Making recommendations by integrating information from multiple social networks

Abstract

It is becoming a common practice to use recommendation systems to serve users of web-based platforms such as social networking platforms, review web-sites, and e-commerce web-sites. Each platform produces recommendations by capturing, maintaining and analyzing data related to its users and their behavior. However, people generally use different web-based platforms for different purposes. Thus, each platform captures its own data which may reflect certain aspects related to its users. Integrating data from multiple platforms may widen the perspective of the analysis and may help in modeling users more effectively. Motivated by this, we developed a recommendation framework which integrates data collected from multiple platforms. For this purpose, we collected and anonymized datasets which contain information from several social networking and social media platforms, namely BlogCatalog, Twitter, Flickr, Facebook, YouTube and LastFm. The collected and integrated data forms a consolidated repository that may become a valuable source for researchers and practitioners. We implemented a number of recommendation methodologies to observe their performance for various cases which involve using single versus multiple features from a single source versus multiple sources. The conducted

PCRM: Increasing POI Recommendation Accuracy in Location-Based Social Networks

“…(3) In [23], user interest, social and geographical influence are taken into consideration, and a USG (User interest, Social and Geographical influence based recommendation) recommendation algorithm is proposed, using a linear fusion framework to integrate these three factors, and combining the location recommendation list generated by each factor. The core of the algorithm is collaborative filtering.…”

“…Experiments show that the recommendation effect of the algorithm is not very good when the data set is sparse, and the algorithm does not consider the influence of the user on the final recommendation in different geographical locations. Algorithm in literature [23] takes into account user's preference, social influence and impact of geography, and presents a USG (User, Social and Geographical influence) recommendation algorithm using a linear integration frame work to integrate these three factors. Each factor was used to generate the location of the recommendation list of fusion to improve the accuracy of the recommendation algorithm.…”

Something’s Missing? A Procedure for Extending Item Content Data Sets in the Context of Recommender Systems

“…(1A): Several recent articles focus on the extension of data with data from a distinct area, for example, data from different domains such as music and film (cross-domain data sets; Abel et al 2013;Ntoutsi and Stefanidis 2016;Ozsoy et al 2016), context information such as time and location (multi-dimensional data sets; Abel et al 2013;Kayaalp et al 2009) or data from different social and semantic web sources such as Wikipedia, Facebook and Twitter (heterogeneous data sets; Abel et al 2013;Bostandjiev et al 2012;Chang et al 2018;Kayaalp et al 2009;Ozsoy et al 2016). These approaches examine whether the diversity of data types leads to improved recommendation quality but do not systematically extend item content data with additional data from the same domain.…”

“…These approaches examine whether the diversity of data types leads to improved recommendation quality but do not systematically extend item content data with additional data from the same domain. (1B): Other works in literature analyze user profiles from different social networks (Abel et al 2013;Li et al 2018;Ozsoy et al 2016;Raad et al 2010).…”

“…However, these works do not focus on item content data at all. (1C): Finally, some recent works focus on the extension of item or user data from multiple data sources in the context of recommender systems (Abel et al 2013;Bostandjiev et al 2012;Bouadjenek et al 2018;Ozsoy et al 2016). These approaches rely on tools such as BlogCatalog, Google Social Graph API, Google Search API or OpenID, which provide information for the matching of users or items.…”