Talk:Semantic reasoner

Untitled

I have been reading the Internet Business Logic Web page and it seems that is an application that uses description logic for Bussiness Inteligence. As I see it, the system does use semantic reasoning, but it is not itself a semantic reasoner (probably the semantic reasoner is part of the system, but the system is not a reasoning engine). This article is about semantic reasoners, not applications that use them. IMHO it would be better to create another article or list of systems that use semantic reasoning. I would like to know what other editors think about it, so feel free to refute me :-) Vastag (talk) 18:40, 26 December 2007 (UTC)

Including JENA as a "semantic reasoner" is very questionable---its documentation lists no less than seven different alternative plugins to provide some kind of inferencing support, but none of them are comparable to the "sound and complete" reasoners included in the remainder of this list. List "yes" on support for entailment testing or consistency checking is thus extremely dubious.

Frankly, the feature table on this page seems completely out of place (and the prior feature-by-feature breakdown of SWRL was particularly ridiculous). Perhaps a separate page comparing OWL or SWRL reasoners would be more appropriate. —Preceding unsigned comment added by 129.67.151.47 (talk) 16:58, 11 November 2008 (UTC)

Article quality

This is a poor quality article, starting with the definition of the term. It seems to think this is more general than inference, but it is not. And the article shows a lack of understanding for the basic concepts in semantics and reasoning. Needs a rewrite. The table is not encyclopedic and includes spam. Needs a rewrite. By the way, has anyone here heard of "references".... ? History2007 (talk) 08:19, 24 February 2010 (UTC)

I agree that the current definition is incorrect. Probably the article name needs to be changed also. I think that it would be more correct to describe these reasoners as "Description Logic Reasoners". What changes do you propose for the table?. Vastag (talk) 13:28, 19 May 2010 (UTC)

Actually the introduction just shows a lack of understanding of the basic concepts of "logic hierarchy", as well as two valued, and multi-valued inference. Just needs to be thrown away and rewritten. There is a page on Description logic which does a semi-reasonable (pun intended) job of clarifying how a description logic is a "restricted segment of first order logic", i.e. it is better than propositional logic, but weaker than first order. There is a red link for inexact description logic in that article, and they call it "fuzzy description logic" which is totally incorrect, given that many of the reasoners are inexact, but non-fuzzy. I may get to that later, but again, it is so far off from correct, that it just needs to be thrown away and rewritten.

Now about the table. I do not know and have not tested all the software packages listed here. Usually software developers promise 20 times more than they deliver, specially on complex items like reasoning, so I do not see how I can fix the table given that the content seems to be driven by reading user manuals, or worse brochures. If it were up to me, I would have this page deleted. History2007 (talk) 15:51, 24 May 2010 (UTC)

Reasoner comparison unreadable

Could anyone please transpose the "Reasoner comparison"-table? It's simply unreadable for normal users. Not everyone has a 3000-pixel-width resolution or more and a wall filling screen to be able to read a 15-column table with text content! Thanks — Preceding unsigned comment added by 141.76.82.62 (talk) 09:48, 22 February 2012 (UTC)

Reasoner comparison

As part of cleanup to save this article, I'm deleting non-notable entries in the lists. However, this table as noted above needs both a transposition and a culling of non-notable entries. So I'm transferring it to the talk page, until we can get around to it. --Mark viking (talk) 18:56, 22 January 2014 (UTC)

	BaseVISor	Bossam	Cyc	Hoolet	Pellet	KAON2	RacerPro	Jena	FaCT	FaCT++	SweetRules	OWLIM	OntoBroker	HermiT
OWL-DL Entailment	No	Unknown	Yes	Yes	Yes	Yes	Yes	No complete reasoner included with standard distribution	Yes	Yes	No	No	Yes	Yes
Supported expressivity for reasoning	R-entailment, OWL 2 RL	Unknown	High-order	Unknown	SROIQ(D)	SHIQ(D)	SRIQ(D-)	varies by reasoner (incomplete for nontrivial description logics)	SHIQ	SROIQ(D)	Unknown	R-entailment, OWL 2 RL	OWL: SHIQ(D) (for OntoBroker 6.1: Subset of OWL-RL); F-logic: normal logic, wellfounded semantics	SHOIQ+
Reasoning algorithm	Rule-based, Rete	Rule-based	First-order with high-order extensions	First-order prover	Tableau	Resolution & Datalog	Tableau	Rule-based	Tableau	Tableau	Rule-based	Rule-based	OWL: Resolution & Datalog; F-logic: Rule-based (BottomUp, MagicSet, Dynamic-Filtering, QSQ)	Hypertableau
Consistency checking	Yes	Unknown	Yes	Yes	Yes	Yes	Yes	Incomplete for OWL DL	Yes	Yes	No	Yes	Yes	Yes
DIG Support	No	No	No	No	Deprecated	Deprecated	Deprecated	Yes	Yes	Yes	No	No	Yes	No
Rule Support	Yes (Own rule format)	Yes (SWRL & own rule format)	Yes (Own rule format)	Yes (SWRL)	Yes (SWRL—DL Safe Rules)	Yes (SWRL—DL Safe Rules)	Yes (SWRL—not fully supported) & own rule format	Yes (Own rule format)	No	No	Yes (SWRL, RuleML, Jess)	Yes (Own format)	Yes (SWRL, RIF, F-logic, ObjectLogic)	Yes (SWRL—DL Safe Rules)
Version	2.0	0.9b45	Unknown	Unknown	2.0.2	2008-06-29	2.0 preview	2.5.4	Unknown	1.6.0	2.1	2.x/3.x	6.1	1.3.6
Licensing	Free to use / closed-source	Free to use / closed-source	Free to use / closed-source (academic and research only) & Non-Free / closed-source	Free / open-source	Free / open-source & Non-Free/ closed-source	Free to use / closed-source	Non-Free/ closed-source	Free / open-source	Free / open-source	Free / open-source	Free / open-source	Non-Free/ closed-source	Non-Free/ closed-source	Free / open-source