Designing Personalisation in a Environmental Recommendation System: Differences of AHP and Fuzzy AHP Katerina KABASSI Department of Environment, Ionian University Abstract. Different methods have been proposed for designing the personalization process in a recommendation system. In the past, multi-criteria decision making theories have been proposed for the design of stereotypes in a recommendation system for environmental awareness. The main objective of this paper is on presenting the main differences when applying the fuzzy AHP and AHP for designing the weights of criteria in a recommendation system that its personalization process is based on multi-criteria decision making theories. Keywords. Multi-Criteria Decision Making, Personalisation, Recommendation Systems, Stereotypes 1. Introduction The research discipline of recommender systems arose to address the problem of information or choice over-abundance, i.e., to help users find information or items that are most likely to be interesting to them or to be relevant to their needs [1]. Recommendation systems have been expanded for use in mobile phones since these devices have been widely spread. Especially, Smartphone crowdsourcing enables real- time data gathering and gives devices access to a wealth of information about each device’s user. As a result, recommendation systems for mobile phones have been developed. The main advantage of those systems is that they can make use of the large number of modalities for interaction between them and their environment [2]. Implicit information coming from Smartphone crowdsourcing can be combined with location and time specific data can further utilize software adaptation [8]. Using all these information, we developed a mobile recommendation system for providing personalized information about Centuries-Old Olive Groves. The system is called OldGrove and is aimed to be used in the Ionian Islands. In order to provide personalized interaction, OldGrove employs user modeling techniques. More specifically, the system uses multi-criteria decision making techniques (MCDM) in combination with individual user modeling and stereotypes for providing personalized interaction. Stereotypes are used for providing the weights of the criteria and individual user modeling the values of the criteria. This paper aims at presenting the design of stereotypes using Analytic Hierarchy Process (AHP) [10]. The main objective of this research is to present a comparative analysis AHP and Fuzzy AHP (FAHP) [3] for designing stereotypes in a recommendation system. Comparison of AHP and FAHP has been implemented before but in completely different context [3, 6, 7]. Novelties in Intelligent Digital Systems C. Frasson et al. (Eds.) © 2021 The authors and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0). doi:10.3233/FAIA210097 223