Omicini Paolo Petta Matjaž Gams
نویسندگان
چکیده
away from the underlying programming languagein favour of using graphical means of programming, suchas wizards and statecharts. Platforms. Closed frameworks such as DESIRE, strongly based on a platform, provide more complete solutions thanothers such as Jadex or TuCSoN. They usually offer anagent architecture and a system model, very useful fornovel developers, together with the communication infrasA SURVEY OF LANGUAGES AND PLATFORMS. . .Informatica 30 (2006) 33–44 41 tructure and a range of robust services, such as directoryfacilitators, agent management services, and monitoringfacilities. As a drawback, closed frameworks limit thedevelopment. For example, the design approach of theframework may not fit certain domain problems. Perhaps that is the reason why most researchers tend to use moreopen solutions. Currently, the most popular solution isto use JADE as underlying agent infrastructure combinedwith some other (higher-level) approach to program theagents’ behaviour. When dealing with more generalframeworks (rather than tied to a platform), their use (i.e.,defining the agents that will run within it, together withthe required services and resources) should be automatedas much as possible, in part to free the developer fromlow-level details (e.g. location of the configuration files,their concrete syntax, etc.). Despite this, few existingframeworks have IDE support. Concerning the paradigmof communication used, there are several on offer, oftenbeing an important issue when choosing which frameworkto adopt. TuCSoN is representative of tuple-centredcommunication, JADE of message passing, and DESIREof data flow among processes. The various approaches mentioned along this survey in-dicate that there is still much work to be done. Among themajor challenges faced by this research community are: – The conception and development of specialised de-bugging tools, in particular for cognitive agent lan-guages; – The integration of agent tools into existing IDEs,rather than starting from scratch; – The separation of MAS frameworks from agent plat-forms, so that each framework can be used for deploy-ing systems on a variety of platforms. – The dissemination of the MAS programmingparadigm, so that programmers have a better un-derstanding of its foundations as well as practical
منابع مشابه
Guest Editors : Andrea Omicini Paolo Petta Matjaž Gams
Educational networks and portals are a formation of thematically gathered data on the web. Structure of national educational networks and portals depends on environment of their origin. In the paper, selected educational networks are analysed according to the following criteria: content-services, navigation, search, user interface, help, credibility, validity and target groups. The criteria wer...
متن کاملThe AgentLink III Technical Forums: Introduction to the Special Issue
PAOLO PETTA, ANDREA OMICINI, TERRY PAYNE and PETER McBURNEY 1 Austrian Research Institute for Artificial Intelligence, Vienna, Austria, EU, and Medical University of Vienna, Austria, EU 2 DEIS, Alma Mater Studiorum—Università di Bologna, Cesena, Italy, EU 3 School of Electronics and Computer Science, University of Southampton, Southampton, UK, EU 4 Department of Computer Science, University of ...
متن کاملComparing Deep and Classical Machine Learning Methods for Human Activity Recognition using Wrist Accelerometer
متن کامل
Co Fide Ce I Terpretatio a D Preve Tio System
This paper presents the interpretation and prevention system as a part of the product developed at the EU FP7 project Confidence. This system is able to detect falls and general disability. The tests show that it is reliable and ready for real-world usage.
متن کاملTowards Robust Fall Detection
This paper presents a method for classifier development by combining domain knowledge and machine learning. The development is performed in two phases: (1) development of initial hypothesis using domain knowledge or interactive machine learning and (2) refinement of the initial hypothesis using genetic algorithms. The method is presented in the domain of fall detection.
متن کامل