Corpus-Based Annotation Editor Developed For Turkish

Studies that have been done in recent years, particularly in cognitive linguistics, are towards the comprehension of human mind structure. Human mind has a structure that includes high-level processes (thinking, sensing, deciding etc.) of the brain. In the other hand, language servesas a bridge between meaning and form by playing a significant role in forming and interpretingof these mental processes. Semantic analysis of language in particular requires the usage of amodel of reality. Ontologies are the most common models that are created by human mind.Due to being naturally large, ontologies carry the potential of hosting error and deficiency. Thus, it is needed that ontologies should be created in a computerized environment by using aformal language. In this context, thematic lattice models created within the framework of Formal Concept Analysis Theory (FCA) could be addressed as an ontology for Turkish language. Through these models, a dataset can be achieved from the syntactic/morphologicaland semantic analysis of Turkish language with the annotation editor developed on the webenvironment. Furthermore, a lexical source for Turkish language in the electronical environment can be achieved based on this dataset by classifying Turkish words types with artificial intelligence algorithms. While creating a significant lexical source for translationsystems, it would also benefit the semantic analysis of Turkish language. In this sense, considering that Turkish language is an agglutinative language, a corpus-based annotation editor that provides accurate analysis of words with their roots and affixes gains a great importance.