Desiderata for Vector-Space Word Representations
It provides foundational guidelines for researchers and practitioners in natural language processing, though it is incremental as it systematizes existing practices rather than introducing new methods.
The paper addresses the lack of design principles for vector-space word representations by detailing desiderata to ensure they are suitable for information processing and data mining techniques.
A plethora of vector-space representations for words is currently available, which is growing. These consist of fixed-length vectors containing real values, which represent a word. The result is a representation upon which the power of many conventional information processing and data mining techniques can be brought to bear, as long as the representations are designed with some forethought and fit certain constraints. This paper details desiderata for the design of vector space representations of words.