An Expert System for Agricultural Research

Abstract

Now-a-days, expert system is widely used in agriculture exclusively for diagnosing and managing various agriculture practices vacillating from those concerned with livestock all the way to crop planting and associated diseases and a well as the physical monitoring and retention of capital resources including knowledge. These issues are mainly dependent upon human experts for their diagnosis and getting recovery.

The involved human experts are very scarce, inconsistent in their day-to-day decisions, unable to comprehend large amounts of data quickly, unable to retain large amounts of data in memory, subject to deliberate or inadvertent bias in their actions and can deliberately avoid decision responsibilities. Human experts are not always available whereas the computer based expert system can be used anywhere, any time.

Expert system offers an environment where the good capabilities of humans and the power of computers can be incorporated to overcome many of the limitations. Expert system increases the probability, frequency and consistency of making good decisions, additive effect of knowledge of many domain experts, facilitates real-time, low-cost expert-level decisions by the non-expert, enhance the utilization of most of the available data and free the mind and time of the human expert to enable him or her to concentrate on more creative activities.

Under these backgrounds, expert system has been developed in various agricultural crops like rice, wheat, tomato, rapeseed and mustard, mango as well as livestock management etc. in order to diagnose various pests and taking management decisions for the benefit of farmers.

Background

Agriculture is the single largest basis of livelihood of about 70 per cent of the population around the globe. It is also the mainstay of the Uganda’s state’s economy. The state can be forced into two physiographic zones which are valley and hills.

It is mainly available from the knowledge and high experience of human experts. Handling uncertain information related to certain specialty, agriculture domain needs an expert system. When an expert in a specific area gives an advice to a less experienced person, they use actual knowledge and experience to generate the piece of information. Therefore, the piece of information results in the combination of different domains and methods. It becomes more precious if the result starts solving problem where different specialists participate to solve it.

This information is either found in written or unwritten forms or generated from data available in the form of audio-video and/or in the form of local and traditional knowledge. In fresh times, technologies and applications of information have emerged as a imaginative measure for up gradation of the entire agricultural fields, ranging from scientific studies to farmers help. Incorporation of expert system as a powerful tool for the stakeholders of agricultural invention has extensive likelihood. Hence, Expert System (ES) technology can play a very important role in generating information from knowledge of human expert. So, Expert Systems can be defined as a tool for information generation from knowledge (Kalita, Sarma et al. 2016).

Expert System (ES) is constructed by obtaining the knowledge from a human expert and coding it into a form that a computer may apply to similar problem. Such types of systems are different from traditional software in the sense that they have dealt with a symbolic knowledge base, not a database or mathematical models. They can be, however, integrated with traditional software to make a complete system if needed. The background of the system begins with the gathering of the rice plant disease symptoms appearing during their life span from agriculture experts, plant pathologists and literature.

Then the acquired knowledge is represented to develop expert system using the Java Expert System Shell (JESS). Keeping the above points in mind, main concern of the research study is to design and develop an expert system providing the control measures of identified diseases. Agriculture is the backbone of Uganda’s economy as it sustains over 60% of the economy and contributes over 50% of the National GDP.

This has led to many youth engaging in the profession so to contribute to the National development through research and innovation (Maçaneiro, da Cunha et al. 2013). National Agricultural Research Organisation (NARO) is the Apex body coordinating Agricultural Research in Uganda with a mandate, “To generate efficient and cost effective and demand driven technologies”.

Currently, NARO has over 250 scientist and 210 technicians supporting the lead scientist and overseeing their work. NARO runs over 60 projects in all fields ( crops, cereals, animals, birds, fisheries, forestry, pastures, machinery, soils, minerals, invasive species and apiary to mention but a few) each having a project leader supervising the a team of a minimum of five staff including research assistants, technicians and support staff.

Problem Area

The productivity of agriculture (crop husbandry and animal husbandry) as the whole industry does not only vary between one countries and another, but also within the same country based on the different agro-ecological zones and production systems used. The gap between the farmers’ yields and those obtained by research stations is still large, even though some reduction has been reported recently. This indicates the various limiting factors affecting the industry productivity and production, ranging from land development, diseases, production and marketing.

