtschaft</collection>
    <collection role="Hochschulbibliografie" number="1">Hochschulbibliografie</collection>
    <thesisPublisher>Technische Hochschule Brandenburg</thesisPublisher>
  </doc>
</export-example>
