International Conference on Information and Communication Technology and Its Applications (ICTA 2016) Federal University of Technology, Minna, Nigeria November 28 – 30, 2016 A Multilingual Translation System for Enhancing Agricultural e-Extension Services Delivery Muhammad Bashir Abdullahi1, Ibrahim Shehi Shehu2, and Yahaya Mohammed Sani3 Department of Computer Science, Federal University of Technology, Minna, Nigeria {1el.bashir02, 2ibrahim.shehu} @futminna.edu.ng, 3saniyahaya84@yahoo.com Abstract—Agricultural extension is the application of new productivity [1]. More so, the various roles and contributions knowledge and scientific research findings to agricultural that policy makers and extension workers continue to make practices through farmer education. As a result, agricultural in the current food production cannot be down played. extension agents or workers are people from government An effective agricultural extension depends on how fast research institutes who educate or pass on information to and useful the extension services reach or meet the farmer’s farmers on how to use the new knowledge and scientific information need. Thus, it is likely that the farmers benefit research findings. However, the conventional method of from agricultural research findings with the eventual goal of communicating the agricultural research outputs or findings to improving agricultural productivity. farmers through only face-to-face meeting has many challenges Unfortunately, in most developing countries today, such as geographic dispersion between farmers and extension workers, poor communication capacity, poor transportation extension agents still depend largely on traditional extension facilities, bad roads, inadequate funding and dialectical approaches of transmitting agricultural information to problems, which create great problems to effective farmers. The traditional approaches of communication are communication of agricultural information to farmers. classified as one-way multipurpose and two-way Furthermore, this mode cannot adequately handle urgent (time multipurpose sources. The one-way multipurpose sources bound) information that should circulate within the farming include: television, radio, public campaign, leaflet, pamphlet, populace. In this paper, a multilingual translation system was newspapers and magazines. While the two-way multipurpose developed to enhance agricultural e-extension services communication sources include: village fairs, field delivery. The system employs a serial integration of rule-based demonstrations, trainings and study tours. Also included in and statistical machine translation techniques to translate this category are: extension workers, private agencies, para- agricultural information or scientific research findings from technicians, farm input dealers, non-governmental the extension workers in English (source language) into organizations (NGOs), credit agencies, fellow progressive farmer’s registered native or preferred language. Four (4) farmers, output buyers/food processors and primary target languages were considered, which include Arabic, cooperative societies [2],[3],[4]. Hausa, Ibo and Yoruba. The system was implemented using These sources of transmitting agricultural information to Per Hypertext Processor (PHP) language version 5.3.5 farmers are no longer effectual for urgent agricultural and Structured Query Language (MySQL) version 5.0.2. research findings needed by farmers. However, the rising The system integration test shows 65% accuracy in face up to farmers in this new millennium is how to manage translating research outputs in English to farmer’s with the information explosion and global trend in agro- registered native language. It is therefore recommended technology. There is the need therefore, for inter-related that the implemented system be adopted for its efficiency and systems to diffuse information and technological innovations accessibility in enhancing agricultural e-extension services to farming populace in developing countries [5]. delivery. In addition to the identified problems, farmers are even more confronted with myriads of challenges now than ever Keywords-language translation; agriculture; e-extension before in relation to information generation and services; farmers; information system communication, which is as a result of lack of sufficient agricultural knowledge and information that will enhance I. INTRODUCTION farmer’s participation, collaboration, and integration in Extension agents or workers are people who possess an agricultural decision making processes. Also, lack of acceptable level of edification, engaged by government to investment from other agricultural stakeholders, for example, bridge the fissure connecting the government and the farmers government agro-allied industries, and non-governmental in terms of agricultural services. These groups of people organizations have lead to low output in both production and educate or pass on information to farmers on how to use sustainability. It is also arguable that farmers’ illiteracy in information derived from government research institutes. Information and Communication Technology (ICT) is so Such information empowers farmers to take control over high in the rural areas as most of the farmers lack access to decision-making processes and resources for increased agricultural technology, innovations and education. There 62 International Conference on Information and Communication Technology and Its Applications (ICTA 2016) exists