Technical

📰 I use a combination of Optical Character Recognition (OCR) technologies to extract and structure content from 1980s–1990s Turkish computer magazines. This workflow transforms scanned legacy print media into searchable, analyzable corpora for computational research.

📊 I apply methods such as Structural Topic Modeling (STM) and semi-supervised learning to uncover latent themes and classify large-scale corpora. These approaches enable both exploratory and theory-informed analysis of digital and historical communication patterns.

🗺️ I use Named Entity Recognition (NER) to extract geospatial data from textual sources. This information feeds into geomapping pipelines that visualize the spatial dimensions of political and cultural discourse over time.