The entire north western region experiences wide variation in soil and climatic conditions. Problems of crop production vary with zones as the soil and climate vary from zone to zone Majority of rural masses and the mainstay of Uganda’s state Economy of western ecological zone, agriculture is the largest source of livelihood. The estimates of State Domestic Product (SDP) fluctuate from year to year according to the success or failure of crop production. An agricultural expert’s opinion is needed to find out the exact type of disease. However, the agricultural experts are limited in number and there are too many problems to be solved at the same time.

These conditions make the requirement of building a system with a capability as an expert. This system must hold the expertise knowledge of the diseases and symptom of plants as agricultural experts has to have. It is, therefore, predictable that the farmers can access the expert system and station staff everywhere at any time to overcome the problem to diagnose diseases/problems and management practices.

It aggregates a number of individual expert systems in agriculture industry design just for a particular knowledge area like Crop Management Advisors and Livestock Management Advisors designed for their crop management and livestock management practices respectively

Problem Statement

The agriculture specialists (project investigators-PIs) and raw experiences are the common sources to provide information that the different stakeholders require for decision making to improve agricultural production. Agricultural specialists’ assistance is not always available when the need arises by the farmers and other junior researchers for their help.

The accurate and real-time identification of livestock and plant diseases/symptoms, associated practices is a problem that requires high degree human expertise, which is costly, and not always available. Therefore, such people are normally few and they often soon retire while their expertise are not yet documented and made widely available to others who may not be experts themselves.

Hence, the problem is to capture human expertise in agriculture practices and make it readily available to others to perform expert work in a standardized high caliber manner.

Objective of the Study

1. The objectives of the research study are as follows;
2. To investigate expert systems applications, especially in agriculture.
3. To identify the common management practices in planting crops and livestock rearing, diseases and their causing agents, predisposing factors of the disease and its control measures.
4. To develop a database that consists of the information about management practices in planting crops and livestock rearing, diseases and their causing agents, predisposing factors of the disease and its control measures as well as resource tracking over time
5. To develop expert system using the PHP and Java Expert System shell (JESS) that can diagnose the diseases associated with crops and livestock provide control measures to control it.
6. To implement the developed prototype expert system Agriculture best practices

The study will cover a period of 4 months concentrating on short term projects. Subject: The study will focus on designing and developing a knowledge based information system for national agriculture research organisation-Uganda Geographical scope: The study will focus on the Western Ecological zone of Uganda covering the Districts of Kyenjojo, Kyegegwa, Kabarole, Bundibudho, Kasese and greater Kibaale.

The knowledge expert system shall benefit a number of people in the Agricultural production chain but mainly directly benefit NARO in a sense of priority setting and resource allocation, the PIs who will be able to multi task instead of being tied down to one project at a time but most of all benefit the farmers both the demo farmers who work directly under the experts during trial and finally the indigenous farmer who taps the knowledge from the Demo farmer for uptake. Timely identification and controlling of the crop and livestock diseases will improved the production in agriculture area by using expert system. The knowledge base of the system is developed using the knowledge from deferent human experts and textbook references for the expertise to be preserved for use in case the experts soon retire.

Expert Systems

Expert systems aims to generate solutions to specific problems through the analysis of a series of facts and rules produced by people with a degree of expertise within an application area.

In the first approach considered in this area, carried out (Mansingh, Reichgelt et al. 2007), an expert system to manage pests and diseases of coffee in a developing country is presented. This system considers a knowledge base, consisting of rules and facts created from the knowledge of experts in coffee pests and diseases, an inference engine and a module for explanation. Similar to the previous work mentioned, in (Lasso, Thamada et al. 2017)an expert system to assist in the diagnosis of coffee diseases is built, based on the analysis of plants exploring carried out by the farmer.

The system structure is based on fuzzy logic techniques and decision trees, used to represent a number of conditions present given the existence of any disease, which are defined by experts. In (Dewanto and Lukas 2014), the system proposed makes use of condition-ending type rules (IF-THEN), builds from expert knowledge. Besides, user can enter different parameters of crops and these variables are used by the generated rules as data source to perform the inference process. Lastly, in (Olesen, Trnka et al. 2011)and (Chevalier, Hoogenboom et al. 2012)expert and decision support systems are developed, which consider climate information obtained from meteorological sensors and knowledge of people with some experience in the domain where they are applied. First, in (V. Rossi et al, 2010) the expert system works in parallel with decision support system. On detection of a risk situation, the expert system is invoked to help the user to take action against the problem. Second, in (R. F. Chevalier et al, 2012) the expert system becomes a guide for users according to the weather forecast obtained by monitoring different meteorological variables.

