Further to Mr. Mikolajuk's message (27 June) -
At the UWI (University of the West Indies) Information Technology Conference
Jamaica, 16th and 17th November, 1999, Dr. Han Reichgelt
([EMAIL PROTECTED]) presented a paper describing the creation of an expert
system addressing diseases of coffee. The conference was available in St.
Lucia over UWI's teleconferencing system, so I was able to attend. Here are
my notes on that paper.
4. Artificial Intelligence and Expert Systems
Professor Han Reichgelt
Dept. Mathematics and Computer Science, UWI
Prof. Reichgelt began by stating the problem of the extensive and improper
use of pesticides and the need to protect soils and water supply. One
solution is integrated pest management but this needs considerable
biological and ecological data. These data are either not available, or
exist only in the minds of a few experts. To create the expert system it is
necessary to interview these experts in depth and filter their knowledge to
identify their "rules of thumb" and the actions they would recommend.
The CPEST expert system examines the depradations of two main pests of
coffee - the leaf miner and the coffee berry borer. The program has a
graphical interface to allow for use by people unaccustomed to computers.
Evaluation was carried out by allowing three experts other than those
originally interviewed, and the system, to evaluate the same problem. There
was only one point of disagreement where the system and two experts agreed
against the third.
Implementation problems
1. The farmers are largely unaware of the consequences to the environment of
pesticides, but rather form a "pesticide subculture" - when in doubt, spray
it!
2. The advice of the system is often "do nothing" or "wait", and the farmers
find this hard to follow.
3. Access to computers is severely limited; to solve this problem access can
be given to the extension officers to act as intermediaries.
One of Dr. Reichgelt's students has just completed a similar system for
control of the diamond backed moth, a pest of cabbages etc.
I believe that similar work has been done in Mauritius on sweet potato. At
an ICT and Agriculture conference here someone from the Commonwealth
Secretariat made a presentation about expert "brainstorming" discussion
lists which bring together international expertise to tackle specific
development problems.
Perhaps a means might be discovered to collect existing systems like these,
and also to make the methodologies available so that new systems could be
developed dealing with development issues generally - health, waste
disposal, soil conservation, etc. as well as agriculture.
Best wishes from St. Lucia
Deirdre Williams
"The fundamental cure for poverty is not money but knowledge"
Sir William Arthur Lewis
Zbigniew Mikolajuk <[EMAIL PROTECTED]> wrote:
>
> I think that the biggest challenge is to build the Gateway in such a way
> that it will bring concrete benefits to people at the grassroots level -
> farmers, local organizations, poor and remote communities. It does not mean
>computers in every village, but a system which will allow to make use of
> knowledge available on the Gateway. People will not "access" the Gateway to
> make queries in databases, browse through web pages or perform other
> "computer operations". The issue here is - can the technology help solve
> concrete problems a farmer in Sri Lanka, a self-help group in India, a
> small enterprise in Pakistan may have.
>
>
------------
***GKD is an initiative of the Global Knowledge Partnership***
To post a message, send it to: <[EMAIL PROTECTED]>
To subscribe or unsubscribe, send a message to:
<[EMAIL PROTECTED]>. In the 1st line of the message type:
subscribe gkd OR type: unsubscribe gkd
Archives of previous GKD messages can be found at:
<http://www.globalknowledge.org>
