IBN SINA manuscript

AVICENNA is a handwriting recognition dataset


The task of AVICENNA is to spot arabic words in an ancient manuscript to facilitate indexing. The data were formatted in a feature representation by the group of Mohamed Chériet (Ecole de technologie supérieure de Montréal, Quebec). Special thanks to all the project participants: Reza Farrahi Moghaddam, Mathias Adankon, Kostyantyn Filonenko, and Robert Wisnovsky. A subset of these data were previously used in the Active Learning Challenge under the name IBN_SINA, but the feature representation was modified to prevent matching the samples.
Among the works of Avicenna, whose Arab name is Ibn Sina, his al-Isharat wa-al-tanbihat received the attention of the later scholars more than others. The reception of this work is particularly intensive and widespread in the period between the late twelfth century to the first half of the fourteenth century, when more than a dozen comprehensive commentaries on this work were composed. These commentaries were one of the main ways of approaching, understanding and developing Avicenna’s philosophy and therefore any study of Post-Avicennian philosophy needs to pay specific attention to this commentary tradition. Kashf al-tamwihat fi sharh al-Tanbihat by Abu al-Hasan Ali ibn Abi Ali ibn Muhammad al-Amidi (d. 1243 or 1233), one of the early commentaries written on al-Isharat wa-al-tanbihat, is an unpublished commentary which still await scholars' attention. Solving the problem of the AVICENNA dataset will help making available more easily these unvaluable resources.

CausalityThis dataset is used in the Unsupervised and Transfer Learning Challenge by the Causality Workbench