4
$\begingroup$

I'm not quite sure what "latent" refers to in this context.

In Computing Semantic Relatedness using Wikipedia-based Explicit Semantic Analysis they say

''Our semantic analysis is explicit in the sense that we manipulate 
 manifest concepts grounded in human cognition, rather than 
 'latent concepts' used by Latent Semantic Analysis''.

What does that mean in simple terms?

I had a look at the related Wikipedia articles, ( Latent (wikipedia.org/wiki/Latent_semantic_analysis), Explicit wikipedia.org/wiki/Explicit_semantic_analysis) ), and I wasn't able to make heads or tails of it.

Perhaps someone with a better appreciate of the nuance involved here might be able to provide me with a clear indication of the similarities and differences, the pros and cons between these two methods for accessing document/text fragment relatedness to a particular concept.

$\endgroup$

1 Answer 1

4
$\begingroup$

The difference is that with ESA, the concepts are already known and labeled (hence, "manifest concepts"), whereas in LSA the concepts are latent (they are undefined and need to be discovered).

Note the following statement from the ESA Wikipedia page:

The name "explicit semantic analysis" contrasts with latent semantic analysis (LSA), because the use of a knowledge base makes it possible to assign human-readable labels to the concepts that make up the vector space.[3][1]

In short, ESA uses prior knowledge of relationships between words and concepts (as well as labels for those concepts), whereas LSA does not.

References cited by Wikipedia for the quote above: [1] Ofer Egozi, Shaul Markovitch and Evgeniy Gabrilovich (2011). "Concept-Based Information Retrieval using Explicit Semantic Analysis" (PDF). ACM Transactions on Information Systems 29 (2). Retrieved January 3, 2015. [3] Gabrilovich, Evgeniy; Markovitch, Shaul (2007). Computing semantic relatedness using Wikipedia-based Explicit Semantic Analysis (PDF). Proc. 20th Int'l Joint Conf. on Artificial Intelligence (IJCAI). pp. 1606–1611.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.