also (i) dialectical disparity in which the research 2) Statistical Machine Translation (SMT) Technique: findings are communicated, (ii) near absence of training and The SMT is a corpus based approach, where translation is re-training of both the extension workers and farmers on new generated on the basis of statistical models whose innovations [6], and (iii) ineffective ICT policies that are parameters are derived from the analysis of bilingual text targeted towards empowering farmers via the deployment of corpora [9]. A massive parallel corpus is required for ICT tools and even where they are available; there is poor access and reception. training the SMT systems. The SMT systems are built based Finally, the present understanding of Agricultural on two probabilistic models: language model and translation extension is supporting people engaged in agricultural model [9]. The merit of SMT system is that linguistic production by facilitating, empowering and linking them to knowledge is not a requisite for building the system. The markets and other players in the agricultural value chain; to complexity in SMT system is creating massive parallel obtain information, skills and technologies to solve their corpus. problems [7]. 3) Hybrid Machine Translation (HMT) Technique: The remainder of this paper is organized as follows: HMT was built owing to the drawbacks of the two section 2 presents the related work. Section 3 describes the approaches and their prospect to be integrated [9]. Statistical research methodology used. The results and discussion of the and Rule-Based are two MT techniques, whose methods of system testing and evaluation were presented in section 4. Section 5 gives the concluding remarks. translation are orthogonal to one another. SMT do not need to learn about the language at all, but RBMT is based on gathering language rules. Due to this difference, integrating or hybridizing SMT and RBMT gives a better performance. II. RELATED WORK The hybrid technique can be used in a number of different ways. In some cases, translations are performed in the first A. Machine Translation stage using a rule-based approach followed by adjusting or Machine Translation (MT) is one of the most essential correcting the output using statistical approach. In the other applications of computational linguistics that uses the way, rules are used to pre-process the input data as well as computer software or web application to translate text from post-process the statistical output of a statistical-based one language to another. One of the benefits of machine translation. This technique is better than the previous two translation is that it helps people to understand an unknown and has more power, flexibility, and control in translation language without the aid of a human translator. However, [9],[10]. MT is often perceived as low quality based on outdated perception created by its use of older translation technologies C. Review of Existing E-Extension Services Delivery or freely available generic translation tools from Google or Systems Bing that have not been customized for a specific purpose Kalna-Dubinyuk [11] developed an electronic extension [8]. Many technology advances have been made in recent service in Ukraine using extension service model that links years that are changing this perception with customized Ukraine’s extension service system and the outside world. machine translation engines [8]. This system has developed the market economy in Ukraine and moves forward in increasing the number of farmers and B. Machine Language Translation Techniques the formation of new forms of agricultural business entities. The major constraint of this system is that it only capitalizes A few different types of machine translation are available on the market economy of agricultural produce and ignores in the market today. According to [9] the most widely used agro-technology innovation and adaptation. techniques include: Statistical Machine Translation (SMT), A mobile-based Agricultural Extension System was Rule-Based Machine Translation (RBMT), and Hybrid developed by [12] in Tanzania called M-FAIS (that is Machine Translation, which combine RBMT and SMT. Mobile Phone Farmers Advisory Information System). The These techniques are briefly explained as follows [9]: system was designed using a GSM modem and Independent 1) Rule-Based Machine Translation (RBMT) Technique: Service Architecture (ISA). The research finding shows that The RBMT relies on countless built-in linguistic rules and the M-FAIS allows farmers to get advice in various millions of bilingual dictionaries for each language pair. The agricultural issues such as agronomic practices, livestock RBMT system parses text and creates a transitional husbandry, post-harvest operations, veterinary services, representation from which the text in the target language is forestry, financial and market support services. The system generated. This process requires extensive lexicons with depends on third party software (Serial Splitter) for sending morphological, syntactic, and semantic information, and and receiving short messaging service (SMS) operations. This makes its use on public domain vulnerable to malicious large sets of rules. The software uses these complex rule sets attack. and then transfers the grammatical structure of the source An implementation of e-extension system that uses language into the target language. [9]. There are no human mobile phones for knowledge sharing was done by [13]. It interventions