Veracity handling and instance reduction in big data using interval type-2 fuzzy sets. (February 2020)

• Record Type: Journal Article
• Authors: Shukla, Amit K.; Yadav, Megha; Kumar, Sandeep; Muhuri, Pranab K.
• Is Part Of: Engineering applications of artificial intelligence. Volume 88(2020)
• Publication Date: 2020-02
• Subjects: Instance reduction -- Big data veracity -- Interval type-2 fuzzy sets -- Cluster centroid -- Footprint of uncertainty; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2019.103315
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Within the aspect of big data, veracity refers to the existing uncertainty in the dataset. The continuous flow of unstructured data with unwanted noise may bring abnormality in the dataset making them unusable. In this paper, we propose a novel method to handle the veracity characteristic of the big data using the concept of footprint of uncertainty (FOU) in interval type-2 fuzzy sets (IT2 FSs). The proposed method helps in handling the veracity issue in big data and reduces the instances to a manageable extent. We have compared the results with the existing clustering based methods and examined the relationship between the clusters and the FOUs by comparing their centroids and defuzzified values. To scrutinize the validity of our results, we have also performed a number of additional experiments by appending extra instances to the datasets. To check its consistency and efficacy, the proposed methodology is assessed from three different aspects. Experimental result validates that the proposed method can suitably handle the veracity issue in big datasets and is efficient in reducing the instances.