Spiral Model for ES Development Key

RP= Research prototype LP = Laboratory Prototype FP = Field prototype PV – Production Version Figure 1: Spiral model for ES development.

Expert Systems in Agriculture

Agriculture is as a complex and semi-structured system. Due to its complexity, it emerges as one of the potential subject areas of Expert System (Kulak 2005). Increased demand of farm productivity and depleted natural resources made the agricultural support system very important interdisciplinary research topic in recent past. There are many different levels of expertise and complexity found in expert systems for agriculture. One can considered all these system under the umbrella of agricultural support systems. These systems encompass computer based solution to manage one or more spatial and temporal variability aspect associated with agricultural system.

Its aim is to improve productivity and profitability of the agricultural system in presence of different variability(Robert 2002), (Naiqian Zhang, 2002). It also helps to conserve the natural resources by their optimum usage. Thus the purpose of such system is to make the overall agricultural system sustainable. In research publication such systems broadly categorized as Decision Support System (DSS), Expert System (ES), and Knowledge based (or Intelligent) DSS, and Web based DSS. The detailed taxonomy and classification are presented in (Manos, Bournaris et al. 2004).

About “management oriented farming” and “the systems approach” as well as a variety of research that recommends “optimal” management under a variety of circumstances. The practical problem with this philosophy and with this available research knowledge is that there is often too much information (i.e. “information overload”). This problem often leads to decisions which produce less than optimal results because of data complexity and time constraints.

As (Beiranvand 2017) point out “agricultural problems are often multi-disciplinary and very complex in nature”. ES in agriculture, therefore, offer great promise because both currently available and future knowledge can be organized within a knowledgebase, as described above, and used within the problem solving process. This encoding of knowledge can then allow for implementation of the “management oriented farming” philosophy by producing decision support systems which leverage management’s input into agricultural production systems. ES in agriculture help people to make complex decisions about agricultural resource systems more effectively and more timely.

Without this technology, many people, often without the desired level of experience or expertise, are forced to make decisions using incomplete information. By helping people to consider all of the relevant information and by assimilating this information into an understandable format, ES assist people in the making of environmentally sound and economically viable farm management decisions. ES technology, therefore, offer the potential to provide a necessary link between both research information and human expertise, and the practical implementation of this knowledge.

Agricultural Expert Systems – Application Areas

After conducting an overview of the literature regarding ES in agriculture, it became apparent that these systems could be categorized into eight general groups.

These groups are as follows:-

• Crop Management Advisors
• Livestock Management Advisors
• Planning Systems
• Pest Management Systems
• Diagnostic Systems
• Conservation/Engineering Systems
• Process control Systems
• Marketing Advisory Systems

Within each of these groups, numerous examples of working systems, both in progress and planned, could be found. This paper will discuss each of these general categories and highlight a few examples of working systems in each.

Crop Management Advisors

This category of ES includes advisory systems that emphasize the management of specific crops. These systems generally attempt to provide a complete and integrated decision support approach that includes most aspects of growing the crop. There are also crop management advisory systems that focus on specific management issues common to most cropping systems and can therefore be used on a wide range of crops within specific geographic regions.

The following ES deals with the growing of a specific crop: GRAPES: An ES for viticulture in Pennsylvania. This ES was developed at Pennsylvania State University in association with Texas A&M University to address the advisory needs of grape growers. The primary motivation for the development of this system was the scarcity of local expertise available to grape growers. Because of the numerous problems associated with viticulture, it was determined that growers should be supplied with a decision support tool which supplied timely, accurate and accessible information.

This system provides grape growers with recommendations regarding pest management (insect, disease and weed control), fertilization, pruning and site selection. The development environment included the rule based shell called Rulemaster on the Macintosh microcomputer (Ahumada and Villalobos 2009). This next ES deals with a specific management issue: An ES to generate fertilizer recommendations. This ES was developed at the University of Manitoba and acts as a fertilizer selection advisor. Its objective is to enable farmers to obtain the best return on their fertilizer investment.

