This image shows the visualization of historical sematic change of three words.

Historical t-SNE visualization

By utilizing word2vectors you can correspond words to their associative relations based on their historical usage and correlational significance.

University of Texas at San Antonio

Properties of embeddings refers: to window discrepancies, historical impact, and optimization of structure with word relation, depending on data being portrayed.

Semantic Properties

An on-going but a helpful book resource about NLP
https://web.stanford.edu/~jurafsky/slp3/

Speech and Language Processing (3rd ed. draft) 

Depth Size is important for both sparse and dense vectors. When vectors are computed with short context windows we collect a syntagmatic association (neighboring words) resulting in relational association, and when vectors are computed from long context windows we collect a Paradigmatic association (similar neighbors) resulting in topic word association.


Association depth

By using respective analogy solutions, a relational property association can be correlated for words. However, this property alone will not suffice on the growing, non-general, complexity of human cognitive processes.

Association by analogy.

Historical Association

Semantic drift involves an erroneous pattern leading to the introduction of the erroneous tuples and the patterns based on these tuples leaning to the creation of more erroneous patterns in the database. 

Learn Before

Related

Learn After