A Chatbot to Support International Students’ Entry to Worklife : Prototype Development and a User Study
Year: 2023
Type: dissertation
Abstract: Chatbots have been employed in various fields, incorporating artificial intelligence (AI) technology to enhance realism.Despite extensive research on recruitment chatbots catering to recruiters, this study investigates the utilization patterns of job seekers interacting with the Generative Pretrained Transformer (GPT)-based chatbot.This study addresses the employment challenges of international degree students in Tampere, Finland, aiming to facilitate their adaptation to the worklife through a GPT-based chatbot prototype and user study.Preceding prototype development, conversation flows are designed, and a Tampere job search Knowledge Base (KB) is constructed via cross-disciplinary research and web content analysis.The Voiceflow platform is then employed for prototype creation.I conducted 3 evaluation workshops with 7 international degree students seeking employment in Tampere.The workshops utilize the chatbot prototype, the User Experience Questionnaire (UEQ) and group interviews to explore challenges associated with job applications, users' expectations for the chatbot, and interaction behaviours.Subsequently, data is systematically organised into formative themes employing Grounded Theory and affinity diagrams.The qualitative results highlight job seekers' challenges, expectations, perceptions of the chatbot, and interaction behaviour.Typically, they contend with real-world challenges and external barriers in the job market.Accordingly, they anticipate the chatbot's role in enhancing job search skills and understanding the local employment landscape.They favourably perceive the chatbot and prefer free-form questions to predefined conversation flows, though they face challenges in formulating coherent questions.Additionally, these results bear implications for prospective research on GPT-based chatbots within the context of job seekers, leading to concentrating on improving conversation flows, and user experience, and evaluating chatbots based on local job market criteria and professional skills.In conclusion, the study indicates that the GPT-based chatbot was effective, encouraging, significant, and valuable in supporting international students with their job-seeking process.Further pertinent research in this domain is deemed valuable.
Author: Liu, Anfang
Institution: Tampere University
Language: English
Topic: AI in Service Interactions
Subfield: Marketing
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7132930189
