Ανάλυση χρηματοοικονομικών δεδομένων με ενίσχυση τεχνητής νοημοσύνης
An AI-enhanced financial data analytics dashboard
View/
Keywords
Ποιότητα δεδομένων ; Καθαρισμός δεδομένων ; Τυποποίηση ; Διαδικτυακή εφαρμογή ; Πρόβλεψη ; Επιχειρηματική ευφυΐα ; FastAPIAbstract
The continuous digitalization of business processes has significantly increased the volume
of available data, without necessarily ensuring that such data can be directly used
for analysis. In practice, files originating from different systems frequently present heterogeneous
structures, field names, date formats, numerical representations and levels of
completeness, making systematic cleaning, standardization and validation necessary. The
aim of this thesis is the design and development of a web-based application for the automated
ingestion, processing, storage, classification, analysis and utilization of business
data. The application follows a separated architecture, using Next.js, React and Type-
Script for the presentation layer, Python and FastAPI for the backend, and Supabase for
relational persistence through PostgreSQL and physical file management through object
storage. Particular emphasis is placed on traceability, since the original file, its metadata,
the raw and cleaned records and the main processing stages remain connected through
unique identifiers and relationships between database tables. In parallel, an independent
time-series forecasting subsystem is developed, in which multiple forecasting models can
be evaluated using common error metrics and backtesting procedures before model selection.
Processed data and forecasting outputs can subsequently be consumed by a business
intelligence environment for the development of interactive reports and analytical visualizations.
The resulting approach establishes a repeatable and extensible workflow from
heterogeneous source files to structured and analytically useful information, reducing repeated
manual processing while improving consistency, auditability and the ability to extend
the system with additional processing rules, data sources and analytical methods in
the future.
Postgraduate Studies Programme
Εφαρμοσμένη ΣτατιστικήDepartment
Σχολή Χρηματοοικονομικής και Στατιστικής. Τμήμα Στατιστικής και Ασφαλιστικής ΕπιστήμηςNumber of pages
118Language
GreekCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα