University of Twente Student Theses


Extracting information from audio data gathered in a virtual reality museum tour to determine personal preferences

Reuvers, E. (2023) Extracting information from audio data gathered in a virtual reality museum tour to determine personal preferences.

[img] PDF
Abstract:This paper explores the possibilities of identifying individual preferences of users through the analysis of audio recordings gathered in a virtual reality museum tour. The audio recordings contain user comments about their observations related to paintings at the museum, as well as inquires for further knowledge. Users were asked to think-out-loud during their visitation. Identification of user preferences would make it possible to design personalized museum tours in the future, for example. The analysis starts with transcribing the audios to texts using the Google Cloud speech-to-text framework. The text files are then analyzed using natural language processing techniques, in particular: part-of-speech tagging, noun phrases extraction and sentiment analysis. The results of the part-of-speech tagging and noun phrase extraction is that most comments contain objective information regarding the objects in the paintings. The sentiment analysis of the complete user comments confirms this. 59.4% of the user comments are neutral and 29.8% is slightly positive, meaning that the comments do not exhibit strong emotional polarity. This suggest that audio recordings based on a thinking-out-loud process may not be sufficient to reliably identify individual preferences. More research is needed to determine if the identification of preferences might be possible with techniques or approaches that have not been utilized in this study.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:02 science and culture in general, 54 computer science
Programme:Computer Science BSc (56964)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page