Linked 'Big' Data: Towards a Manifold Increase in Big Data Value and Veracity

Abstract

The Web of Data is an increasingly rich source of information, which makes it useful for Big Data analysis. However, there is no guarantee that this Web of Data will provide the consumer with truthful and valuable information. Most research has focused on Big Data's Volume, Velocity, and Variety dimensions. Unfortunately, Veracity and Value, often regarded as the fourth and fifth dimensions, have been largely overlooked. In this paper we discuss the potential of Linked Data methods to tackle all five V's, and particularly propose methods for addressing the last two dimensions. We draw parallels between Linked and Big Data methods, and propose the application of existing methods to improve and maintain quality and address Big Data's veracity challenge

Big and open linked data analytics ecosystem: Theoretical background and essential elements

“…They claimed that the technical infrastructure is essential for employing data analytics in a decentralized manner on the Web. Similarly, Debattista et al (2015) explored the potential of data analytics in the Web of Data focusing on the parallels between methods used for big data and linked data.…”

“…Their contribution and importance in value creation varies. Some researchers emphasize data processing and analysis as the key phase (Debattista et al, 2015;Günther et al, 2017;Miller and Mork, 2013), while the others consider data publication, sharing, and reusing as the most important value generating mechanisms (Attatd et al, 2016;Charalabidis et al, 2014;Jetzek et al, 2014). However, both these phases are complementary since the output of each phase is provided as a feedback to the other.…”

Methods for Assessing, Predicting, and Improving Data Veracity: A survey

“…Some have related data veracity to the confidence level (i.e., integrity, availability, completeness, consistency, and accuracy) of the data and its sources (Berti-Équille, 2015;Herrera et al, 2019;Patgiri and Ahmed, 2016;Moyne and Iskandar, 2017;Batista et al, 2017). Debattista measured veracity in terms of dereferenceability and consistency (Debattista et al, 2015). In the database area, veracity has been related to data inconsistencies, duplicates, and missing or incomplete data (Berti-Equille and Borge-Holthoefer, 2015).…”

“…Natural language processing (Rubin, 2006), (Rubin and Lukoianova, 2013), ensemble methods (Berti-Équille, 2015); Reservoir sampling (Debattista et al, 2015), bloom filter (Debattista et al, 2015); Health information classification;…”

A survey on scholarly data: From big data perspective

“…Considering that data will be collected from heterogeneous sources and may exist in different formats, the concept of data linking can be used. Debattista et al [45] gave useful insights on how 'Linked Big Data' can significantly improve the veracity and value dimensions of data, also drawing parallels between methods used for big data and linked data. Linked data is a useful concept for finding events of interest and solving queries that were otherwise not possible [155] [156].…”