IIIT-Hyderabad Telugu dependency treebank consisting of around 1600 sentences is made available in ICON 2009 tools contest. Around 1000 sentences are taken from the HCU Telugu treebank and these sentences are manullay validated and added to IIIT-Hyderabad treebank. The treebanks are annotated using Paninian dependency grammar. We clean up the treebank by removing sentences with wrong format or incomplete parse trees etc. The final treebank consists of around 2400 sentences and this is used in parsing experiments.

The intra-chunk dependencies are annotated using the treebank expander made available at https://github.com/ltrc/Shift-Reduce-Chunk-Expander based on Context Free Grammar (CFG) rules for Telugu. The grammar rules for Telugu are available at https://github.com/ltrc/telugu_treebank/blob/master/cfg_grammar/telugu_grammar_anncorra.json

The intra-chunk annotated treebank is made available in CoNLL-X format in WX notation.