during the conversion from one language to uses social network tools to share quick and instant another language. Human intervention only takes place, if at information. It maximizes the use of information and all, after translation to manually correct errors in the communication technology to attain a modernize aquaculture machine translation output. sector. It also focuses on creating an electronic and 63 International Conference on Information and Communication Technology and Its Applications (ICTA 2016) interactive bridge where farmers, fishers, and other editing, and statistical error checking. These modules are stakeholders rally and transact to enhance productivity, explained as follows: profitability and global competitiveness. The shortcoming of 1) Start: This is the beginning of the process. this system is that not all farmers have internet access and 2) Input Text: When a source text/sentence (in English ICT literacy to operate in the social media. language) of Agricultural Information (AI) is entered as An ICT-based agricultural extension service delivery input, the following process ensued: system was proposed for Nigeria by [14]. The system is to be 3) Deforming and Pre-Editing: This is a preprocessing used in agriculture center(s) in village(s) where extension service is required. The purpose is to stimulate farmers module. In this module, deforming is performed as a process driven extension; by allowing farmers to request for in which the machine checks the part of the source AI guidance and assistance based on their unique needs. text/sentence that does not require translation such as However, it does not adopt the use of phone-based pictures, figures, diagrams and identifies only the portion of application and not all farmers have access to internet in the the source text/sentence that can be translated. Similarly, rural communities, even where the facilities are available. pre-editing involves fixing up the punctuation marks that E-sagu was implemented by [15]. It is a web-based does not require translation. This is to make the machine agricultural expert advice dissemination system, which language translation of the AI easier, faster and efficient. farmers can use to send a digital photograph of their 4) Analysis: In this module, the source text of AI is problems to an agricultural expert. The role of extension analyzed based on the linguistic information provided to agent is excluded on the system since the agricultural expert has direct link with the farmers and knows about their produce a complete parsing of a source language sentence. problems. The shortcoming of this system is however that of Thus, it comprises of two components: tagger and parser. dialectical problem. a) Tagger: This component identifies the linguistic Shrikant and Shinde [16] developed a web-based property of individual word of AI in the source text through information and advisory system for agriculture using the following processes: software engineering’s classic life cycle method. Classic life (i) Morphological Analysis: This aspect cycle is also called linear sequential model and it is a widely determines the form of AI word such as used paradigm for system development. The system provides number, tense or part of speech (POS) tagger. farmers with relevant and updated crop information. The (ii) Syntactic Analysis: This determines whether information it provides is restricted to crops. Information AI words are subject verb or object. regarding Livestock, market and weather are not provided by this system. In addition, not all farmers can have access to b) Parser: This component breaks AI words/sentence web-based facilities, since in many rural communities, where into smaller elements, according to a set of linguistic rules majority of the farmers reside, have no internet facilities. that describe its structure through semantic and contextual E-agriculture framework was developed by [17]. The analysis which determines the proper interpretation of AI framework proposes an implementation of an e-farming text/sentence from the result produced by syntactic analysis. system that can be used in aiding sustainable agricultural This is achieved by using lexical and semantic analyzer farming practices. The incorporation of IT into farming created by parser. involves the integration of diverse technologies, with each 5) Transfer: In this module, the syntactic/semantic capable of positively impacting the efficiency of farming structure of the AI source text is then moved in to the activities, thereby, promoting sustainability in agricultural syntactic/semantic structure of the AI target languages. practices. This framework has overcome farmers’ literacy level problems since the framework proposed is meant to 6) Generation: In this module, lexical transfer (the compliment and replace the traditional extension services mapping of a source-language lexical item with an delivery. The major limitation of this framework is that equivalent target-language item) occurs and mapping language translation from source language of the agricultural dictionary entries into appropriate inflected forms to yield a information to the target local language understood by the target-language equivalent term. This is achieved using rural farmers is not taken in to account. Arabic lexicon, Hausa lexicon, Ibo lexicon and Yoruba lexicon to ensure proper interpretation. 