Personalised web services for agricultural domain: a case study for recommending organic seeds to farmers and growers

Abstract

In this paper, we introduce RecOrgSeed, a model for producing interesting recommendations to organic farmers/growers. The whole process has been distinguished into two stages, one off-line that includes data preparation, ontology creation and usage mining and one online that concerns the recommendations production. The knowledge about farmers/growers (users) and seeds (products) is extracted from usage mining data and semantic annotations in conjunction with user-product ratings and matching techniques between users. The preliminary evaluation experiments show that even in the case of "cold-start problem" where no initial behavioural information is available, the approach can provide users with logical and relevant recommendations.