This system uses an extensive explanation facility which allows users to better understand the underlying assumptions, facts and reasoning used to generate the recommendations. The development environment included Object Oriented ExperCommon Lisp by Expert intelligence on a Macintosh microcomputer platform. This system was subsequently converted to C++ and implemented on an IBM PC platform (Evans and Mondor. 2000)

Livestock Management Advisors

Similar to the crop management advisors, livestock management advisors can focus on integrated management advice for specific species of livestock. The following ES is an example of such an integrated advisory system: Application of conditional causality in an integrated knowledge- based system for dairy farms. This ES was cooperatively developed at Utrecht University and Erasmus University in the Netherlands. Its primary goal is to provide farmers with an integrated management advisory system including Health, Production and Financial modules for daily farm management decisions. This system is very large and complex and uses the IEEE development environment (Hogeveen et al. 2002).

Planning Systems

Planning is the key to any successful business and there is some exciting developments regarding production planning ES in agriculture. These systems deal with identifying and suggesting various management options aimed at optimizing production efficiency within a variety of constraints over a longer period of time and therefore set the future direction of the production operation.

One such system is as follows: CROPS: A whole-farm crop rotation planning system to implement sustainable agriculture. This ES was developed at The Virginia Polytechnic Institute and State University. Its primary objective is to address the growing need for whole-farm planning systems in response to the environmental, economic and legislative pressures that farmers are facing. The system aims to provide both sustainable and profitable farm cropping plans which meet desired production goals within the various constraints imposed on farming systems.

This system was developed with the Object Oriented C++ programming language (Saini, Kamal et al. 2002). 2.3.4 Pest Management Systems Of all the systems reviewed, pest management ES were by far the most common, numerous, and widespread. These systems provide farmers and researchers with integrated pest management strategies which include all relevant factors in order to adequately and cost effectively control pests. This requires that many factors such as population dynamics, weather, cost, pesticide susceptibility, and the environment be considered in order to reach optimal decisions.

One such system is as follows: POMI: An expert system for integrated pest management of apple orchards. This system was developed cooperatively at the Istituto per la Ricerca Scientifica e Technologica and the Istituto Agrario S.Michele in Italy. The system provides the apple producer with help on first classifying observations and then providing recommendations on appropriate actions. The KEE development environment was used to construct this system (Gerevini, Perini et al. 1992)

Diagnostic Systems

As with human disease diagnosis ES such as MYCIN, various diagnosis systems have been developed in agriculture using ES technology. One such system follows: An abductive reasoning expert system shell for plant disorder diagnosis: This ES was developed co-operatively between Laval University and Agriculture Quebec in the province of Quebec. It was designed as a generic tool capable of handling any appropriately structured diagnostic problem. In this sense it is a “domain specific” generic tool developed for wide applicability in diagnosing plant disorders. This system still requires work but has been used to develop a reasonably successful diagnostic tool for

Conservation/Engineering Systems

There were a number of ES encountered which dealt with engineering solutions to conservation problems. The following system was developed locally and represents a good example of just such a system: Development and validation of an expert system for soil erosion control planning in Prince Edward Island: This system was developed at the Technical University of Nova Scotia as a doctoral dissertation.

Its primary function was to conserve soil by recommending the appropriate engineering solution to control soil erosion within typical cropping systems on Prince Edward Island. It is a comprehensive system developed using the rule-based PC plus Expert System Shell by Texas Instruments (Robinson 1996) Process Control Systems A number of ES can be classified as process control type systems.

These systems are generally involved with monitoring a number of factors and then, through instrument interfaces, undertaking corrective actions. One such system is as follows: Determination of greenhouse climate setpoints by SERNSTE: This ES was developed in France at the Institut National de la Recherche Agronomique. Its function is to maintain daily climatic setpoints in a glasshouse for the winter production of tomatoes. It achieves this by monitoring a number of environmental factors and adjusting these to meet crop requirements. The system was developed using the object oriented KAPPA development tool on lBM-compatible (Martin—Clouaire, Boulard et al. 1993)

Marketing Advisory Systems

The final general category of ES in agriculture are systems that advise farmers on the marketing of specific commodities. In reviewing the literature, two such systems were encountered as brief summaries of each system without the fuIl text of the paper. These are as follows: Cattle Marketing: This ES was developed at the University of Arizona and helps farmers select marketing altematives for their cattle. These alternatives include local auctions, video sales, association sales and contracting. This system was developed using HyperCard (Crassweller, Travis et al. 1993) GMA (Grain Marketing Advisory)’: This ES was developed at Purdue University in Indiana and it helps farmers select and analyze marketing altematives for their grain.