7) Reforming and Post-Editing: This is a post- processing module. In this module, once the AI text is III. RESEARCH METHODOLOGY translated, the target text is reformed after post-editing. This involves re-incorporation of non-translated portion of the A. Multilingual Translation System source AI to target text for quality and adequate target AI to The proposed multilingual translation system is based on be disseminated to famers. the hybrid machine translation technique, which is a serial 8) Statistical Error Checking: In this module, to ensure integration of the rule-based and statistical machine accurate grammatical matching of the target output translation techniques. It is designed to translate source text from English language into any of the four preferred target produced by the rule-based approach into its statistical languages: Arabic, Hausa, Ibo and Yoruba. The operational approach equivalent. Translation error checking is done to process of the multilingual translation system, as shown in ensure good quality translation of the target texts of the AI. Fig 1, is divided into six modules: deforming and pre- But in a situation where the rule-based translation technique editing, analysis, transfer, generation, reforming and post- does not correspond with its statistical machine translation 64 International Conference on Information and Communication Technology and Its Applications (ICTA 2016) equivalent, a decision is taken to consider another set of with the agricultural extension agent, receive/read linguistic rules starting through the analysis, transfer down agricultural information from their mobile phones and also to the generation stage using the rule-based technique until send their comment/query/request to AEA. the target text matches that of statistical based target text. 9) Output Text: The output target texts/sentences are the Start proper and accurate equivalent translation of the source AI text (English) in to target AI texts (that is Arabic, Hausa, Ibo Input text or Yoruba) meant to reach the farmers according to their (that is English) registered target languages with the AEA. 10) Stop: This mark the end of the process. Deforming and Pre-editing B. System Framework Design Analysis The proposed framework for a multilingual translation system to enhance agricultural e-extension services delivery Tagger - Morphological Analysis is shown in Fig. 2. The system connects three major -Syntactic Analysis stakeholders of an agricultural extension services namely the farmer, researcher/expert and agricultural extension agent. Parser Semantic and The role of each stakeholder is explained as follows: Contextual Analysis 1) Researcher/Expert: This stakeholder provides critical research output in response to specific needs of the farmers. Transfer: The research output cut across different aspect of farming Re-Ordering Syntactic and Semantic of source text including crop farming, livestock farming and the rest. Whenever a research request get to the research institute, it Generation Syntactic and Semantic is handled by an expert in the area requested who after his generation of target texts using findings relays a feedback through extension agents to target languages lexicons farmers. In this system, the researcher/expert login to the Arabic lexicon system, read information/research request and communicate research request and finding to the AEA. The medium of Hausa lexicon communication between the researcher and AEA is through Yes Ibo lexicon the web and their language of communication is English. 2) Agricultural Extension Agent (AEA): This is a trained Yoruba lexicon expert who serves as a link between the farmers, research institute and farm input firms. They convey information responsive to the requirement of any component in the Reforming and post-editing system. The AEA usually pay regular visit to farmers, interact with them in order to know their problems and concerns and then send it to an expert requesting for a Statistical error checking using corpus to check for language translation conformity solution. The functions of the AEA in this system includes with the result obtained by rule-based approach login, update of information, register and manage farmers, send updates of agro information via SMS to farmers and Arabic Corpus also send research request to researcher/expert. In addition, Hausa Corpus the AEA interact with researcher through the web-based system and send agro information to farmers from the Ibo Corpus system to their mobile phones (it could be ordinary phones Yoruba Corpus or android phones). 3) Farmers: This is the consumer of agricultural information. This is the component that receives agricultural Error information from research institutes and farm input firms via agricultural extension agent relevant to their farming No requirement(s). This is because timely delivery of Output target agricultural information to farmers means empowerment to texts them. The multilingual translation system for enhancing agricultural e-extension services delivery is aimed to Stop provide farmers with this timely information they required based on their registered local languages (Arabic, Hausa, Figure 1. Flowchart for Multilingual Translation System Ibo or Yoruba). The role of the farmers includes: registering 65 International Conference on Information and Communication Technology and Its Applications (ICTA 2016) Farmers · Register with the AEA · Request for agricultural information · Receive/Read agricultural information Multilingual Translation · Login Technique System Environment · Read agricultural research request · Update agricultural research