Feature-Wise Ranking of Candidates through Maximum Degrees in Hidden Bipartite Graphs

In this day and age of technological breakthroughs, electronic recruitment tools have gained much recognition due to their increasing popularity among recruiters. Many methods like Learning To Rank and Multi-Criteria Decision making have been employed inside these tools to enhance the process. The ranking is one of the most important parts of e-recruitment on which these methods and techniques are applied. Among these methods, the research area of graphs has not been explored enough in the context of ranking. Keeping this in view, this paper uses the k-Most Connected Vertices (kMCV) Problem of connected graphs to propose ranking in e-recruitment. Specifically, we map the finding of hidden bipartite graphs to the application of e- recruitment by first modifying it in the form of a graph and then apply to rank. Considering the hidden edges of bipartite graphs, matching between candidates and job descriptions to find the most suitable candidate for a particular job is done. To perform this task, we extend the use of the Switch-on-Empty (SOE) algorithm on a modified bipartite graph to propose a solution. We describe an algorithm, namely, SOE++, that adaptively employs the matching of composite and atomic nodes to solve the kMCV problem. We further explore the application of the algorithm in the domain of e-recruitment for ranking. The single run of this algorithm reveals information of the whole node of one set corresponding to a single feature node of the other set. Theoretical analysis shows that significant gains in performance are achieved when compared to the previous algorithm.