Sie sind auf Seite 1von 3

2013 4th International Conference on Intelligent Systems, Modelling and Simulation

JESS Based Expert System Architecture For Diagnosis Of Rice Plant Diseases:
Design And Prototype Development
Khumukcham Robindro#1, Shikhar Kr. Sarma*2
#Department of Computer Science, Gauhati University
Guwahati, Assam, India – 781014.
rbindkmchm@gmail.com
*Department of Information Technology, Gauhati University
Guwahati, Assam, India – 781014.
sks001@gmail.com

Abstract - An expert system is computer program composed of expert system providing the control measures of identified
knowledge base, inference engine and user-interface. Its
technical aspect involves the design and implementation of the
diseases. The system development begins with the disease
architectural model of an expert system namely the knowledge symptoms collection of the rice plant appearing during their
base component, the graphical user interface component, the life span from agriculture experts, plant pathologists and
application component and the database. This paper presents literatures related to the rice plant diseases. Then the
the importance of the design and development of an expert acquired knowledge is represented to develop expert system
system for diagnosis of diseases in rice plants. Further, such an using the Java Expert System Shell (JESS).
expert system is very essential for the farmers as experts are
not always available with the farmers and the diseases are one II. EXPERT SYSTEM IN AGRICULTURE
of the important factors for unproductivity. When a farmer
Expert System can generate the information to the farmers
needs advice of an expert, this expert system would provide the
decision support to the farmer. Here, we have implemented the by using its knowledge base and reasoning mechanism from
User Interface component using Java, the application human experts and other sources. Thus, it can solve the
component using the Java Expert System Shell (JESS) and the above said problem to a large extent. When a user enters the
Java IDE of Netbeans 7.0 while the database component using data of his plantation to the system, the appropriate advice is
SQL. generated based on its knowledge base and reasoning
mechanism. There is no limitation on the number of
Keywords - Expert System, Rice Diseases, JESS, Knowledge generated recommendation. Therefore it overcomes the
Base. problem of static information provided in extension
documents.
I. INTRODUCTION The integration of knowledge and experiences of different
More than half of the world's population depends on rice, specialists are facilitated on the expert system. When
which is an important staple food (FAO, 2004), occupying problems occur related to the nutrition, plant pathology,
145 million hectares of land. Its annual yield worldwide is entomology, breeding and production, the expert system can
approximately 535 million tons. Almost 90 percent of the help the users in identifying the cause of the problem in an
world’s rice is produced and consumed in the six Asian efficient way, thereby consulting a document that handles
continental countries which are namely China, India, the specific problem.
Indonesia, Bangladesh, Vietnam and Japan. They comprised Expert System makes easier for user to identify the
80% of the world’s production and consumption of rice. describe symptoms like image bases or textual bases
Over 300 million acres of Asian land is used for growing information as it is very difficult to describe in words. It can
rice. Rice production is very important to Asian cultures. also be integrated with textual database which can be used
Such an important food crop is occasionally infected by for explanation purposes of basic terms and operations to
many diseases and reduces the productivity. For the confirm and to reach conclusion in some situations.
identification, a disease is an abnormal condition which Confirming the diagnosis of the rice plant and the cause of
affects the rice plants making it to function improperly. certain disorder can also be done through image suitability.
Though identification of a disease was not an easy task, it The applications of Expert System in agriculture are
can be recognized from the symptoms associated and mainly found in the area of diseases diagnosis and pest
changes in the plant. Finally, it concludes that identification controls. Many domain specific expert systems are being
of diseases needs experience and knowledge of the rice used at different levels in agriculture. “An Expert System for
plants and their diseases. A person can also depend on a diagnosis of diseases in Rice Plant,” is an application of
system for accuracy in describing the symptoms of the expert system in agriculture and developed by Shikhar Kr.
diseases and identifying any disease, making right decision Sarma, Kh. Robindro Singh & Abhijeet Singh [1]. In this
and choosing the right treatment. Keeping the above points system, the expert system is developed with rule-based
in mind, our main concern is design and development of an expert system, using ESTA. Another expert system

