The Triplet Dataset of Mongolian Pentad-Livestock Economic Domain

Mongolian information processing is the downstream task of natural language processing. Five-livestock culture is an important part of Mongolian traditional culture. Mongolian WordNet can comprehensively reflect the semantic characteristics of Mongolian words. Based on Mongolian WordNet, by collecting and sorting out the knowledge of five livestock, this data set has established various relationships between the concepts of Mongolian five-livestock words, such as partial whole relationship, superior subordinate relationship and disease treatment. It has gathered a lot of five-livestock knowledge, completed the construction of five-livestock structured Mongolian semantic database, and manually processed it into triples to describe concepts and relationships, which is conducive to knowledge representation and reasoning based concept retrieval, whether basic research or application development, all these five-livestock knowledge can play an important role in language engineering.