Towards an Automatic Data Value Analysis Method for Relational Databases

Abstract

Data is becoming one of the world's most valuable resources and it is suggested that those who own the data will own the future. However, despite data being an important asset, data owners struggle to assess its value. Some recent pioneer works have led to an increased awareness of the necessity for measuring data value. They have also put forward some simple but engaging survey-based methods to help with the first-level data assessment in an organisation. However, these methods are manual and they depend on the costly input of domain experts. In this paper, we propose to extend the manual survey-based approaches with additional metrics and dimensions derived from the evolving literature on data value dimensions and tailored specifically for our use case study. We also developed an automatic, metric-based data value assessment approach that (i) automatically quantifies the business value of data in Relational Databases (RDB), and (ii) provides a scoring method that facilitates the ranking and extraction of the most valuable RDB tables. We evaluate our proposed approach on a real-world RDB database from a small online retailer (MyVolts) and show in our experimental study that the data value assessments made by our automated system match those expressed by the domain expert approach.

Towards a taxonomy for business capabilities determining data value

“…While qualitative data valuation focuses on generating contextual knowledge about the data value [34,38,50], quantitative data valuation concentrates on numerical information [34,35,57]. The existing literature shows that a combination [34] of both characteristics to different extents is also possible.…”

Health Datasets as Assets: Blockchain-Based Valuation and Transaction Methods

“…Data value is determined by various factors such as data complexity, number of records in the dataset, number of variables, and quality of the data (25). Furthermore, unidentified health records are less valuable because researchers need dates and geocodes to contextualize disease progression and comorbidities (26).…”

“…Vezyridis and Timmons (26) described that within the National Health Services electronic health record systems, the billing codes used to record the same disease could vary widely between healthcare practices. In addition, the massive volume of electronic health datasets also affects quality because it is challenging to implement data standards and ranges (25).…”

“…As there is a growing interest in managing health datasets as assets, blockchain technologies can improve data valuation and asset management (8). However, there are no uniform approaches to valuation (26) or assetization of health information (25). Thus far, most research conducted on data value has focused on the factors that can influence perceived data value (25).…”

“…However, there are no uniform approaches to valuation (26) or assetization of health information (25). Thus far, most research conducted on data value has focused on the factors that can influence perceived data value (25). However, there is a need to draft effective algorithms that consider data valuation methods, industry sectors, and data.…”