Agent Based Introduction

Introduction

An intelligent agent is a software that assists people and act on their behalf. Intelligent agents work by allowing people to delegate work that they could have done, to the agent software. Agents can perform repetitive tasks, remember things you fo rgot, intelligently summarize complex data, learn from you and even make recommendations to you.

What kind of problems Intelligent Agents can solve

To understand how intelligent agents work, it is best to examine some of the practical problems that intelligent agent can help solve. An intelligent agent can help you find and filter information when you are looking at corporate data or surfing the Internet and don't know where the right information is. It could also customize information to your preferences, thus saving you time of h andling it as more and more new information arrived each day on the Internet.

Applications of Intelligent agent

Here are some of the examples that use intelligent agent which illustrate some of the important ways intelligent agents can help solve real problems and make today's computer system easier to use. 

Customer Help Desk 
Customer help desk job is to answer calls from customers and find the answer to their problems. When customers call with a problems, the help desk person manually look up answers from hardcopy manuals, but those hardcopy manuals have been replaced with s earchable CD-ROM collections, and some companies even offer searches over the Internet. Instead of hiring help desk consultants, or having the customers search through the internet for an answer, with intelligent agent, customer describe the problem and the agent automatically searches the appropriate databases (either CD-ROM, or the Internet), then presents a consolidated answer with the most likely first. This is a good example of using intelligent agent to find and filter information. 

Web Browser Intelligent 
A web browser intelligent, such as an IBM Web Browser Intelligent is an agent which helps you keep track of what web site you visited and customizes your view of the web by automatically keeping a bookmark list, ordered by how often and how recent you vis it the site. It allows you to search for any words you've seen in your bookmark track, and takes you back to the site allowing you to find and filter quickly. It also help you find where you were by showing you all the different track you took starting at the current page. It also let you know by notifying you when sites you like are updated, and it could also automatically download pages for you to browse offline. 

Personal Shopping Assistant 
IBM's Personal Shopping Assistant uses intelligent agent technology to help the Internet shopper or the Internet shop owner to find the desired item quickly without having to browse page after page of the wrong merchandise. With the Personal Shopping Ass istant, stores and merchandise are customized as the intelligent agent learned the shopper's preferences as he/she enters in any on-line mall or stores or looking at specific merchandise. It could also arrange the merchandise so that the items you like t he most are the first one you see. Finally, Personal Shopping Assistant automates your shopping experience by reminding you to shop when a birthday, an anniversaries, or item that is on sale occurred. 
 

Characteristic of Intelligent Agent

All agents are autonomous, which means that an agent has control over its own actions. All agents are also goal-driven. Agents have a purpose and act accordance with that purpose. There are several ways of making goals known to an agent, and are listed below: 

  • An agent could be driven by a script with pre-defines action which would then define the agent's goals.
  • An agent could also be a program and as long as the program is driven by goals and has other characteristics of agents.
  • An agent could also be driven by rules, and the rules would define the agent's goals.
  • There is also embedded agent goals, such as "planning" methodologies, and in some cases the agent could change its own goals over time.

 

An agent could also senses changes in its environment and responds to these changes. This characteristic of the agent is at the core of delegation and automation. For example, you tell your assistant "when x happens, do y" and the agent is always wai ting for x to happen. An agent continue to work even when the user is gone, which means that an agent could run on a server, but in some cases, an agent run on the user systems. 

In a Multi-Agent System, agents are social, this means that they communicate with other agents. Some agents learn or change their behavior base on their previous experiences. Some agents are mobile, meaning they move from machine to machine to be clo ser to data they may need to process and do so without network delays. Finally, some agents attempt to be believable, such that they are represented as an entity visible or audible to the user and may even have aspects of emotion or personality.

Enhance search engine performance with Intelligent Agent

Before we rush into how an intelligent agent is used to enhance a search engine performance, let's look at what makes an agent different from a search engine. Some of the search engine such as Yahoo, Lycos, and WebCrawler seem to match the description s of basic intelligent agent. The main difference is that an agent is more interactive and can perform many tasks at different locations. First of all, for example, if you search using a search engine, such as Lycos or Yahoo, you may get a list of match es, which you might have to follow and possibly not get your information. Secondly, using a search engine may increase the percentage of those matches that might not be relevant to the inquiry. However, if you were to use an agent, the agent could submi t your keyword(s) to many different search engines and follow those corresponding links and gather the information without any intervention from the user. 

An intelligent agent uses such technology as the spider, which is also used in the traditional web search engines. However the spider will be a tool which will be used and trained by the user to search the web for specific types of information resourc es. The agent can be personalized by its owner so that it can build up a picture of individual likes, dislikes and precise information needs. Over time, an agent will build up an accurate picture of a user information needs. It will learn from past exp eriences, as a user will have the option of reviewing search results and rejecting any information sources, which are not relevant or useful.

Agent communication languages

Some of the agent communication languages include KQML (Knowledge Query and Manipulation Languages), AOP (Agent Oriented Programming) and Agent Talk. 

  • KQML is a language and protocol used for exchanging information and knowledge. KQML is both a message format and a message-handling protocol to support run-time knowledge sharing among agents. KQML can be used as a language for an application progra m to interact with an intelligent system or for two or more intelligent systems to share knowledge in support of cooperative problem solving.

 

  • AOP is an interpreter for programs written in a language called AO. AO is a programming language for the paradigm of Agent-Oriented Programming. It is currently under development at Stanford.

 

  • Agent Talk is a coordination protocol description language for multiagent systems. Agent Talk allows coordination protocols to be defined incrementally and to be easily customized to suit application domains by incorporating an inheritance mechanism.

Tools & Languages used to implement Intelligent Agent

There are many tools and languages used to implement intelligent agent and here are some of the tools and languages listed below: 

  • Aglet, which is programming code that, can be transported along with state information. Aglets are Java objects that can move from one host on the Internet to another. That is, an Aglet that executes on one host can suddenly halt execution, dispatch i tself to a remote host, and resume execution there. When the Aglet moves, it takes along its program code as well as its data.

 

  • Facile, which is a high-level, higher-order programming language for systems that require a combination of complex data manipulation and concurrent and distributed computing. It combines Standard ML (SML), with a model of higher-order concurrent proc esses based on CCS. Facile is being used at ECRC to develop Mobile Service Agents.

 

  • Penguin, which is a Perl 5 module that provides a set of functions to (1) send encrypted, digitally signed Perl code to a remote machine to be executed; and (2) receive code and, depending on who signed it, execute it in an arbitrarily secure, limited compartment. The combination of these functions enable direct Perl coding of algorithms to handle safe internet commerce, mobile information-gathering agents, "live content" web browser helper apps, distributed load-balanced computation, remote software update, distance machine administration, content-based information propagation, Internet-wide shared-data applications, network application builders, and etc.

 

  • Python, which is an interpreted, interactive, object-oriented programming language. It is often compared to Tcl, Perl, Scheme or Java. It's used quite a bit as an embedded or extension language in hypermedia projects, and is used quite a bit for the s orting of text processing and administrative scripting that Perl is often used for.

 

Conclusion

In conclusion, intelligent agent has been around for years, but the actual implementation is still in its early stages. As agents gain a wider acceptance and become more sophisticated, they will become a major factor in the future of the Internet. In telligent agents will not completely replace surfing altogether, but they will make information gathering much easier for the users or consumer. Instead of searching through lists and lists of unwanted sites, the user could ask their agent to start searc hing, and in a few moments, it come back with the information that is needed immediately.