Agreed on more substantive detail needed. I was surprised at the lack of mention of many of the basic techniques and domains that a person interested in should consider learning about.
The following are all germane but not mentioned: text analysis/mining, controlled vocabularies, indexing, taxonomies, ontology, semantic web, latent semantic analysis, latent dirichlet allocation, corpus analysis, document similarity analysis, tf-idf, ngrams, and skip grams just to mention a few.
In general the article is a good idea but their needs to be more of a description of the domain landscape and then "paths" plotted through that landscape that lead to interesting and useful competency.
That's a great point. I wonder if there would be a better way to introduce meaningful, actionable topics of study to an introductory-level audience of people who may have never heard of NLP.
The following are all germane but not mentioned: text analysis/mining, controlled vocabularies, indexing, taxonomies, ontology, semantic web, latent semantic analysis, latent dirichlet allocation, corpus analysis, document similarity analysis, tf-idf, ngrams, and skip grams just to mention a few.
In general the article is a good idea but their needs to be more of a description of the domain landscape and then "paths" plotted through that landscape that lead to interesting and useful competency.