request or finding to the AEA Web Interface Web server database Expert/Researcher · Login · Update personal profile · Register/manage farmers · Post updates on agricultural research information via SMS to farmers · Send agricultural research request to experts Agricultural Extension Agent (AEA) Figure 2. Proposed system framework MySQL database management software that store C. System Framework Representation collection of information and organized them so that it A three (3) tier model was adopted for the proposed can easily be accessed, managed, and updated. MySQL multilingual translation system which consists of front-end command is used to insert, update, fetch and delete data in (presentation tier), middle tier and back-end (data tier) as this system database. explained below: 1) Presentation Tier: This is also known as the front IV. SYSTEM TESTING AND EVALUATION end and at this level, information is presented to client (i.e. Researcher or AEA) via browsers. This tier was A. System Integration Analysis developed using Per Hypertext Processor (PHP) language To ascertain the workability between unit functions of version 5.3.5. the implemented system (interoperability of function 2) Middle Tier: This tier is also known as server side. between components), system integration testing was It is used for processing request through the My Structure carried out. Four (4) test cases were tested hundred (100) Query Language (MySQL). MySQL 5.0.2 was used. This times each. The test result shows how each component is also where the language translation takes place. responded to an event that identifies specific functions of 3) Data Tier: This tier is also known as back-end. the design to whether or not the responses are as expected. This is the data center for the system which uses the Integration testing analysis is shown in Table I. 66 International Conference on Information and Communication Technology and Its Applications (ICTA 2016) Test case 01 shows that 80.3% SMS from AEA were Some questions were directed to the selected farmers, sent and received while 19.7% were not. Test case 02 extension agents and experts and their responses were shows that 74.6% of the web messages from AEA to collected instantly. Simple percentage method (SPM) was experts and from experts to AEA were successful while used for the calculations as shown in Table II. Out of the 25.4% were not. forty eight (48) numbers of validation forms distributed On test case 03, 73% experts and extension agent’s only forty three (43) were filled and returned. login to the web application were authenticated by Results obtained from the evaluation of the usage of granting access to the web application functions while the developed system shows that respondents believed the 27% was not. Test case 04 tested the accuracy of system will bridge the information gap amongst translation from source language (research output in researchers, extension agents and farmer functionalities, English) to farmer’s registered native language, 65% 41(95.3%) are satisfied that the linkage provided by the translation was achieved to each of the target languages implemented system is adjudged by the respondents as the while 35% were not. best ever, while 2(4.7%) thought otherwise. Meanwhile, 33(76.7%) are satisfied with usability and accessibility of B. Evaluation and Acceptance Satisfaction Analysis the system for enhancing agricultural e-extension services The evaluation of the implemented system was done in delivery while 10(23.2%) were unsatisfied. order to validate what the research work proposes, and to 24(55.8%) were satisfied that the system provides have a thorough understanding of how well it is working. outmost confidentiality on the researchers, extension In other to ascertain the effectiveness, efficiency and agents and farmers information as well as security of the capability of the implemented system for enhancing system. 19(44.2%) were not satisfied. Similarly, agricultural e-extension services delivery using the 32(74.4%) believe the implemented system is effective multilingual translation technique, the implemented and efficient in enhancing agricultural e-extension services system was used as a pilot scheme with forty eight (48) delivery while 11(15.6%) were not satisfied. respondents. That is thirty (30) farmers, ten (10) extension agents and eight (8) experts or researchers were selected within Suleja and Minna, Niger State, Nigeria. TABLE II. RESULTS OF SYSTEM EVALUATION ANALYSIS TABLE I. ANALYSIS OF INTEGRATION TEST OF THE IMPLEMENTED SYSTEM S/N Question Response Response (in (in number number and Test Test event Description of Expected Result in and %) %) case test result Percentage Yes/Satisfie No/Unsatisfie d d 01 Farmers AEA sends Farmer’s 80.3% sent 1 Are you satisfied that the 41 (95.3% ) 2 (4.7% ) mobile phone research mobile and system has bridged the receives SMS findings SMS phone received. communication gap from AEA via the system receives 19.7% not among researcher, to farmer’s research sent and extension agent and mobile phone findings received farmers? SMS 02 AEA sending Sending Send and 74.6% sent 2 Are you satisfied with the 33 (76.7% ) 10 (23.2%) research research receive and user friendliness (easy to request and request to web received. use) and accessibility of receiving experts messages 25.4% not the system as adequate research sent and for enhancing agricultural information received e-extension services and findings delivery? via web 3 Are you satisfied with the 24(55.8% ) 