In computer science, '''human-based computation''' or simply '''human computation''' is a technique whenwhere a computational process employs humans to perform its function, using human abilities to solve a certain problem or a set of problems. This approach explores differences in abilities and alternative costs between humans and computer agents to achieve symbiotic human-computer interaction. ▼{{wikify-date|August 2006}}
In traditional computation, a human employs computercomputers to solve a problem: a human provides a formalized problem description to a computer, and receives a solution to interpret. In human-based computation, the roles are reversed: a computer asks a person or often a large number of people to solve a problem, then collects, interprets, and integrates their solutions. Precursors of this idea are interactive programs requesting input from a user, e.g. asking a confirmation to delete a file . However, this concept in its explicit form appeared at the intersection of computer graphics and evolutionary computation. [Sims 91] used human visual perception and esthetic ability to implement evaluation function in evolutionary programming application and evolve pieces of graphic art this way. TheA crucialmore differenceexplicit hereexample isof agency:Human-based Sim'scomputation programis wasthe no[[ESP longerGame]]. anESP agentis ofa itspopular user,online but rather a coordinator ofgame: many humanpeople evaluatorsplay whoover became40 agentshours ofa the programweek. Human-basedBy geneticplaying, they algorithmprovide (HBGA)meaningful, isaccurate akeyword logicallabels extensionfor and a quite general model of this approach basedimages on the ideaWeb; offor outsourcing. Thusinstance, inan HBGAimage humanswith alsoa candog contribute their innovative solutions intogets the processlabel and"dog." haveThese morelabels controlcan overbe theused, functionsamong theyother arethings, performing.to Mostimprove implementations of HBGA also have somethe kindaccuracy of motivationimage systemsearch. that encourages humans to participate.▼▲In computer science, '''human-based computation''' is a technique when a computational process employs humans to perform its function, using human abilities to solve a certain problem or a set of problems. This approach explores differences in abilities and alternative costs between humans and computer agents to achieve symbiotic human-computer interaction.
Human computation focuses on harnessing human time and energy for addressing problems that computers cannot yet tackle on their own. Although computers have advanced significantly in many respects over the last 50 years, they still do not possess the basic conceptual intelligence or perceptual capabilities that most humans take for granted. By leveraging human skills and abilities in a novel way, the hope is to solve large-scale computational problems that computers cannot yet solve and begin to teach computers many of these human talents.
▲In traditional computation, a human employs computer to solve a problem: a human provides a formalized problem description to a computer, and receives a solution to interpret. In human-based computation, the roles are reversed: computer asks a person or often a large number of people to solve a problem, then collects, interprets, and integrates their solutions. Precursors of this idea are interactive programs requesting input from a user, e.g. asking a confirmation to delete a file. However, this concept in its explicit form appeared at the intersection of computer graphics and evolutionary computation. [Sims 91] used human visual perception and esthetic ability to implement evaluation function in evolutionary programming application and evolve pieces of graphic art this way. The crucial difference here is agency: Sim's program was no longer an agent of its user, but rather a coordinator of many human evaluators who became agents of the program. Human-based genetic algorithm (HBGA) is a logical extension and a quite general model of this approach based on the idea of outsourcing. Thus, in HBGA humans also can contribute their innovative solutions into the process and have more control over the functions they are performing. Most implementations of HBGA also have some kind of motivation system that encourages humans to participate.
The following table from [Kosorukoff 00] uses the evolutionary computation model to describe four classes of computation, three of which rely on humans in some role. The classification is in terms of the roles (innovation or selection) performed in each case by humans and computational processes. This table also has a third dimension, determining how agents are organized. Here we assume that organization and coordination are performed by a program.
<center>
<table border=1 cellpadding=0 cellspacing=0 bgcolor=white><tr><td>
<table width=640 border=0 cellpadding=5 cellspacing=1 bgcolor=lightblue>
<tr><th colspan=4>General class of organizational evolutionary methods</th></tr>
<tr bgcolor="#E0E0FF"><th bgcolor=white><font color="#A0A0FF">Selection agent</font><br><hr><font color="#FFA0A0">Innovation agent</font></th><th>Computational</th><th>Human</th></tr>
<tr><th bgcolor="#FFE0E0">Computational</th><th>Parallel GA</th><th>Interactive GA</th></tr>
<tr><th bgcolor="#FFE0E0">Human</th><th>Computer-Aided Design</th><th>Human-based GA</th></tr>
</table>
</table>
</center>
==Methods of human-based computation==
* [[Darwin]] [Vyssotsky, Morris, McIlroy 61] and [[Core War]] [Jones, Dewdney 84] These are gamesactivities where several programs written by people compete in a tournament (computational simulation) in which fittest programs will survive. Authors of the programs copy, modify, and recombine successful strategies to improve their chances of winning.
* [[Interactive genetic algorithm]] [Sims 91] IGA enables the user to create an abstract drawing only by selecting his/her favorite images displayed on the computer screen, so human performs fitness computation. [Unemi 1998] Simulated breeding style introduces no explicit fitness, just selection, which is easier for human. In a typical IGA system, a computer program generategenerates an image and presents it to a human user for evaluation/classification.
* [[Wiki]] [Cunningham 95] enables editing the web content by multiple users, but more importantly, provides the mechanism for reversing the changes that allows for selection among several versions of the page. Wiki is sometimes interpreted as the backronym for "What I know is", which describes the knowledge contribution, storage and exchange function.
* [[Human-based genetic algorithmalgorithms]] [Kosorukoff 98] HBGA uses human selection in the same way as IGA, but also enables users to take part in innovation by performing intelligent mutation and crossover. In this method, all operators of a typical [[genetic algorithm]] are outsourced to humans (hence the name '''human-based'''). The first HBGA is [[3form]] Free Knowledge Exchange that implements collaborative problem-solving mechanism.
* [[Captcha]] (reverse Turing test) [Lillibridge 98, von Ahn 00] Automated tests to distinguish a human user from a computer program using open problems in AI that have no suitable algorithmic solutions. In some sense, they are also a reverse of IGA: a computer program generates an image and presents it to an agent for evaluation/classification, but uses a reply of the agent to classify it into human or non-human category.
* InteractiveGames onlinewith [[guessinga game]]sPurpose [Burgener 99, von Ahn 03]: These are programs that extract knowledge from people in an entertaining way. Some ofexaples theminclude: canThe be[[ESP describedGame]], in termswhich players contribute accurate keyword labels for images; [http://www.peekaboom.org Peekaboom], in which players locate objects within images; and Verbosity, which collects common-sense facts. The output of HBGAeach modelof these games has multiple applications, including training artificial intelligence (AI) systems.
==See also==
* [[Social software]]
* [http://video.google.com/videoplay?docid=-8246463980976635143 Lecture on human computation by Luis von Ahn]
==References==
| 