Managing Agriculture Knowledge Through Localized Community Expert System
Despite of new knowledge sharing web-based and Information Communication Technologies (ICTs) tools flourishing in the Egypt over the recent years, there is still a pronounced gap between research results and the knowledge seekers. Thus, agriculture and rural development research results undergone by different institutes and agencies in Egypt and the MENA region are not sufficiently echoed on the level of small-scale farmers and vulnerable rural poor. In addition, the indigenous knowledge that the rural people have inherited and practiced over time is not in the hand of the researchers and knowledge providers to be able to fortify their findings with empirical knowledge and successful practices. The Expert System is expected to overcome the gap between the availability of mass of agriculture information online (Un-managed knowledge) and the knowledge management and in agriculture sector in Egypt, create a vibrant network for exchange of ideas and experiences among farmers and academia professionals. Providing updated adequate Knowledge to farmers can lead to optimal utilization of limited resources, and maximizes the acre productivity. The project will also open the road for Knowledge economy in agriculture sector in Egypt.
Goal & Objectives
This project aims to investigate the most efficient agriculture knowledge management and utilizations solutions that can empower farmers with the expertise knowledge whenever needed in an easy optimized and updated way to improve their agriculture productivity; hence share in overcoming the problem of food security in Egypt.
- Analyzing the Expert Systems methods and tools
- Studding the following three dimensions (the existing agriculture portal www.aradian.net , the needed knowledge base of human expertise for problem solving, and the best available Experts/sources of clarifying the uncertainties)
- Designing the Expert System patterns and develop a series of technical procedures of know-how knowledge system
- Developing partnerships to transfer knowledge and ensure integration and sustainability of the System
- Raising the awareness of farmers and extension officers on how to make the best use of the developed Expert System
- Building the Capacity of partners and experts to integrate efforts and deliver high level of know how to farmers
- Defining the impact assessment factors of applying expert system on farmers
The Foundation Phase (March – December 2012)
This phase aimed at building a foundational framework for the project by studying international, regional and local expertise in building knowledge bases, identifying challenges and opportunities in the Egyptian context, ensuring pre-requisites, and producing Knowledge Bases prototypes for piloting during the remaining of the project. This phase comprised the key interventions including: Situation analysis, Developing Expert System Tool, Selection of micro topics, Selection of knowledge experts, Developing knowledge engineering training suite prototype, Building capacity in knowledge engineering, and Producing Knowledge Bases prototype.The Piloting Phase (January – September 2013)
This phase started by putting in place the prototype products which have been developed in the foundation phase for piloting the capacity building efforts among a wide group of knowledge experts. During the project piloting phase, five (5) capacity building and validation workshops were conducted raising the awareness and building the capacity of more than 60 participants. Different training approaches have been used during these capacity building workshops in order to identify the best approach for training knowledge experts on developing expert systems.
- Extension Workers
- Knowledge experts
- Agro-knowledge Scientists
This workshop was conducted in partnership with Salasel project to build the capacity of five (5) teams of knowledge experts for collectively developing their expert systems. The idea of a group expert system was originated in light of the findings from the foundational phase that highlighted the merits and drawbacks of both academics and field workers in developing expert systems. This workshop aimed at joining the efforts of both academics and field workers and getting them to work together in groups for developing group expert systems.
The purpose of this workshop was to train a group of researchers from the center of Agricultural researches not only on expert system development, but also on expert systems verification and validation methodology- an additional part that was added to the training module. Training knowledge experts on verifying and validating their expert systems was a decision taken by the project team following discussions with project partners (i.e. CLAES & Salasel) and technical consultants on testing the efficiency of the developed expert systems.
Egypt's ICT Trust Fund, affiliated with the Ministry of Communications and Information Technology (MCIT) and the United Nations Development Program (UNDP), witnessed yesterday the closure of the events of the Agricultural Knowledge Management project through developing the local expert system, which aims to support employees in the agriculture sector.
ICT Trust Fund offers this program with a grant from the International Development Research Centre (IDRC), in cooperation with KariaNet project, funded by the International Fund for Agricultural Development (IFAD). During the project, six specialized training workshops were organized. They resulted in identifying a practical model to build knowledge bases which help diagnose and solve agriculture problems. Professional and academic experts worked on building and making these bases accessible online. Moreover, the project led to learning skills of knowledge engineering, evaluating developed expert systems and training more than 50 agriculture experts among practitioners and academics. The fruitful project provided 37 agriculture expert systems, available for free on the internet and easy surfing on mobile phones.
