A manuscript titled, “On the Efficient Representation of Datasets as Graphs to Mine Maximal Frequent Itemsets” authored by Dr. Zahid Halim et al. has been accepted for publishing in the upcoming issue of the IEEE Transactions on Knowledge and Data Engineering (TKDE). The transaction is ranked # 1 under the domain of databases and information systems. The TKDE is classified in the gold category by the reputed Journal Quality Ranking System (JQRS). Additionally, it is a Q1 journal according to the Web of Science (Clarivate Analytics).
The work presents a graph-based approach for representing a complete transactional database. The proposal enables to store all relevant information, for extracting Frequent Itemsets (FIs) of the database in one pass. The said work also developed an algorithm that extracts the FIs from the graph-based structure. Experimental results are reported comparing the proposed approach with 17 related FIs mining methods using six benchmark datasets. Results show that the proposed approach performs better than others in terms of time. This is a groundbreaking research utilizing graphs as a data structure for data mining purposes.
This work was conducted under the umbrella of one of the Institute’s research group – MInG (https://www.minrg.org). MInG is working in close collaboration with national and international partners to solve problems in the domain of Artificial Intelligence, Data Mining, and Machine Learning. The conducive environment at GIK has always been helpful to produce high-quality research. The research group has published their applied and theoretical research at top-notch venues.
The said publication is an outgrowth of a research project conducted at the MInG. The project team comprised of Mr. Omer Ali and Mr. Muhammad Ghufran Khan. Omer is presently pursuing his Ph.D. and working as a Research Fellow at Oslo University Hospital (world rank 131-QS). Whereas, Ghufran has recently secured a fully funded Ph.D. in the domain of AI at TU Dresden, Germany (world rank 157-QS). This is depictive of the potential of GIK graduates.
The article can be read at: https://doi.org/10.1109/TKDE.2019.2945573