Quantitative stock analysis application

In this thesis, an end-to-end automated quantitative stock analysis system is developed by integrating multi-source financial data from official platforms such as the Public Disclosure Platform (KAP) and the Banks Association of Turkey (TBB). The system is built upon a layered architecture using ASP.NET Core, with scheduled data retrieval through Playwright and Selenium, automated preprocessing, SQL-based data storage, and Python-supported visualization components. The platform generates four distinct types of PDF reports—Quantitative, Banking, Insurance, and Factoring—targeting both real sector and financial institutions. It also produces daily market summaries and delivers real-time notifications through a Telegram bot integration. Experimental findings show that the integration of heterogeneous data sources significantly expands the analytical scope, while the inclusion of Excel and PDF modules enhances the level of detail in reports. Performance evaluations indicate up to a 65% reduction in reporting time, leading to faster analysis and improved sensitivity and accuracy in portfolio decision-making through sector-based rating comparisons. The discussion section analyzes challenges related to multi-source integration, such as synchronization issues, timestamp mismatches, and version control complexity. The system architecture, supported by API-based services, embeddable components, and a multi-tenant infrastructure, positions the platform as a commercially viable, full-scale SaaS (Software as a Service) product. In conclusion, the developed system consolidates data accuracy, processing speed, and user interactivity into a unified platform, offering a scalable, sustainable, and extensible decision support tool tailored to the Turkish financial markets.