OpenITI: Machine-Readable Corpus of Premodern Islamicate Texts

OpenITI is a multi-institutional effort to construct the first machine-actionable scholarly corpus of premodern Islamicate texts. It currently contains almost exclusively Arabic texts, with a more limited number of Persian texts, and is supported by the European Research Council and Qatar National Library. The corpus is built from texts first assembled within the OpenArabic project and the Persian Digital Library pilot.

Use Cases

• [Text mining and analysis] based on a corpus of premodern Arabic and Persian texts
• [Historical linguistic studies] based on machine-readable Islamicate texts
• [Building NLP models for classical languages] based on the Arabic and Persian text corpus
• [Digital scholarship and new forms of textual analysis] based on the machine-actionable corpus infrastructure

Strengths

• Corpus is described as the first machine-actionable scholarly corpus of its kind
• Project involves multiple institutions including the University of Vienna and Aga Khan University
• Supported by funding from the European Research Council Horizon 2020 programme

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download
• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment

Provenance

Source

Open Islamicate Texts Initiative (OpenITI), with texts from the OpenArabic project and Persian Digital Library pilot.

Collection Method

Texts were assembled and compiled by academic projects, with Arabic texts from OpenArabic (2013-2017) and Persian texts from the Persian Digital Library pilot (2015-2016).

Time Range

Covers premodern periods, specific dates not provided.