giorgos202 Goto Github PK

crowding_system

hangman

recommendersystems_thesis

Recommender Systems is a subject that has occupied the business and research world to a great extent. It is a widely used technology based on methods of machine learning and information retrieval. Recommender Systems, starting in 1995, have developed rapidly in terms of the variety of problems they face, the techniques they use and their practical applications. Such implementations can be found on very popular online systems such as Netflix, Amazon, Pandora and many more. The majority of Recommender Systems are based on the Collaborative Filtering technique. The Collaborative Filtering technique is a process of filtering or evaluating items using the opinions of other users and is based on the assumption that if a person A has the same opinion as a person B on an issue, A is more likely to have B's opinion on a different issue than that of a randomly chosen person. Such methods have greatly occupied the research world and therefore the amount of techniques and algorithms that have been developed around this field is great. Specifically in this dissertation we conduct our study on Recommender Systems in social networks. A social network is a set of users who, in addition to interacting with objects, also develop interactions with each other. Our study and experiments are carried out in such systems where the source from which we derive information for the provision of recommendations, extends beyond the user ratings to items, to the relationships that users have developed with each other. In addition, we use techniques known as Link Prediction which are suitable for evaluating similarity between users within a graph, in order to enrich our data.

reservations

rests-m-d

Manage and Direct Restaurants