These alternatives include the cash market, futures markets, forward contracts and basis contracts. This system was developed using Personal Consultant Figure 2: Integrated knowledge production and implementation cycle (Brian, 2012)

Advantages of Expert System

The significant advantages in the above mentioned expert systems of different crops are given below.

• The system can be used by extension personnel, researchers and farmers to identify crop diseases and enable to proceed their management.
• User can easily identify the disease on the basis of photographs of symptoms and text descriptions of disease.
• The user friendly software developed using windowing environment, thus provides enough facilities to identify the disease and to suggest the remedy conveniently.
• Provide consistent answers for repetitive decisions, processes and tasks.
• Hold and maintain significant levels of information.
• Reduce employee training costs.
• Centralize the decision making process.
• Create efficiencies and reduce the time needed to solve problems.
•  Combine multiple human expert intelligences.
• Reduce the amount of human errors.
• Review transactions that human experts may overlook.

 Limitations of Expert System

Various limitations in the Expert Systems of different crops are listed out below.

• Many farmers in the country are illiterate and knowledge of computers in rural areas is still unreached.
• It needs to be expanded and updated to accommodate new diseases and ailments of important crops in the locality.
• There is a need to include other disease diagnosis techniques such as, laboratory tests, soil test report, tissue test, plant analysis report, etc.
• The integration of nutrient deficiency module with the knowledge base needs to be included
• If the picture used in expert system is poor quality, the confusion in diagnosis of the problem will be happened and ultimately decision making will not be done properly. Therefore, the picture quality is required to be enriched.
• The complexities arising in managing rules for large knowledge base. It is difficult to write knowledge-based rule and place them in proper sequence for larger number of parameters. Verification of large numbers of rule-based system is difficult.
• Since the computer is lack of common sense, the programmer should develop the expert system in efficient way. If he or she does mistake, everything will be collapsed.
• In the developing countries, lots of farmers are not competent to English language, such expert systems need to be developed in regional languages.
• The expert systems are to be demonstrated to village area through blocks or village administration unit so that farmers can get a chance to develop their own expertise.
• Adding speech interface to the system may be proved to be more beneficial for the farmers of the remote area.
• Will not be able to give the creative responses that human experts can give in unusual circumstances.
• Lack of flexibility and ability to adapt to changing environments.
• Not being able to recognize when no answer is available.
• Knowledge acquisition remains the major bottleneck in applying expert system technology to new domains. xv. Maintenance and extension of a rule base can be difficult for a relatively large rule base (beyond 100 rules).
• Expert systems are not as compact as neural network and genetic algorithm systems. This makes them harder to embed in other systems, as the inference engine and working memory must be part of the system at run-time.

Conclusion

The farmers often rely on agricultural specialists and advisors to provide information for decision making to get rid of problems related to pests and diseases. But, due to non-availability of agricultural specialists or extension workers, the decision making process will be delayed. In such cases, the losses due to problems will be increased more within the delayed time. Therefore, the time saving and immediate decision making can be done effectively by using expert system.

The expert systems in agriculture help a lot in increasing the crop production and reducing the yield losses. The successfully developed expert systems should be demonstrated to farmers for the benefit of them. The impact studies of expert systems in different crops are required to be incorporated in due course. Expert system is computer program which can be used as virtual expert to guide the Growers.

Expert system is a technological way to deliver agricultural knowledge from books, research papers, thesis etc to actual implementation level i.e. at Growers. As a result, application of expert system in agriculture sector become popular and many nations took initiative to develop different expert system. But most of expert system concentrate on particular aspects of crop management like pest or fertilizer management etc. For E.g. in India, for Soybean, none of the expert system is available which will give guideline to Growers from soil preparation up to harvesting.

So researchers have to try to develop such an expert system which will guide to Growers to take decision into different aspects of crop management like soil preparation, seed selection, pest management, fertilizer management, weed control, irrigation management, nutrition managements well as livestock management etc.

