Abstract
dc:descriptionAs analytics moves towards a model of proactive computing, the requirement for forecasting acquires more importance. Systems with forecasting capabilities can play a significant role in assisting users to make smart decisions as soon as critical situations are detected. Being able to forecast that certain patterns in a stream have a high probability of being detected before they are actually detected can help an analyst focus early on what is important and possibly take a proactive action. The need for event forecasting as a means for proactive behavior has led to proposals about how forecasting could be conceptualized and integrated within a complex event processing system. However, such proposals still remain largely at a conceptual level, without providing concrete algorithms. On the other hand, there is a substantial body of work on the related field of time-series forecasting. However, time-series analysis is usually applied on numerical data streams, where each element of the stream corresponds to a measurement of some variable of interest. Moreover, these measurements are often assumed to take place at time intervals of constant length. On the contrary, event processing systems need to be able to additionally deal with symbolic/categorical streams, where each element might be accompanied by arguments, either numerical or symbolic, arriving at unspecified timepoints. The goal of this work is to provide a theoretical basis and build a prototype system for forecasting the occurrence of complex event patterns. This will be achieved by advancing the state-of-the-art in automaton models, going beyond classical automata, so that patterns can have the expressive power required by complex event processing applications. A probabilistic framework will subsequently be used, so that the behavior of these automaton models may be quantified in a way that allows for producing forecasts with confidence. The resulting system will be evaluated against synthetic datasets (for verification purposes) and against real-world datasets in order to demonstrate the applicability of the proposed system.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Ilias, Alevizos
- Αλεβίζος, Ηλίας
Subjects
dc:subject × 27- Επεξεργασία σύνθετων γεγονότων
- Αναγνώριση σύνθετων γεγονότων
- Συμβολικά αυτόματα
- Μαρκοβιανές αλυσίδες
- Μαρκοβιανές αλυσίδες μοτίβων
- Πεπερασμένα αυτόματα
- Κανονικές εκφράσεις
- Άπειρα αλφάβητα
- Μαρκοβιανά μοντέλα μεταβλητής τάξης
- Επιθεματικά προβλεπτικά δέντρα
- Complex event processing
- Complex event recognition
- Symbolic automata
- Markov chains
- Pattern Markov chains
- Finite automata
- Regular expressions
- Register automata
- Infinite alphabets
- Variable-order Markov models
- Prediction suffix trees
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Τεχνητή νοημοσύνη
- Computer and Information Sciences
- Natural Sciences
- Artificial Intelligence
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/51160