]> 2008/06/10 OpenCyc Knowledge Base Copyright© 2001-2008 Cycorp, Inc., http://www.cyc.com/, Austin, TX, USA This file contains an OWL representation of information contained in the OpenCyc Knowledge Base. The content of this OWL file is licensed under the Creative Commons Attribution 3.0 license whose text can be found at http://creativecommons.org/licenses/by/3.0/legalcode. The content of this OWL file, including the OpenCyc content it represents, constitutes the "Work" referred to in the Creative Commons license. The terms of this license equally apply to, without limitation, renamings and other logically equivalent reformulations of the content of this OWL file (or portions thereof) in any natural or formal language, as well as to derivations of this content or inclusion of it in other ontologies. externalID A unique, language-neutral, variable-sized identifier for a concept that can be used to refer unambiguously to that concept across OWL exports or across Cyc inference engines. label A natural-language representation for a concept that is both human readable and readable by the Cyc inference engine. These terms are not guaranteed to refer to the same concept across time but are guaranteed to be consistent within a particular OWL export. Use 'cycAnnot:externalID' for unambiguously referring to a concept across OWL exports or across Cyc inference engines. MachineLearning machine learning A collection of events in which a program is running and it could develop new information (in some representational form) based on existing information. This collection is neutral on the success of the process. Thus, if it may not generate any new information given a particular set of initial data. http://en.wikipedia.org/wiki/Machine_learning discriminative weight learning A specialization of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rgBzbIHYqQdmfo5QMiMv_Hw" class="cyc_term">MachineLearning</a>. Each instance of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rygYIHJp2EdqAAACQJw5drg" class="cyc_term">DiscriminativeWeightLearning</a> is an event in which a computer program learns weights for formulas using a discriminative (as opposed to generative) algorithm. ("Discriminative," as it is being used here, is a technical term which we will not attempt to define in this comment.) See also <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4r23ndqJp2EdqAAACQJw5drg" class="cyc_term">GenerativeWeightLearning</a>. DiscriminativeWeightLearning generative weight learning A specialization of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rgBzbIHYqQdmfo5QMiMv_Hw" class="cyc_term">MachineLearning</a>. Each instance of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4r23ndqJp2EdqAAACQJw5drg" class="cyc_term">GenerativeWeightLearning</a> is an event in which a computer program learns weights for formulas using a generative (as opposed to discriminative) algorithm. ("Generative," as it is being used here, is a technical term which we will not attempt to define in this comment.) See also <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rygYIHJp2EdqAAACQJw5drg" class="cyc_term">DiscriminativeWeightLearning</a>. GenerativeWeightLearning A sub-collection of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rgBzbIHYqQdmfo5QMiMv_Hw" class="cyc_term">MachineLearning</a> in which a particular program is trying to develop <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvViA-JwpEbGdrcN5Y29ycA" class="cyc_term">implies</a> rules based on some body of knowledge. machine rule induction MachineRuleInduction MachineLearning machine learning A collection of events in which a program is running and it could develop new information (in some representational form) based on existing information. This collection is neutral on the success of the process. Thus, if it may not generate any new information given a particular set of initial data. ComputerActivity computer activity A collection of events, and a specialization of both <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvVioQZwpEbGdrcN5Y29ycA" class="cyc_term">Action</a> and <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvVjpjZwpEbGdrcN5Y29ycA" class="cyc_term">PhysicalEvent</a>. In each instance of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvViFUJwpEbGdrcN5Y29ycA" class="cyc_term">ComputerActivity</a>, at least one instance of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvVjXzZwpEbGdrcN5Y29ycA" class="cyc_term">Computer</a> is somehow involved. An instance of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvViAkpwpEbGdrcN5Y29ycA" class="cyc_term">Person</a> may or may not be involved. (Instances of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvViFUJwpEbGdrcN5Y29ycA" class="cyc_term">ComputerActivity</a> where no <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvViAkpwpEbGdrcN5Y29ycA" class="cyc_term">Person</a> is involved fall under the more specialized collection <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvVj8AZwpEbGdrcN5Y29ycA" class="cyc_term">ComputerProcessRunning</a>.) A wide variety of activities falls into this collection - from sending an email (see the more specialized collection <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvdMvgJwpEbGdrcN5Y29ycA" class="cyc_term">EMailSending</a>), to scanning a set of ports (see <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rwPae0JwpEbGdrcN5Y29ycA" class="cyc_term">PortScan</a>), to a computer crash (see <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvVjkXZwpEbGdrcN5Y29ycA" class="cyc_term">ComputerCrashing</a>). MachineLearning machine learning A collection of events in which a program is running and it could develop new information (in some representational form) based on existing information. This collection is neutral on the success of the process. Thus, if it may not generate any new information given a particular set of initial data. MachineLearning machine learning A collection of events in which a program is running and it could develop new information (in some representational form) based on existing information. This collection is neutral on the success of the process. Thus, if it may not generate any new information given a particular set of initial data. MachineLearning machine learning A collection of events in which a program is running and it could develop new information (in some representational form) based on existing information. This collection is neutral on the success of the process. Thus, if it may not generate any new information given a particular set of initial data. type of temporally stuff-like thing that is also a type of durative event A <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rHQdVmB_TEdaAAABQ2rksLw" class="cyc_term">SecondOrderCollection</a>. <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rsptUGhfiQdmQ1dIv6htA7Q" class="cyc_term">ProcessType</a> is the collection intersection of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvVitN5wpEbGdrcN5Y29ycA" class="cyc_term">TemporalStuffType</a> and <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rwNyXxJwpEbGdrcN5Y29ycA" class="cyc_term">DurativeEventType</a>; that is, a thing is an instance of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rsptUGhfiQdmQ1dIv6htA7Q" class="cyc_term">ProcessType</a> if and only if it is an instance of both <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rvVitN5wpEbGdrcN5Y29ycA" class="cyc_term">TemporalStuffType</a> and <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rwNyXxJwpEbGdrcN5Y29ycA" class="cyc_term">DurativeEventType</a>. Instances of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rsptUGhfiQdmQ1dIv6htA7Q" class="cyc_term">ProcessType</a> can be constructed from instances of <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4rv48ocpwpEbGdrcN5Y29ycA" class="cyc_term">AccomplishmentType</a> using <a href="http://sw.opencyc.org/2008/06/10/concept/Mx4r0VWpmNnEQdiJDq0mDlTBkQ" class="cyc_term">ProcessFn</a> (q.v.). ProcessType