Jump to content

Talk:Semantic search

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Needs work

[edit]

This article strikes me as rather outdated, which is surprising given the rather hot topic. Here is a very recent comprehensive survey on the topic: http://ad-publications.informatik.uni-freiburg.de/FNTIR_semanticsearch_BBH_2016.pdf . You should find that useful to improve the article. If you want assistance, please ask, I am one of the co-authors. 23:09, 12 October 2016 (UTC)

This article needs work. Please contribute.--Searchmaven 23:19, 17 July 2007 (UTC)[reply]

My apologies for any offense caused by my of the word "leverage" as a transitive verb. I did consult a dictionary (http://www.answers.com/leverage). Nonetheless, I do prefer your re-wording. :-) Dtunkelang (talk) 18:47, 2 February 2009 (UTC)[reply]

[edit]

I have stumbled upon this page as a result of reading Google's announcement via the Wall Street Journal, March 15th 2012, that

"the company is aiming to provide more relevant results by incorporating technology called "semantic search," which refers to the process of understanding the actual meaning of words."

In my opinion, this imminent breakthrough (over the next couple of years?) will require a major revamp of this current Wikipedia page on the subject. One first step to take is to add in the Wall Street Journal reference. --Михал Орела 12:03, 6 April 2012 (UTC) — Preceding unsigned comment added by MihalOrela (talkcontribs)

paywall

[edit]

Reference 5 (Domain Specific Data Retrieval on the Semantic Web) appears to point to a Springer paywall. aetilley 07:47, 14 May 2015 (UTC)

How is Bing semantic?

[edit]

In what specific sense is Bing a semantic search? It appears to be a fairly popular model of syntactic search, plus features in limited, selected areas that incorporate some meaning, in much the same way that Google does. I would like to note those features on its entry. ...In general, I think there's value in distinguishing general-purpose search sites like Google and Bing from sites that focus on a specific domain (e.g. Swoogle), and also distinguishing the latter from sites that are effectively front-ends to a database in which the semantics are static (I think GoPubMed is basically this, although I could be mistaken). Thoughts? (Should I produce some candidate text?) PaulBrinkley (talk) 23:12, 16 June 2015 (UTC)[reply]

Needs to be overhauled or deleted

[edit]

This entry needs a complete rewrite, or perhaps it should simply be deleted. The definition of "semantic search" relies on a unreliable source, and indeed the term is pretty fuzzy. Beyond that, the opening section of the entry is run-on of a few references without a coherent summary of the definition and aspects of the term. The remaining sections are even worse. It's not clear why disambiguation should be a top-level section. And the lists in the next two sections are arbitrary and not particularly informative.

This entry should either be completely rewritten or deleted. I have nominated it for deletion. Dtunkelang (talk) 04:50, 1 December 2018 (UTC)[reply]

  • Comment I'm not sure why you have nominated it for PROD rather than AfD, or, preferably, tagging it for Cleanup and/or Expert-subject input. "Semantic search" is clearly a term - 6 of the references use it in their titles, as does one of the external links, and earlier discussion on this Talk page indicates that readers and editors have engaged with it. RebeccaGreen (talk) 15:25, 1 December 2018 (UTC)[reply]
  • Comment I agree that it is a phrase that people use, but the question is whether that term is more than a buzzword. That said, I'm coming around to that possibility, especially after seeing the [Semantic Search on Text and Knowledge Bases] survey by Bast et al. But can we agree that, at the very least, the article needs a complete rewrite? If so, let's update the tags to reflect that. I did tag the article for cleanup initially, and then nominated it for deletion. If there's opposition to deleting it, then how about removing the questionable content, reducing this to a stub (perhaps relying on Bast et al as a definition), and then fleshing it out as people find the time to do so? Dtunkelang (talk) 18:29, 1 December 2018 (UTC)[reply]
  • Comment I have gone ahead and removed the deletion and cleanup tags, instead reducing the entry to a stub with a definition citing a reliable source. Dtunkelang (talk) 18:54, 1 December 2018 (UTC)[reply]
    • While I agree with these concerns, I have restored a brief paragraph that helps to introduce the topic for lay readers in addition to the very dry and academic first intro sentence. Of course tone and source may not be ideal, but the information is useful nonetheless for a basic understanding of the topic. GermanJoe (talk) 16:53, 31 January 2019 (UTC)[reply]
    • I think the restoration is good. The entry as it stands is generally useful and does not seem to have any egregious issues. Dtunkelang (talk) 18:39, 4 February 2019 (UTC)[reply]