Brought to you by EarthWeb
IT Library Logo

Click Here!
Click Here!

Search the site:
 
EXPERT SEARCH -----
Programming Languages
Databases
Security
Web Services
Network Services
Middleware
Components
Operating Systems
User Interfaces
Groupware & Collaboration
Content Management
Productivity Applications
Hardware
Fun & Games

EarthWeb Direct EarthWeb Direct Fatbrain Auctions Support Source Answers

EarthWeb sites
Crossnodes
Datamation
Developer.com
DICE
EarthWeb.com
EarthWeb Direct
ERP Hub
Gamelan
GoCertify.com
HTMLGoodies
Intranet Journal
IT Knowledge
IT Library
JavaGoodies
JARS
JavaScripts.com
open source IT
RoadCoders
Y2K Info

Previous Table of Contents Next


6. FUTURE TRENDS

In effect, the future trends are highly related to the research issues discussed in the previous section. It is expected that the agent-based approach will be extensively used in solving the problems resulting from the integration of expert systems with other software components. The agent-based approach will address other issues related to heterogeneous components that are distributed on different platforms. The agent-based approach is a general approach suitable for any domain. In agriculture, we can see its usage in integrating GIS, multimedia, and simulation models with expert systems.

Developing domain-specific tasks in agriculture is a very important future trend that will help in knowledge sharing and reuse and in automating the knowledge acquisition process. Successful results in these two directions will expedite the development procedure of expert systems. Machine learning will help also in automating the knowledge acquisition process and more attention will be given to it in the near future.

Sophisticated user interfaces for different media types are expected to be an important issue. Different user interface models will also be investigated in as much as agricultural expert systems have different types of users: researchers, extensionists, and growers. Their requirements and needs are different.

More sophisticated explanations facilities will be provided once domain-specific models are well established. An explanation facility should not match what expert systems are providing now -- the why and how primitives, but it should be intelligent enough to generate the explanation based on the user level. Intelligent agent-based approach may also be used in developing such an explanation facility.

Another issue that will play an important role is the usage of the Internet to access expert systems developed in different locations. There is a trend now to develop tools to facilitate the dissemination of expert systems through the WorldWide Web. It is expected in the near future that shells and tools will enable developers to put their expert systems on the Web.

SUMMARY

This chapter has discussed the needs of expert systems in agriculture and revealed their importance as tools for information transfer through information generation from knowledge and expertise. The advantages of an expert system over traditional methods include: providing the growers with dynamic information related to their actual situations, taking into consideration different specialties and different sources of information; shortening the update time of information, especially if the expert system is centralized and accessible from different locations; and transferring real experience that is not documented in any form of media by acquiring it from its sources: extensionists, highly experienced growers, and/or researchers.

The expert systems selected to demonstrate applications in the 1980s revealed that these systems actually addressed issues that need further research. The usage of machine learning to generate the knowledge base was addressed in the soybean diagnosis system. The integration of simulation model with expert system was considered and implemented in the cotton expert system. The concept of a task-specific package was raised in the CALEX system. The importance of an expert system as an appropriate technology to expedite agricultural desert development was emphasized in the late 1980s by Rafea and El-Beltagy (1987).

In the 1990s, expert systems have been expanded to include other commodities and disciplines such as agroforestry and meteorology. The second-generation expert systems methodologies have been applied. More sophisticated expert system shells were used, such as NEXPERT/OBJECT and KEE. Programming languages such as Prolog and Small Talk were also used. The integration of multimedia was first addressed due to the advances made in this technology. Expert systems in the field for real use have been deployed.

Research issues in agricultural expert systems are categorized under these topics: integration of software components with agricultural expert systems, knowledge sharing and reuse, intelligent retrieval of agricultural data, and automatic knowledge acquisition. The future trends in research and development of agricultural expert systems are expected to be: using agent-based approaches to solve the integration problem of different software components, developing domain-specific tasks that will contribute to knowledge sharing and reuse, and automatic knowledge acquisition. Sophisticated user interfaces and explanation facilities that depend on the user level are expected to be seriously considered. The dissemination of expert systems through the Internet is also anticipated.


Previous Table of Contents Next

footer nav
Use of this site is subject certain Terms & Conditions.
Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details.