A New Approach for Scholars Matching Using Universal Quantifier Queries

Universal quantifier queries on recursive relation are defined as the set of queries that query the database to get pairs of records (r1, r2) from the same relation such that the second record of each pair must intersect with the first record in a set of required attributes. Such query remains to be an interesting type of queries specially today with the appearance of many applications that need them. Today there are many databases that include scholarly data like DBLP, those databases though include information about papers, conferences, authors, and many other useful information, they don´t correlate authors with similar interests or research directions. Nevertheless, the current explosion in the amount of data has driven the need for new techniques and technologies as traditional database techniques are no longer adequate to manage, store, and query those large amounts of data. Thus, the use of cloud emerged as a solution for several big data problems. Using clusters of commodity machines turn to be an optimal solution. Recently, there has been considerable interest in designing new algorithms using inverted index to efficiently answer different types of queries over big data. In this paper, we present a new technique for evaluating universal quantifier queries on recursive relation on large scholarly datasets using the popular MapReduce framework and inverted index. In addition, we present experimental results that show the performance of our proposed technique over the famous scholarly DBLP data-set.