Towards pertinent evaluation methodologies for word-space models

摘要: This paper discusses evaluation methodologies for a particular kind of meaning models known as word-space models, which use distributional information to assemble geometric representations similarities. Word-space have received considerable attention in recent years, and begun see employment outside the walls computational linguistics laboratories. However, such remain infantile, lack efforts at standardization. Very few studies critically assessed used evaluate word spaces. attempts fill some this void. It is central goal answer question “how can we determine whether given space good space?”