Relational intelligence recognition in online social networks — A survey

Abstract Information networks today play an important, fundamental role in regulating real life activities. However, many methods developed on this framework lack the capacity to adequately represent sophistication contained within the information it carries. As a result, they suffer from problems such as inaccuracies, reliability and performance. We define relational intelligence as a combination of affective (Cambria, 2016; 2015 [1] , [2] ; Hidalgo et al., 2015 [3] ), sentimental (Ferrara and Yang, 2015 [4] ; Wang et al., 2013 [5] ; Madhoushi et al., 2015 [6] ) and ethical (Vayena et al., 2015 [7] ; Nunan and Di Domenico, 2013 [8] ; Anderson and Guyton, 2013 [9] ) developments reflected in the evolving patterns of online social structures. These developments involve the ability of actors to adaptively regulate emotions, values, interest and demands between each other in an online social scene. In this paper, we provide a state-of-the-art overview of approaches used in recognizing relational intelligence — with special focus given to Online Social Networks (OSNs). The important core processes of data mining, identification (extraction), detection (labeling), classification, prediction and learning which empower machine recognition tasks will be discussed in detail. In addition, widely affected applications like recommending, ranking, influence, topic modeling, evolution, etc. will also be introduced along with their basic concepts uncovered to a detailed degree. We also include some discussions on more advanced topics that point to further interesting future research directions.