The Agricultural Knowledge Management project via developing a local expert system started in march 2012 with the aim to make agricultural expertise available through an easy and quick system trusted by users like farmers, guiding persons and all people who are concerned with agriculture in Egypt, using advanced technology tools to achieve financial revenues and development outcomes for all participants including the agricultural expert or the farmer himself.
Situation analysis And Need assessment study:
Literature review has been conducted at the outset of the project along with a SWOT analysis in order to identify the key components of building Knowledge Bases through experts systems and the pre-requisites for such a process. The review identified the main causes behind the lack of a workable model for building efficient Knowledge Bases. The causes are outlined under four main headings; causes related to Tools and Technologies, Domain Experts, Knowledge Acquisition, and System Usage.
Developing Expert System Tool:
The process of selecting and adapting the Expert System Tool started with an assessment of the elements of human interface that should exist in such a tool; namely, intuition, user-friendliness, easy-to-learn, visual interface, efficiency and reliability. In light of these elements, a tool has been selected from the available outsource technology for expert systems. The project adopted a work approach that employed different rounds of assessment for the Expert System Tool; assessment by an Expert System Technical Advisor, a number of knowledge engineers at CLAES (Central Laboratories for Expert Systems) and by the knowledge experts who used the tool to develop their knowledge bases.
In developing expert systems, the need becomes to respond to the interest of the growers; selecting micro topics that respond to problems facing the growers, structuring and ordering questions that could be easily answered by the growers, using clear questions and language familiar to the growers in describing the symptoms of the disorders, including all possible questions and answers that growers could possibly observe and describe, reviewing and updating the expert systems for better ways to respond to growers' needs for diagnosing plants' disorders. In a nut shell, each step in developing, reviewing and updating expert systems should be for more convenient use of expert systems by the end users; the growers. This is what this research project has identified as the ultimate criteria for efficient expert systems.
Developing Knowledge Engineering Training Suite Prototype
During the foundational phase, a partnership has been established with CLAES (Central Laboratories for Expert Systems), part of the Agri-expert systems network of the Ministry of Agriculture. The main tasks of CLAES in this research project were to develop knowledge engineering training package, build the capacity of knowledge experts in knowledge engineering, and assist knowledge experts in developing valid and verified expert systems. A training package was developed that includes presentations and training material on related topics such as expert system life cycle, ultimate users' needs, knowledge elicitation, knowledge analysis, knowledge engineering, expert system development and expert system validation and verification. The training material was piloted during the foundational phase with the two selected knowledge experts. The training approach and material were adopted accordingly for the following rounds of training.
Experimental and Validation workshops
During the project piloting phase, five (5) validation workshops were conducted raising the awareness and building the capacity of more than 60 participants. Different training approaches have been used during these capacity building workshops in order to identify the best approach for training knowledge experts on developing expert systems.
Building Capacity on Knowledge Engineering
Building the capacity of knowledge experts in knowledge engineering is at the heart of the workable model. Previous attempts for building expert systems relied on knowledge engineers for knowledge acquisition, analysis, encoding and decoding knowledge on the expert system tool which proved difficult for the knowledge engineers to build several expert systems in different knowledge domains and in updating the developed expert systems. This research project puts a high weight on the knowledge experts to build their own expert systems in light of the fact that it is more effective to train knowledge experts on knowledge engineering than training knowledge engineers in a specialized knowledge domain. Given the significance of this component in the workable model that has been elicited during this research project, a training module has been designed and piloted with six groups of knowledge experts during the foundational and piloting phases of the research project. The training module includes training sessions on expert system life cycle, ultimate users' needs, knowledge elicitation, knowledge analysis, knowledge engineering, expert system development, expert system validation and verification.
Project Collective Results
- 6 Capacity Building and Validation Workshops
- 50+ participants
- 37 Expert Systems developed
- 4 Focus group meetings with knowledge experts
- 1 Focus group meeting with CLAES
- Tool Questionnaire
- Training Questionnaire
- Peer Review Questionnaire
- Expert System Rating Scale
- Training Package
- ICT Skills KIT
- Process reports
- Process/Working Model document
- On-going feedback collection mechanism
22 March 2013
27 March 2017
10 September 2013
24 December 2013