πRecommendation Engine
As for recommendations, the algorithm is built based on collaborative filtering, which is the core concept of the recommendation system. In the last decade, the concepts of machine learning and deep learning have been integrated into collaborative filtering [21, 22]. The algorithmic system combines user embedding with item embedding to obtain a single score that indicates the userβs preference for a particular item.
The final step of Ovisorβs algorithm is based on previous feature learning and time series predictions. Ovisor uses a collaborative filtering algorithm to connect users and services, which is the core of robo-advisory [19, 20]. Under the rapid development of AI, machine learning and deep learning techniques have been widely used for recommendation systems. Thus far, K-nearest-neighbor algorithm, deep neural networks and graph neural networks have been used in explainable recommendation and multi-object recommendation.
Last updated