ORIGINAL ARTICLE Social movie recommender system based on deep autoencoder network using Twitter data Hossein Tahmasebi 1 • Reza Ravanmehr 1 • Rezvan Mohamadrezaei 1 Received: 28 October 2019 / Accepted: 4 June 2020 Ó Springer-Verlag London Ltd., part of Springer Nature 2020 Abstract Recommender systems attempt to provide effective suggestions to each user based on their interests and behaviors. These recommendations usually match the personal user preferences and assist them in the decision-making process. With the ever-expanding growth of information on the web, online education systems, e-commerce, and, eventually, the emergence of social networks, the necessity of developing such systems is unavoidable. Collaborative filtering and content-based filtering are among the most important techniques used in recommender systems. Meanwhile, with the significant advances in deep learning in recent years, the use of this technology has been widely observed in recommender systems. In this study, a hybrid social recommender system utilizing a deep autoencoder network is introduced. The proposed approach employs collaborative and content-based filtering, as well as users’ social influence. The social influence of each user is calculated based on his/her social characteristics and behaviors on Twitter. For the evaluation purpose, the required datasets have been collected from MovieTweetings and Open Movie Database. The evaluation results show that the accuracy and effectiveness of the proposed approach have been improved compared to the other state-of-the-art methods. Keywords Social recommender system Á Deep learning Á Deep autoencoder network Á Collaborative filtering Á Content-based filtering Á Social influence 1 Introduction In recent years, a sudden increase in online information has caused confusion among users. Recommender systems are information filtering tools that direct users toward their interested, relevant products or services in a personalized way. Among various systems available to access informa- tion, the recommender system plays an important role in improving businesses and facilitating users’ decision- making [1]. In general, the list of recommendations is created based on users’ preferences and interests, items’ properties, users’ past interactions, and some additional information such as time and spatial data [2]. Recom- mender system models, based on the types of input data, are mainly classified into collaborative filtering (CF), content-based filtering (CB), and hybrid [3]. Recommender systems are evolving alongside the web; in recent years, they have been increasingly used in many areas of internet, such as e-commerce [4] and intelligent marketing [5], electronic book recommendation [6], music/movie recommendation [7], and tourism industry [8]. The objective of a recommender system is to make recommendations appropriate for each user, and therefore, these systems seek to balance accuracy, novelty, disper- sion, and stability of the recommendations. Considering the ever-increasing amount of information in the virtual world, and the confusion of users in searching and selecting the required information, the use of recommender systems is increased. One type of recommender system is social recommender systems [9]. The purpose of these systems is to reduce the information overhead by providing the most attractive and the most relevant content to the users of social media. In & Reza Ravanmehr r.ravanmehr@iauctb.ac.ir Hossein Tahmasebi hos.tahmasebi.eng@iauctb.ac.ir Rezvan Mohamadrezaei rez.mohamadrezaeilarki.eng@iauctb.ac.ir 1 Department of Computer Engineering, Central Tehran Branch, Islamic Azad University, Tehran, Iran 123 Neural Computing and Applications https://doi.org/10.1007/s00521-020-05085-1