Your Text
Paste the text you want to summarize into the field below.
Summary
The generated summary will appear here once processed.
Your summary will appear here.
A Turkish automatic text summarization system - Start generating your online Turkish summary
Paste the text you want to summarize into the field below.
The generated summary will appear here once processed.
Your summary will appear here.
This is a working implementation of lexical chain summarization for Turkish, available free on the web and as mobile apps. The notes below describe what it does and where its limits lie.
Lexical chaining comes out of computational linguistics research on text cohesion. The premise is that a coherent text maintains its topic through chains of related words, and that sentences sitting on the strongest chains are the ones most representative of the text.
The system implements this in stages: morphological analysis to reduce Turkish word forms to roots, chain construction over semantically related terms, sentence scoring by chain strength, and selection of top-scoring sentences in their original order.
The same morphological analysis supports two additional functions: a noun extractor and a verb extractor, both of which identify word classes in Turkish text after resolving suffixes to roots.
Being extractive, the system cannot state a conclusion the source text never states in a single sentence. It performs best on expository prose with consistent terminology, and less well on narrative text where topic continuity is carried by plot rather than vocabulary. Documents with heavy formatting artifacts — page headers, reference lists, table contents — produce noisier results unless cleaned first.
The tool is offered through the web interface and the mobile apps. There is no public API at present.
No. Input is processed to produce the summary and is not stored afterward.