Designing an AI-powered multilingual chatbot for enhancing library user engagement

Thesis Abstract

The rapid proliferation of digital technologies has transformed the landscape of library services, necessitating innovative approaches to enhance user engagement through accessible and interactive platforms. Despite the growing availability of digital interfaces, many libraries face challenges in effectively communicating with diverse user populations, particularly those with multilingual needs and varying levels of digital literacy. This study aims to design, develop, and evaluate an AI-powered multilingual chatbot to improve user interaction, information retrieval, and overall satisfaction within a university library context. The specific objectives include (1) identifying user requirements and language preferences; (2) developing a chatbot with multilingual capabilities leveraging natural language processing (NLP) and machine learning algorithms; (3) assessing the usability and effectiveness of the chatbot in facilitating access to library resources; and (4) analyzing user engagement metrics and satisfaction levels post-implementation. The research adopts a mixed-methods sequential exploratory design, integrating qualitative data obtained through user interviews and focus groups with quantitative data derived from usage logs and structured surveys. The target population comprises undergraduate and postgraduate students, faculty members, and library staff at a university with a student body of approximately 25,000. A stratified random sampling technique is employed to select a sample size of 300 users for the survey, ensuring representation across different user categories. Data collection instruments include semi-structured interview protocols, digital usage analytics tools, and standardized questionnaires measuring user satisfaction and engagement. The development phase of the chatbot incorporates iterative design based on user feedback, utilizing NLP frameworks such as Google Dialogflow and translation APIs to support at least five major languages English, Mandarin, Arabic, Spanish, and French. Data analysis encompasses descriptive statistics, regression analysis to determine factors influencing user engagement, and thematic analysis of qualitative feedback, guided by the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT) as theoretical frameworks. Expected findings suggest that a multilingual AI chatbot significantly enhances users' ability to access information efficiently, increases overall satisfaction, and fosters greater engagement with library resources. The analysis is anticipated to reveal that ease of use, perceived usefulness, and language convenience are key determinants of user acceptance, corroborating TAM and UTAUT predictions. Additionally, usage analytics are expected to demonstrate increased interaction frequency and reduced inquiry response times compared to traditional reference services. This research makes a substantial contribution to library and information science by providing empirical evidence on the efficacy of AI-enabled multilingual chatbots in academic library settings. It extends existing knowledge on digital engagement strategies, highlighting the importance of linguistic inclusivity and intelligent conversational agents in expanding access and improving user experience. Practically, the study offers a replicable framework for library practitioners to deploy similar AI solutions tailored to their specific multilingual contexts. The main conclusion asserts that AI-powered multilingual chatbots are effective tools for augmenting library user engagement, particularly in linguistically diverse communities. Recommendations emphasize the integration of such platforms into broader digital literacy initiatives and ongoing user feedback mechanisms to ensure continuous improvement. The study advocates for further research on scalability, long-term impact, and the incorporation of advanced conversational AI features such as sentiment analysis and personalized content delivery to deepen user engagement and operational efficiency in diverse library environments.