19 (44.2% ) 03 Users login to Authenticate Users Login was level of security and web user (experts achieved confidentiality of application and 73% to the farmers, extension agents agents) web while and researchers have 27% was information on the access to not system? web-based system 4 Are you satisfied with the 32 (74.4%) 11 (15.6%) efficiency and 04 System Research Research 65% effectiveness of the translates output output translation system for enhancing research translated from should be was agricultural e-extension outputs to English to received achieved to services delivery? farmer’s native farmers native in farmers each of the language language registered target 5 Are you satisfied with the 34 (79.06%) 9 (20.9%) native languages functionalities of the language while 35% system? failed 67 International Conference on Information and Communication Technology and Its Applications (ICTA 2016) On the general functionalities of the system [5] A.O. Ephraim, andO.O. Gloria . Information and Communication 34(79.06%) are satisfied with the performance of the Technology and Enhancement of Agricultural Extension Services in the New Millennium, Journal of Educational and Social system functionalities, as such believe the implemented Research, vol.3 No.4, pp. 155-159, July, system is a veritable tool for enhancing agricultural e- 2013,Doi:10.5901/jesr.2013.v3n4p155 extension services delivery while 9(20.9%) were not. [6] I.S. Lawal, B.Y. Boadi, O. Oladokun and T. Kalusopa. The Generation and Dissemination of Agricultural Information to V. CONCLUSION Farmers in Nigeria: A Review,Journal of Agriculture and Veterinary Science, vol.7 No.2,pp. 102-111, April, 2014. The research work implemented a multilingual [7] E.D.Kristin. Agriculture and Climate Change: An Agenda for translation system for enhancing agricultural e-extension Negotiation in Copenhagen, The Important Role of Extension services delivery that ensures real time agricultural Systems, Vision 2020, Food for Agriculture and Environment, information is provided to farmers irrespective of their International Food Policy Research Institute,USA, pp. 1-2, May, 2009. geographical location and language. The implemented [8] D. Bhalla, N. Joshi and I. Mathur. Rule Based Transliteration system translates the agricultural information from a Scheme for English to Punjabi, International Journal of Natural source language (English) into four(4) other native Language Computing, vol. 2 No. 2 ,pp. 67-73, April, 2013. languages (Arabic, Hausa, Ibo and Yoruba) depending on [9] Mamta.A Review of Various Approaches Used for Machine which the native farmer reads and understand. The Translation, International Journal of Advance Research in implemented system has also brought all the stakeholders Computer Science and Management Studies, vol.3 No. 2, pp. 108- (researcher, agricultural extension agent and farmers) in 113, February, 2015. agricultural information generation and dissemination [10] M. D. Okpor. Machine Translation Approaches: Issues and Challenges, International Journal of Computer Science,vol. 11 No. together by enabling the AEA to send farmers research 2, pp. 159-165, September, 2014. request to researchers or experts and receive research [11] T. Kalna-Dubinyuk. The Development of Electronic Extension findings from the researchers via the web-based Service in Ukraine on the International Platform, Econtechmod. An application. International Quarterly Journal, vol.1 No.3, pp. 29-33, July, 2012. In addition, farmers receive instant text messages from [12] C. Sanga, M.Mussa, S. Tumbo, M.R.S. Mlozi, L. Muhiche, &R. the AEA via their mobile phones and on requests, queries Haug. Development of the Mobile based Agricultural Extension made on agricultural information. The acceptance System in Tanzania: A Technological Perspective, International evaluation of the implemented system shows that the Journal of Computing and ICT Research, vol.8 No.1, pp. 49-67, June, 2014. implemented system is efficient and effective for [13] J.A. Antonieta.E-Extension and the Use of Mobile Phones in enhancing agricultural e-extension services delivery. Knowledge Sharing, pp.1-44, 2012. [14] L.O. Durojaiye, S.Z. Abubakar, Z. E. Omeneza, S. Muhammed, A.A Wahab, Ismail and R.A.Musa.An ICT-Based Agricultural REFERENCES Extension Service Delivery for Nigeria, Journal of Agricultural [1] N.P. Vidanapathirana. Agricultural Information Systems and their Extension, vol. 17 No.2, pp. 16-22, December, Applications for Development of Agriculture and Rural 2013,http://dx.doi.org/10.4314/jae.v17i2.3. Community: A Review Study.pp.1- 14, 2012. [15] P. Jaisridhar,R.SKumar andS. Sangeetha e – Sagu: A Web-Based [2] P. Adhiguru, P.S. Birthal and B.G.Kumar. Strengthening Agricultural Expert Advice Dissemination System from Pluralistic Agricultural Information Delivery Systems in India, Information and Communication Perspective, International Agricultural Economics Research Review,vol22, pp. 71-79, Journal of Food, Agriculture and Veterinary Sciences, vol.2 January-June, 2009 No.3, pp. 99-104, December, 2012. [3] M.WilliamandK.Trywell. Evaluation of the Agricultural [16] G.J. Shrikant and G.N Shinde. A Web Based Information & Information Service (AIS) in Lesotho, International Journal of Advisory System for Agriculture, International Journal of Information Management, pp. 4-11, 2010. Innovative Technology and Creative Engineering, vol.1 No.2, pp. [4] Final Report: FGN-AETA (2009). Agricultural Extension 1-7, February, 2011. Transformation Agenda, pp. 1-62 68