Abstract
dc:descriptionBased on the 2012 ACM Computing Classification System (ACM CCS), recommender systems are sub-category of Information Systems and Information Retrieval, bridging concepts from those two categories. Recommender systems are basically tools that traditionally predict the likelihood of a user’s preference for an item. They are typically used in a variety of applications such as movies, music, news, books, research articles, search queries, social tags, and products in general. The goal of such a system is to understand the user’s preferences and taste and then recommend items that the user is likely to enjoy or find useful. Modern recommender systems reveal an expansion on the application areas such systems are being used for various scenarios. This brings us to what we call in the context of this dissertation “recommender systems with real-life applications” (or real-life recommender systems in short), that forms the main research direction of this dissertation. The domain of research is the subject of “Recommender Systems”, focusing though on a problem with a greater amount of challenges and extensions. The scope of this research is to provide a new type of recommender systems that spans beyond the traditional virtual frameworks of social networks and web applications. Real-life recommender systems are fed with real-world data (i.e. user trajectories, IoT sensor data, etc.), extract patterns that are correspond to user habits and recommend real-life actions to the user. This thesis, attempts to lay down the fundamental characteristics of such systems and provides a hands-on development of a demo framework for the case of energy consumption and sustainability, providing energy-related action recommendations for reducing users’ energy consumption.
Degree
thesis:*- Grantor dc:publisher
- Harokopio University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Sardianos, Christos
- Σαρδιανός, Χρήστος
Subjects
dc:subject × 21- Ανάκτηση πληροφοριών
- Συστήματα συστάσεων
- Συστάσεις Πραγματικού Χρόνου
- Εξατομίκευση συστάσεων
- Ενεργειακή απόδοση
- Έξυπνα συστήματα
- Information retrieval
- Retrieval tasks
- Recommender systems
- Real-life Recommendations
- Personalized Recommendations
- Energy efficiency
- Intelligent systems
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Τεχνητή νοημοσύνη
- Πληροφοριακά συστήματα
- Computer and Information Sciences
- Natural Sciences
- Artificial Intelligence
- Information Systems
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/53418