‘Big Data’ collaboration: Exploring, recording and sharing enterprise knowledge

Abstract

As data sources and data size proliferate, knowledge discovery from 'Big Data' is starting to pose several challenges. In this paper, we address a specific challenge in the practice of enterprise knowledge management while extracting actionable nuggets from diverse data sources of seemingly related information. In particular, we address the challenge of archiving knowledge gained through collaboration, dissemination and visualization as part of the data analysis inference and decision-making lifecycle. We motivate the implementation of an enterprise data discovery and knowledge recorder tool called SEEKER based on a real-world case study. We motivate the implementation of an enterprise data discovery and knowledge recorder tool called SEEKER (Schema Exploration and Evolving Knowledge Entity Recorder) based on the queries and the analytical artifacts that are being created by analysts as they use the data. We show how the tool serves as a digital record of institutional domain knowledge and as documentation for the evolution of data elements, queries and schemas over time. As a knowledge management service, a tool like SEEKER saves enterprise resources and time by (1) avoiding analytic "silos" (i.e., a separate set of data that is not included in an enterprise's data administration), (2) expediting the process of multi-source data integration and (3) intelligently documenting discoveries from collaborating analysts.

Approach for mapping data to client projects

“…According to the authors, "the bottleneck begins when datasets are shared in different formats, hosted across different infrastructures, or in different schemas." They further mention how the procedure of finding relevant data, analyzing it and sharing its insights is a tedious and poorly documented effort [9]. The challenges that they face are similar to the challenges faced by the UVa Open Miner.…”

“…They are challenged with archiving knowledge gained through dissemination and visualization of data. In Sukumar and Ferrell's paper [9], they address the challenges of data volume, variety, velocity and veracity seen in 'Big Data' for data-driven enterprises [10]. According to the authors, "the bottleneck begins when datasets are shared in different formats, hosted across different infrastructures, or in different schemas."…”

“…According to Sukumar, et al, "this application extracts the table and field structures from commercial and open source databases, identifies key related relationships in different databases and displays those relationships graphically for interactive exploration… the SEEKER acts as a virtual warehouse of data sources within and across enterprise datasets of interest to the user" [9]. The way in which this tool acts similar to the UVa Open Miner in this paper is that both tools allow the users to explore and interpret previously unseen entity relationships within the datasets.…”

Big Data E Gestão Do Conhecimento: Definições E Direcionamentos De Pesquisa

“…Além disso, os estudos indicam que as fontes de dados também têm aumentado, assim como o tamanho de dados. Logo, os dados em si tornam-se um recurso fundamental para uma empresa, pois oportunidades deliberadas são criadas para gerar dados dos próprios dados, de forma crescente, podendo criar valor e proporcionando uma vantagem competitiva para a empresa (SUKUMAR;FERRELL, 2013;BATRA, 2014). Fonte: Elaborado pelos autores.…”

The Characteristics of Big Data in Successful of Software Development Life Cycle for Mobile Application

“…The Variety related to the type or type of data that can be processed from structured data to unstructured data. In fact, while Velocity is related to the speed of processing data generated from various sources, ranging from batch data to real time, while the characteristics of Veracity (truth) and Value (value) are related to data uncertainty and the value of benefits from the information produced Sukumar & Ferrell (2013).…”