Formal Concept Analysis Applied to Professional Social Networks Analysis

Abstract

From the recent proliferation of online social networks, a set of specific type of social network is attracting more and more interest from people all around the world. It is professional social networks, where the users' interest is oriented to business. The behavior analysis of this type of user can generate knowledge about competences that people have been developed in their professional career. In this scenario, and considering the available amount of information in professional social networks, it has been fundamental the adoption of effective computational methods to analyze these networks. The formal concept analysis (FCA) has been a effective technique to social network analysis (SNA), because it allows identify conceptual structures in data sets, through conceptual lattice and implication rules. Particularly, a specific set of implications rules, know as proper implications, can represent the minimum set of conditions to reach a specific goal. In this work, we proposed a FCA-based approach to identify relations among professional competences through proper implications. The experimental results, with professional profiles from LinkedIn and proper implications extracted from PropIm algorithm, shows the minimum sets of skills that is necessary to reach job positions.

ImplicPBDD: A New Approach to Extract Proper Implications Set from High-Dimension Formal Contexts Using a Binary Decision Diagram †

“…The proper implications set is useful because it provides a minimum representation of the data, for applications where the goal is to find the minimum set of attributes to achieve specific purposes. For example, the work developed in [18] aimed to identify relationships between professional skills of LinkedIn users. The proper implications set was applied to identify the minimum set of skills (premise) that imply in professional competence (conclusion).…”

“…Unlike Impec algorithm, which finds its proper implications considering the set of attributes M with support greater than or equal to 0, the PropIm algorithm [18] generates its proper implications (also considering the entire set M of attributes) but only with support greater than 0. Implications with support equal to 0 are valid implications (obtained from other implications, for example, Armstrong's axioms); however, in formal contexts, there is no occurrence of such implications.…”

“…Although the Impec and PropIm algorithms extract the proper implications, the computational performance of these algorithms processing high-dimensional contexts (in the number of instances) becomes impracticable. An algorithm called ProperImplicBDD, based on the PropIm [18], was proposed in [21]. The ProperImplicBDD algorithm uses a data structure known as Binary Decision Diagram (BDD) in order to manipulate high-dimensional formal contexts efficiently.…”

“…Both algorithms used the Binary Decision Diagram (BDD) as a data structure. However, the algorithm ImplicPBDD presents a novel pruning heuristic (described in Section 4) to find the premises of the implications more efficiently, when compared to [18,21].…”

“…• to use BDD as a data structure in order to extract proper implications from high-dimensional contexts in the number of instances; • to process a greater volume of data compared to the algorithms proposed in [16,18]; • to handle datasets that partially fit the FCA challenge contexts [20], namely the context containing 120,000 instances.…”