Difference between revisions 945171198 and 946623703 on enwiki

{{AFC submission|d|mergeto|Computational intelligence|u=ManuelRodriguez|ns=118|decliner=AngusWOOF|declinets=20200309160440|ts=20200309093942}} <!-- Do not remove this line! -->

(contracted; show full)

==History==

The phrase ''soft computing'' was coined in the 1990s by [[Lotfi A. Zadeh]], a pioneer in the field. Zadeh was also a pioneer in the related field of ''Fuzzy logic'' and the concept of [[fuzzy sets]].
 He improved a range-based logic system into a [[universal science]] which stands in opposition to Artificial Intelligence{{clarify}}. According to soft computing [[advocacy group]]s, hard computing is not equipped  to analyze cognitive processes.{{sfn|Gudwin|1999|p=160}} 

Soft computing, as opposed to traditional computing, deals with approximate models and gives solutions to complex real-life problems such as [[fuzzy control system]]s, fuzzy graph theory{{sfn|Sunitha|Sunil|2013}}, [[fuzzy systems]], and so on.   Zadeh observed that people are good at 'soft' thinking while computers typically are 'hard' thinking.{{sfn|Jin|2014}}  People use concepts like 'some', 'most', or 'very' rather than 'hard' or precise concepts, values and quantities. People want a 'warm' glass of milk, not one that is 41 degrees C. In general, people are good at learning, finding patterns, adapting and are rather unpredictable.  In 'hard' computing, by contrast, machines need precision, determinism and  measures, and although pattern recognition happens, there is a brittleness if things change - such systems cannot easily adapt. 'Soft' computing by contrast embraces chaotic, neural models of computing that are more pliable.  Soft computing involves using a combination of methods that are designed to approximate human learning, decision making and intelligence. 

Although soft Computing became a formal area of study in Computer Science in the early 1990s, the concepts underpinning the field had been developed since earlier in the century.{{sfn|Zadeh|1994|pp=77-78}}{{failedverification|date=January 2020}} eEarly soft   computational approaches could model and precisely analyze only relatively simple systems. More complex systems arising in [[biology]], [[medicine]], the [[humanities]], [[management science]]s, and similar fields often remained intractable to conventional mathematical and analytical methods.  Complexity of systems is relative and many conventional mathematical models have been very productive in spite of their complexity.{{cn|date=January 2020}}

Soft computing deals with imprecision, uncertainty, partial truth, and approximation to achieve computability, robustness and low solution cost. As such it forms the basis of a considerable amount of [[machine learning]] techniques. Recent trends tend to involve evolutionary and swarm intelligence based algorithms and [[bio-inspired computation]].{{sfn|Yang|2013}}{{sfn|Chaturvedi|2008}}

Generally speaking, soft computing techniques resemble biological processes more closely than traditional techniques, which are largely based on formal [[logical system]]s, such as [[sentential logic]] and [[predicate logic]], or rely heavily on computer-aided numerical analysis (as in [[finite element analysis]]). Soft computing techniques are intended to complement each other.

Unlike hard computing schemes, which strive for exactness and formal proof, soft computing techniques exploit the given tolerance of imprecision, partial truth, and uncertainty for a particular problem. Another common contrast comes from the observation that [[inductive reasoning]] plays a larger role in soft computing than in hard computing.

Research in soft computing have demonstrated some practical examples,{{sfn|Zhou|2000|pp=238-250}}{{sfn|Karray|2002}} such as Neuron [[MOS transistor]]s {{sfn|Shibata|1999|pp=648-656}} and emotional pets.{{sfn|Dote|2001}} Critics of soft computing argue that the existing mathematical tools employed in hard computing  are sufficient to fulfill current and future demands.{{cn}}

Within the history of Artificial Intelligence there is a trend towards including [[linguistics]] and probability theory into existing software frameworks.{{sfn|Seising |2013}} The highly successful innovation of [[deep learning]] is based on the idea, that input data is transformed into abstract representation. This development shows that so called hard computing has sufficient tools to deal with imprecise information and there is no need to introduce fuzzy logic.{{cn}}

==References==
{{reflist}}

==Bibliography==
*{{cite book |last1=Samir |first1=Rov |last2=Chakraborty |first2=Udit |title=Introduction to soft computing : neuro-fuzzy and genetic algorithms |date=3 June 2013 |publisher=Pearson |isbn=978-8131792469 |ref=harv }}
(contracted; show full)
* {{cite journal |last1=Zadeh |first1=Lotfi A. |url=https://go.galegroup.com/ps/i.do?id=GALE%7CA15061349&sid=googleScholar&v=2.1&it=r&linkaccess=abs&issn=00010782&p=AONE&sw=w |title=Fuzzy Logic, Neural Networks, and Soft Computing| journal= Communications of the ACM|date=March 1994 |volume=37 |number=3| pages =77-84 |ref=harv}}

{{Authority control}}

[[:Category:Scientific modeling]]
[[:Category:Artificial intelligence]]
[[:Category:Semantic Web]]
[[:Category:Soft computing]]