2166-0662/13 $26.00 © 2013 IEEE 674


DOI 10.1109/ISMS.2013.54
“Developing an Expert System for plant disease diagnosis,” are usually written for specific functions in Java
is also an expert system developed by S.S. Abu-Naser, environment.
K.A.Kashkash and M. Fayyad[2]. The system is for the
purpose of plant disease control using the shell CLIPS. V. MATERIALS AND TECHNICAL METHODS
“Expert Systems Applications: Agriculture”, is also the The technical methods of the developed system involved
application of expert system in the agriculture domain the design and implementation of components namely the
developed by Ahmed Rafea[3]. “Web based Expert System User component, User interface component, Application
for Diagnosis of Micro Nutrients Deficiencies in Crops”, by component (JESS) and Database component (SQL) as
S.S.Patil, B.V.Dhandra, U.B.Angadi, A.G.Shankar, and shown in fig.1. The Graphic User Interface (GUI)
Neena Joshi also describes application of expert system in component of the system is implemented using the Java
agriculture particularly in the area of nutrient deficiencies in Programming language. We have used the Java Expert
crops [4]. The system is a web based system using the System Shell (JESS) and the Java IDE of Netbeans 7.0 to
ServCLIPS tool. implement the Application component and the
implementation of database component is using SQL.
III. BUILDING OF THE EXPERT SYSTEM
During the course of an expert system development,
knowledge base development is the most important part.
Knowledge base defines the quality of an expert system. In
our system knowledge Base development is taken care with
the help of domain specific expert with JESS. In order to
build a comprehensive expert system, the following four
steps were recommended. Fig.1 Framework for Rice Plant Disease Diagnosis Expert System
Identify the problem and detect if there is a need
The expert system for rice plant disease diagnosis is
for the application. Determine if expert systems are
developed using Netbeans 7.0; JESS (Java Expert System
a viable solution to the problem.
Shell) for the rule/knowledge base and SQL was used as the
Select an expert system development tool whether
Database engine. The JESS file is called in the Netbeans
programming language, an environment or a shell
environment and the Database also. When JESS rules are
according to the flexibility, power, speed of
executed in JAVA, JESS library files are also to be loaded
execution needed and the ability of the developer
into the class path of Java. In JESS two JAR files (JESS.jar,
(programmer) himself.
jsr94.jar) are present. These two jar files should be included
Create a prototype. Demonstrate the credibility and
in order to execute a Java file which has JESS commands
feasibility of the prototype.
embedded in it. Instance of JESS Rule engine will be
Refine and extend the prototype. Customize the
created in Java code and instance of JESS can then be
user interface.
reused. JESS Rule Engine is having Special API to execute
IV. KNOWLEDGE BASE MODELLING USING JESS (JAVA rules.
EXPERT SYSTEM SHELL) For creating an instance of JESS rule engine,
engine = new Rete();
In order to implement rules for modelling the knowledge engine.reset();
base, we are using the Java Expert System Shell (JESS) [5, engine.batch("agri.clp");
6, 7, 8]. JESS is a rule-based language for specifying expert Here, Rete engine = new Rete() is an instance of Rete
systems which can be tightly coupled to code written in the engine by which JESS is embedded with Java and
powerful, portable Java language. Jess is architecturally a engine.batch("agri.clp"); is to load the diagnosis rules.
production system executing a rule-based program and is a After implementing JESS in Java, the symptom data is
scripting language written entirely in Sun Microsystems’s loaded in the working memory.
Java Language. Jess language is therefore a declarative database = aDatabase;
language. A JESS document is commonly initiated in text engine.addAll(database.getSymptoms());
editor, as well as the windows platform editor, Notepad. After completing the above processes, the run command
Although it can work alone and could be run on the is given and then all the rules matching the facts will be
windows command prompt, it is usually incorporated into a fired.
Java program for functionality. A Jess file is normally saved engine.resetToMark(marker);
with a “.clp” extension in opposition to the normal “.txt” engine.run();
extension. It comprises a JAR file which links the JESS to Here, engine.resetToMark(marker); is to remove all the
the Java IDE environment. As soon as the Jess is referenced previous data and engine.run(); is to fire the rules that apply
to the code, it would run predefined instructions subject to to the given symptoms.
user’s input from the Java Interface. The JESS is typically
run and manipulated on the Java interface. Program codes

