Stems terms contained in the input documents with the Zemberek stemming algorithm, terms will be disambiguated and reduced to their stem. The Zemberek stemming algorithm works for Turkish texts only.
Warning: It is highly recommended to use this node only with documents that have been tokenized with the Zemberek TurkishTokenizer . Otherwise term information (letter case, tags etc.) might be lost. Please double-check the node configurations of the preceding nodes.
- Type: TableDocuments to preprocessThe input table which contains the documents to preprocess.