Methodology

This is a web based expert system with java server pages (JSP/PHP) as the front end and MySQL as the backend. Tomato crop expert advisory system is aimed at a collaborative venture with eminent Agriculture Scientist and Experts in the area of Tomato Plantation with an excellent team of computer Engineers, programmers and designers. The program is divided into two aspects Information System, Advisory System in Information system, the user can get all the static information about different species, Diseases, Symptoms, chemical controls, Preventions, Pests, Virus of Tomato fruits and plants. In Advisory System, the user is having an interaction with the expert system online. The user has to answer the questions asked by the Expert System.

Depends on the response by the user the expert system decides the disease and displays its control measure of disease. This web application is expected to have the following features:

1. this web application provides time-to-time updates of tomato information to the users at their doorsteps regarding diseases, virus and its control measure, which leads to good yields.
2. This site contains four major sections named Information Systems of Tomato crop, Tomato Advisory System, other services related to web application and an additional feature is links to other agriculture systems.
3. The web directory service, articles and the discussion forum service provided in the website will help the tomato fraternity in a greater way to interact each other to produce better findings in the area of tomato field.

Proposed Framework for Agriculture Expert System

The advantage of model based approach to prepare the agricultural expert system is that it facilitates to develop generalized system for all agriculture knowledge producing best practices. The proposed modeling approach help to develop expert system for optimum knowledge retention generated by project investigators (PIs) and probability of occurring of diseases to the farmers.

User Interface Designing/Prototyping

The UI is the portion of software with which a user of referral system directly interacts. It models UI requirements; requirements evolved through analysis and design to result in the final UI for the system. It introduces technology to prototyping efforts that is most likely made a design decision about the implementation technology. The following parts shows UI prototype in developing Electronic referral management system.

1. Welcome Page This is the page that displays categories of users who can login to the NARO expert system. By putting one of the user category in mouse focus, a login screen will pop up as reflected in the next interface below
2. Login Screen This page displays when the user category proceed to the NARO expert system. It provides interfaces to accept users ‘account information which is subject to verification.
3. Successful Login Interface By taking project investigators as user category sample space, a screen below displays when the PI has successfully logged in. It provides a menu to housing a huge critical amount of information which in turn converted in to knowledge capital.
4. Administration Page This screen below shows the tasks that the farm manager/ administrator of NARO expert system as well as the information that he/she can view and alter.
5. Database Design The screen below show the high level major database management system objects like tables and indices, reflecting how physically the data is stored in database.
6. Auto Generated Data Dictionary for NARO Expert System

Actual Implementation Requirements 

User interfaces developed with P HP and HTML5/XML/CSS/C++, The user interface allows the user to interact with the system for defining the parameters and getting the results MYSQL for database development

Suitability and Feasibility of Expert Systems and Validation Rules

The problems already raised can to large extent be solved using expert systems to generate the information to the growers by using its knowledge base and reasoning mechanism acquired from human experts and other sources. The expert system generates the advice based on its knowledge base and its reasoning mechanism that are actually behind all developed extension documents, and more.

Consequently, when a user enters the data of his/her plantation to the system, the appropriate advice is generated. There are no limitations on the number of generated recommendations. Therefore expert system overcomes the problem of static information provided in extension documents. The knowledge acquisition process for building an expert system, facilitates the integration of knowledge and experiences of different specialties. For example, an agricultural diagnostic expert system requires the integration of specialists in nutrition, plant pathology, entomology, breading, and production.

Therefore, when a problem occurs, the system can help the user in identifying the cause of the problem in a much more efficient way than consulting a document that handles a specific problem. Expert systems can be integrated with other information sources such as images bases and/or textual bases to make use of these sources. For example, images can be used for describing symptoms as it is very difficult and very confusing to describe them in words. Images can also be used for confirming the diagnosis of the cause of a certain disorder. Expert systems can also be integrated with textual data bases that may be the extension documents related to the specialty and/or commodity handled by an expert system.

This textual data base can be used for explanation purposes of basic terms and operations. It can also be used to confirm the reached conclusion in some situations. The updating problem is also found in expert systems. However, the knowledge base can be maintained more efficiently than maintaining manual documents. The problem of updating the versions in the field can be eliminated in case that the expert systems are stored on a central computer and accessed through a computer network. The undocumented experience and knowledge can be acquired and stored in the knowledge base of an expert system for a certain specialty and/or commodity.

