Embeddings models for Buddhist Sanskrit: Evaluation Datasets

Evaluation Dataset used for the study published as Embeddings models for Buddhist Sanskrit, LREC 2022 proceedings. It contains a semantic similarity dataset and an analogy dataset, as well as the published study and a ReadMe file containing the guidelines used for scoring semantic similarity and some notes about the manual scoring task. The evaluation datasets have been prepared by Ligeia Lugli, Bruno Galasek-Hul, Luis Quiñones and Jai ParanjapeThis study was funded by a NEH Digital Advancement Grant level 2 (HAA-277246-21