Every real-life environments where users interact with items (products, films, research expert profiles) have several development phases. In the Cold-start phase, there are almost no interactions among users and items content-based recommendation systems (RS) can only recommend based on matching the attributes of the items. In the transition state, items start to collect user interactions but still a significant number of items have too small number of interactions, RS does not allow users to discover cold items. In a regular state, where most of the items in the system have enough interactions, the recommendations often suffer from low diversity of the items within a single recommendation. This article proposes a general recommendation algorithm based on Ontological-similarity, which is designed to address all the above problems. Our experiments show that recommendations generated by our approach are consistently better in all environment development phases and increase the success rate of recommendations by almost 50% measured using ontology-aware recall, which is also introduced in this article.
Every real-life environments where users interact with items (products, films, research expert profiles) have several development phases. In the Cold-start phase, there are almost no interactions among users and items content based recommendation systems (RS) can only recommend based on matching the attributes of the items. In the transition state, items start to collect user interactions but still a significant number of items have too small number of interactions, RS does not allow users to discover cold items. In a regular state, where most of the items in the system have enough interactions, the recommendations often suffer from low diversity of the items within a single recommendation.This article proposes a general recommendation algorithm based on Ontological-similarity, which is designed to address all the above problems. Our experiments show that recommendations generated by our approach are consistently better in all environment development phases and increase the success rate of recommendations by almost 50\% measured using ontology-aware recall, which is also introduced in this article.
Every real-life system has several development phases. In the beginning, it has no interactions; wecall this the cold-start problem. After a while, where user interactions occur in the system, the sys-tem gets to a state where one part of the catalogue has some interaction that could be enough touse CF algorithms. However, there’s still a larger amount of items with a small amount of inter-action, so-called cold-item. We call this state a transition state. If most of the item in the systemhas interactions, we call this a regular state. The system needs a slightly different behaviour fromthe recommendation system at each stage. At the cold-start stage, we need exploration of thecatalogue and the involvement of the maximum number of cold-items to use CF as soon as pos-sible. In the transition phase, we need diversification because there are still many items with asmall amount of interaction. In the regular phase, we want to offer new relevant items to usersto surprise them and keep them from getting into the information bubble and thus have a betterexperience using the system. Our approach uses Ontology, which can provide semantic item simi-larity if the text attributes contain the exact words but even if the text attributes look different.This paper presents a Universal Ontological-Based Algorithm (OBACS) that uses diversificationmethods such as serendipity and diversity to improve user experience in all system developmentphases. We verify our approach experimentally on the Movie Database and describe the results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.