1. Introduction: Agriculture in India is the means of livelihood of almost two thirds of the work force in the country. It has always been INDIA'S most important economic sector. From a nation dependent on food imports to feed its population, India today is not only self--sufficient in food production, but also has a substantial reserve. Agriculture and allied activities constitute the single largest contributor to the Gross Domestic Product(GDP), almost 33% of it. This increase in agricultural production has been brought about by bringing additional area under cultivation, extension of irrigation facilities, the use of improved high yielding variety of seeds, better tools and techniques evolved through agricultural research, water management, and plant protection through judicious use of fertilizers, pesticides and cropping practices. However, Indian agriculture is still facing a multitude of problems to maximize productivity. Due to several reasons, the majority of the farming community is not getting upper bound yield despite successful research on new agricultural practices like inventing new crop varieties, crop cultivation and pest control techniques. One of the reasons is the present Agricultural Extension Services employed is limited. The information dissemination traditionally follows a push model which fails to share information where and when needed, or are constrained by time slots, etc (committing human experts). Farmers, who do not have direct access to scientific knowledge about farming often relies on peers for the same and hence, may get incomplete and/or distorted information. Furthermore, the complexity of a whole farming process is growing because it is constrained by many factors such as requirements, goals, regulations, etc that farmer must satisfy or consider[Jose Lopez-Collado, 1999]. Thus, manual evaluation of all the possible combinations of factors that affects the farm planning is impractical and prone to errors. Because of these complexities involved to achieve an optimal crop plan, computer-based systems such as Advisory System is required to automate many activities like pest control, weed control, crop variety selection, crop rotation, irrigation scheduling, etc. in a planning process. An Advisory System for farmers provides expert advices to farmers on many activities in a farming process. With this system, farmers can access virtual agricultural experts as and when needed.
On the other hand, developing such an advisory system for farmers is not so easy. The applicability of such a system across different regions may not be possible. The farming process and techniques may be varied from region to region. Depending on the geographical and atmospheric difference, the type of crop to be planted may be varied among regions. In addition, some farmers may prefer to their local varieties of crop. Thus developing such an advisory system for farmers needs deep knowledge of the agriculture domain and huge knowledge acquisition from mainly experts and farmers. Example, Cha-hao(black rice), a special paddy type, is a local rice variety grown in Manipur. The technique and farming process for Cha-hao is different from other paddy variety. These farming process and techniques might not be aware by experts and farmers living in other parts of India.
2. ES and CBR for Advisory Systems
Expert systems(ES) are defined as an intelligent computer program that uses knowledge and inference procedures to solve problems those are difficult enough to require significant human expertise for their solution. Most ES are rule-based which works as a production system in which rules encode expert knowledge. This type of ES is known as Rule-based Expert System[KR Chowdhary]. On the other hand, Case-Based Reasoning(CBR) is a popular AI technique for problem solving which relies on solving problems based on past solutions, just as humans use experience to solve...