Ionian University
Τέχνης έρευνα στο διαδίκτυο: ανάλυση της συμπεριφοράς χρηστών και των τεχνολογιών αναζήτησης
Abstract
dc:descriptionThe main objective of this dissertation is the comprehensive study of user behavior when searching online for content related to art and culture. This task was accomplished through the collection of a variety of qualitative and quantitative data from members of the wider community and the deduction of conclusions about the habits and needs of these users, as well as the manner in which they engage with modern searching technologies. In order to collect such data, emphasis was given to the design and development of a federated search engine, whose main functionality involved the aggregation of content from a variety of public digital repositories, the integration of this content and its presentation to the end-users for the purpose of fulfilling their search needs. The design of this platform was based on the needs of art searchers, as they were expressed during a series of semi-structured interviews with interested parties from the fields of art and culture. Enhancing the basic search functionality of the platform with voice recognition and synthesis technologies, visual search technologies through the use of pattern recognition and personalization and community features implemented by a simple user system, were integral parts of the design and development of the platform. Beyond fulfilling its main objective, the platform also implemented a data collection system designed specifically for the collection of quantitative measurements regarding the interactions between users with the platform’s various features. After an initial lab study, that also played the role of an evaluation and involved both qualitative feedback in the form of focus group interviews, and also quantitative feedback through statistical analysis of usage metrics, the platform, which was named “Art Boulevard”, was made publicly available both as a Web application and a smartphone app. Starting with its launch and for a period of six months, the platform collected a variety of quantitative data from over ten thousand search sessions, while simultaneously being promoted through marketing techniques and social media, in order to achieve the widest possible outreach. After suitable processing, the collected data formed a dataset containing important information, which was then studied both through means of descriptive statistics and through employing methodologies based on machine learning techniques. During this analysis, special emphasis was given to the differences between art searcher behavior and the behavior of general-purpose search engine users, as well as the factors that influence this behavior. Users, searching online for content related to art and culture, used more keywords per search query focusing on general concepts and creator names, repeated a search query more often in an effort to acquire specialized information, showed interest for a wider range of results trying to gain a more spherical approach of the subject matter, and displayed greater preference for desktop or laptop devices than mobiles over general purpose searchers, thus underlining the nature of art search as a focused and deliberate process. At the same time logged-in users displayed a higher engagement with advanced features and technologies highlighting the value of personalized engagement and users using desktops achieved a higher search success rate confirming the importance of efficiency during the search process. Through machine learning data analysis, search sessions were classified into four archetypes: substantial sessions, short sessions, direct result sessions and expert sessions establishing user intent as a definitive factor on engagement. Moreover, the ability to accurately predict the success of a search session using quantitative behavioral metrics was documented, highlighting the value of using behavioral modeling to accomplish better search interfaces and streamlining the user experience. Finally, a thematic classification of content for which searchers displayed explicit interest was established, showcasing among other things the comparative advantage of traditional visual content in terms of digitization and immediate access to the artistic subject. All of the above findings contributed in creating a better understanding of the distinct habits and needs of art searchers. By incorporating this new knowledge into their design process, content providers operating in the fields of art and culture can improve their offered services. This type of evolution may lead to an overall upgrade of the art searching experience on the Web, which will not only be for the benefit of the searchers themselves and their produced works, but will also offer the content providers the means with which to ensure a wider diffusion of their content and a larger outreach.
Degree
thesis:*- Grantor dc:publisher
- Ionian University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Pergantis, Minas
- Περγαντής, Μηνάς
Subjects
dc:subject × 28- Παγκόσμιος ιστός
- Μηχανές αναζήτησης
- Διαδύκτιο
- Τέχνη
- Πολιτιστική κληρονομιά
- Συμπεριφορά χρηστών
- Τεχνολογίες
- World wide web
- Search engines
- Internet
- Art
- Cultural heritage
- User behavior
- Technologies
- Τέχνες (Τέχνες, Ιστορία της Τέχνης, Ερμηνευτικές Τέχνες, Μουσική)
- Ανθρωπιστικές Επιστήμες και Τέχνες
- Τέχνες, διεπιστημονική προσέγγιση
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Διάδραση ανθρώπου-υπολογιστή
- Διεπιστημονικές εφαρμογές επιστήμης ηλεκτρονικών υπολογιστών
- Arts (arts, history of arts, performing arts, music)
- Humanities and the Arts
- Arts, interdisciplinary
- Computer and Information Sciences
- Natural Sciences
- Human-Computer Interaction
- Computer Science Interdisciplinary Applications
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/54405