National Technical University of Athens (NTUA)
Προδιαγραφική αναλυτική δεδομένων με διαδραστική ενισχυτική μάθηση στη λήψη αποφάσεων
Abstract
dc:descriptionPrescriptive analytics is often considered as the next step towards increasing data analytics maturity and leading to optimized decision making ahead of time for business performance improvement.The current thesis focuses on the utilization of the predictive analytics outcomes in combination with the data-driven decision making, machine learning algorithms and big data technologies to support complex prescriptive analytics problems of various domains. To that end, it develops a prescriptive analytics framework that can guide the development of various prescriptive methods. These methods can support decision making by providing adaptive, automated, constrained, optimal, and time-dependent decisions.The proposed framework is realized through the prescriptive analytics method that consists the main contribution of the current thesis. The proposed method employs Interactive Multi-Objective Reinforcement Learning algorithms. It models the decision making problem dynamically and generates optimized decisions, taking into consideration multiple objectives, while at the same time, through interactive interaction, incorporates the decision maker's preference into the decision process. Furthermore, the method has been implemented as an information system. The system has been employed in a stock market case study in order to evaluate the proposed approach’s performance against the traditional stock trading decision methods and its adaptability according to the user’s preference for different trading profiles.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Lepenioti, Aikaterini
- Λεπενιώτη, Αικατερίνη
Subjects
dc:subject × 12- Προδιαγραφική Αναλυτική
- Ενισχυτική Μάθηση Πολλαπλών Κριτηρίων
- Διαδραστική Ενισχυτική Μάθηση
- Prescriptive Analytics
- Multi-Objective Reinforcement Learning
- Interactive Reinforcement Learning
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Τεχνητή νοημοσύνη
- Computer and Information Sciences
- Natural Sciences
- Artificial Intelligence
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/52850