This experience can be available to all growers using the system. The feedback from the usage of the system can be used as a source of information when analyzed by researchers, the knowledge behind it can be identified, and the knowledge base can be updated continuously. Expert systems can also help in overcoming the problem of the relatively few numbers of experts relative to the demand from the growers. Expert systems technology can help to transfer the information of experts, and experienced growers to farmers through the extension system.

Validation

An approach to validation will be based upon the detection of anomalies is presented, and related to the concepts of consistency, completeness, correctness and redundancy (Rule verification and validation: structure). Automated tools for expert system verification are reviewed. Considerable attention is then given to the issues in structuring the validation process, particularly the establishment of the criteria by which the system is judged, the need to maintain objectivity, and the concept of reliability.

This will be followed by a reviewing of validation methods for validating both the components of a system and the system as a whole. Validation will be done by the primary expert who was involved in the systems knowledge base development and knowledge representation. Validation will provide the final opportunity to evaluate an expert system prior to testing by additional experts or other identified end users.

The primary purpose of validation is to have the expert concede to the development of a credible prototype which provides a reasonably accurate diagnostic ability. Although validation is an essential phase to expert system development, problems of access to expert assistance, time and resource constraints can often make validation procedures impractical or limited

Recommendations 

Mobile application development tech-Android and sq-LITE version should be part of the future work of building integrated agriculture expert system ii. The research issues such as software components with the agricultural expert systems, agricultural knowledge sharing and reuse, intelligent retrieval of agricultural data and automatic knowledge acquisition need to be addressed.

Reference

1. Ahumada, O. and J. R. Villalobos (2009). “Application of planning models in the agri-food supply chain: A review.” European journal of Operational research 196(1): 1-20.
2. Beiranvand, Z. (2017). “Integration of expert system and fuzzy theory for diagnosis wheat plant diseases.” Revista QUID 1(1): 1924-1930.
3. Chevalier, R. F., et al. (2012). “A web-based fuzzy expert system for frost warnings in horticultural crops.” Environmental modelling & software 35: 84-91.
4. Crassweller, R., et al. (1993). “The future use and development of expert system technology in horticulture.” HortTechnology 3(2): 203-205.
5. Dewanto, S. and J. Lukas (2014). Expert system for diagnosis pest and disease in fruit plants. EPJ Web of Conferences, EDP Sciences.
6. Gerevini, A., et al. (1992). “POMI: an expert system for integrated pest management of apple orchards.”
7. Kalita, H., et al. (2016). Expert system for diagnosis of diseases of rice plants: Prototype design and implementation. Automatic Control and Dynamic Optimization Techniques (ICACDOT), International Conference on, IEEE.
8. Kulak, O. (2005). “A decision support system for fuzzy multi-attribute selection of material handling equipments.” Expert systems with Applications 29(2): 310-319.
9. Lasso, E., et al. (2017). “Expert system for coffee rust detection based on supervised learning and graph pattern matching.” International Journal of Metadata, Semantics and Ontologies 12(1): 19-27.
10. Maçaneiro, M. B., et al. (2013). “Drivers of the adoption of eco-innovations in the pulp, paper, and paper products industry in Brazil.” Latin American Business Review 14(3-4): 179-208.
11. Manos, B., et al. (2004). “A decision support system approach for rivers monitoring and sustainable management.” Environmental Monitoring and Assessment 96(1-3): 85-98.
12. Mansingh, G., et al. (2007). “CPEST: An expert system for the management of pests and diseases in the Jamaican coffee industry.” Expert systems with Applications 32(1): 184-192.
13. Martin—Clouaire, R., et al. (1993). “Using empirical knowledge for the determination of climatic setpoints: an artificial intelligence approach.” The Computerized greenhouse: automatic control application in plant production: 197.
14. Olesen, J. E., et al. (2011). “Impacts and adaptation of European crop production systems to climate change.” European Journal of Agronomy 34(2): 96-112.
15. Robert, P. C. (2002). Precision agriculture: a challenge for crop nutrition management. Progress in Plant Nutrition: Plenary Lectures of the XIV International Plant Nutrition Colloquium, Springer.
16. Robinson, B. (1996). “Expert systems in agriculture and long-term research.” Canadian journal of plant science 76(4): 611-617.
17. Saini, H. S., et al. (2002). “Web based fuzzy expert system for integrated pest management in soybean.” International Journal of Information Technology 8(1): 55-74.