675
The developed expert system provides a simple way in rice plant diseases can be handled by a reasoning approach
putting the symptoms and predisposing facttors based on the based on its knowledge base and it is easy to be accessed by
user’s description. In addition to the abbove inputs, the the users. The results obtained from m this expert system does
picture of the particular symptoms are also integrated to get not only provide the diagnosis of rice plant diseases through
the details about the particular disease annd to provide the a rule based diagnostic approach, but b also shows the future
treatment advice from the system. This systtem allows a user prospects of the application of different sub-fields of
to browse the cases that are similar to the reequested case and artificial intelligence to various infectious disease research.
to compare the symptoms of the requesteed case with the
stored ones. Then the user will choose the m most similar case REFERENC
CES
and browse its information to emphasize his/her choices. [1] Shikhar Kr. Sarma, Kh. Robindro Siingh and Abhijeet Singh, “An
Expert System for diagnosis of diseasees in Rice Plant,” International
The system will process user’s choice andd make a suitable
Journal of Artificial Intelligence and Expert
E Syatems (IJAE) ISSN :
decision. 2180 – 124X Volume 1, Issue 2, 2010, pp.26-31.
[2] S.S. Abu-Naser, K.A.Kashkash and M. Fayyad, “Developing an
Expert System for plant disease diaagnosis,” Journal of artificial
Intelligence 1(2): 78-85, 2008, ISS SN 1994-5450 ©2008 Asian
Network for Scientific Information .
[3] Ahmed Rafea, “Expert System Applications: Agriculture,” Central
Fig.2 Main Menu of Rice Plant Disease Diagnosi s Expert System Laboratory for Agricultural Expert Systems,
S P.O.Box 100 Dokki
Giza, Egypt.
The above fig.2 shows the main menu of the rice plant [4] S.S.Patil, B.V.Dhandra, U.B.Angadi, A.G.Shankar,
A and Neena Joshi,
disease diagnostic expert system which hass the menus: File, “Web based Expert System for Diagnosis D of Micro Nutrients
Diagnosis and Help. In the File menu, wee have stored the Deficiencies in Crops”, Proceedingss of the World Congress on
items like Agent, Disease, Symptom and Exit items. The Engineering and Computer Science 2009 Vol I WCECS 2009,
October 20-22, 2009, San Francisco, USA.U
causing agents of the diseases are stored inn the agent menu;
[5] Olugbenga Oluwagbemi, Esther Adeoy ye, Segun Fatumo, “Building a
the diseases occurred in the rice plant in thheir life span are
Computer-Based Expert System for Malaria Environmental
stored in the disease menu; all the possible symptoms of the Diagnosis: An Alternative Malaria Control Strategy”, Egyptian
diseases are stored in the symptom menu aand the exit menu Computer Science Journal Vol.33 No.1 1 September 2009.
is the exit of the system; the diagnostic tool for the rice plant [6] S.S. Abu-Naser, H. El-Hissi, M. Abu-R Rass and N. El-khozundar, “An
diseases is stored in the Diagnosis menu. Inn the Help menu, Expert System for Endocrine Diagnosis and Treatment using JESS,”
Journal of artificial Intelligence 3(4): 239-251, 2010, ISSN 1994-5450
the contents and about the system is stored.
©2010 Asian Network for Scientific In nformation.
VI. CONCLUSION [7] Martin Strauss, “JESS, The Java Exp pert System Shell,” Universitat
des Saarlandes, Matrikelnummeer:2512940, Seminar “AI
The expert system here we have deeveloped in this Tools,”Wintersemester 2006/2007, Bettreuer: Michael Kipp.
research work, is a rule-based system andd contained in its [8] Charles Hunn, “Employing a Java Exp pert System Shell for Intelligent
knowledge base, some important rules on ddisease causative Support in Exploratory Activities,” Bizz Logic Solutions.
agents, symptoms and predisposing factorss which favor the [9] P. Terry, “Expert Systems: Manageer's Prospective,” Journal of
spread of diseases in rice plant. It also offerrs solution to how Management in Engineering, vol. 7, no. 1, 1991.

676

Das könnte Ihnen auch gefallen