e-tourism is an user adopted plan to trip will first initiate a search through web, idetifying the locations, map the distances, communication facilities exists etc. are searched through web. we already have an application to tmplement this ystem through Multi Agent System. These systems llows new users to enter into systems any time, also interact with travel agents to map their Trips.
* The agent should also be able to reason with out all the information, because many times not all the information is accessible. That situation should not stop the problem solving task and should try to look for alternative solutions.
* In the e-tourism domain, the main electronic information sources will be accessible through web. i.e easily managed by the travel agent.
* Some popular approaches used in the web have been the Web Agents, especially the softbots and spiders. Systems named search engines have been very successful in the last few years, be
It is very useful to the travel agents to mange very easily.
* Now a days it is possible to find systems that allow defining and proving agents with different skills, language communications like multimedia etc. The objective of Zeus project was to facilitate the rapid development of new multi agent applications by abstracting into a tool kit, the common principles and components underlaying some existing multi agent systems.
There are different approaches to work with the information stored in the web. This situation analysis some of them that uses the Artificial Intelligence techniques. The key Abstraction used in these systems is the agent concept. i.e Multi-agent systems(MAS) are are sub field of distributed Artificial Intelligence.
*They are able to solve big-size problems.
*They allow different systems to work interconnected and cooperate.
*They provide efficient solutions where information is distributed among different places.
*They allow software reusability, there fore there is more flexibility to adopt different agent capabilities to solve a problem.
* UserAgent: UserAgent pays attention to the users queries and shows the solution. It analysis the problem and obtains an abstract representation. Subsequently it requests to a planner agent for a solution to that problem. The usergent has different skills like communication with planneragents and users, or learning the users profiles necessary to customize the systems and answer the queries.
* PlannerAgents: The main goal of PlannerAgent is reasoning about UserAgents and other Planneragents problems, and find the set of possible solutions. These agents have different communication skills, planning and learning t
* WebBot: This agent fills in the details that are requested by PlannerAgents and obtaining the required information from internet. Different partial solutions given by the Webagents are combined by the Planneragents.
TravelPlan has a cooperative architecture where different agents need to cooperate to reach solutions obtained by the PlannerAgents. travelPlan success needs both characterstics sharing knowledge to obtain new solutions.
Shared Information amoung Agents:
TravelPlan agents needs to implement a communication language that tells how Data and Information transmitted from UserAgents to Planneragents to WebBot and vice versa, by using Request, Accept, Inform, Reject, Request-to-do.
In communication process each agent acknowledges the messages and generate the performative accept. when processed they send answer message to the agent. the Planneragent sends a proposal for cooperation, to the WebBot, and this agentjejects the request. When the PlannerAgent really needs the information sends a new request and finally obtained the desired information.
TravelPlan like other MAS uses a set of grahical user interface (GUI) to communicate with the users.
Conclusion: The UserAgent has different user interfaces to communicate with the user. The first interface is to Input the Problem from the User and second one is optional and it it used to define the user preferences. when the System solves the problem, the PlannerAgent returns sample solutions to the user. When the system solves the problem example, the PlannerAgent returns sample solutions to the user. The system rejects a set of Possible solutions in order go gain in efficiency and only a subset of solutions is shown to the user. Travelplan recommends a particular solution. If this solution matches with the learned preference from the user. To do this, the UserAgent that pays attention to the users, extracts the useful characterstics from the old stored solutions by the users, and uses them to classify all the possible solutions.