Industrial Informatics: Assertion of Knowledge from Raw Industrial Data

Industrial Informatics: Assertion of Knowledge from Raw Industrial Data

Iram Shahzadi, Qanita Ahmad, Imran Sarwar
DOI: 10.4018/978-1-4666-0294-6.ch010
(Individual Chapters)
No Current Special Offers


Correct and timely access to business information is the key to success in industry. However in industry, data is generated on daily basis and increases exponentially. Therefore, managing it is a challenging task for every organization. To deal with this phenomenon of information overload, organizations are in dire need to find and set up potential means for the analysis of raw industrial data (i.e. texts) and draw necessary information from it. This information can result in knowledge and knowledge leads towards wisdom, the essence of every business. This chapter is concerned with the use of knowledge management systems to cater information overload hassles, the organizations are facing today. As a solution, a detailed study of currently existing open source data and knowledge management systems is conducted. Hence, this chapter discusses the state of the art tools and technologies in this domain, and highlights the need and importance of semantic applications for industrial data processing.
Chapter Preview


Extraction of relevant knowledge, from the gathered information, is the most significant and effectual factor for any business to run successfully and to compete in current era. However the data and information generated in most industrial organizations, recently, is not in proper format to build knowledge base for these organizations. Generally, industrial data exists in unstructured format. So companies need to perform many activities to manage this data and information, and manipulate it to generate useful knowledge. Knowledge worth more than information and the ability to make precise decisions based on the available knowledge has the highest value. In industry, knowledge can be extracted from a vast source of resources of data and texts including web pages, local and distributed repositories, databases, files etc. Thus knowledge generation and management is the central focus for the success of any industry. This knowledge can play its role as a valuable asset in organization’s policy making and can result in improved resource management, business production or revenue generation.

However, as a significant portion of industrial data is not structured, it is not easy to find, access, analyze, or use this data to locate pieces of useful information and extract meaningful knowledge from it. This requires the transformation of unstructured data into organized information and thus the information leads towards knowledge generation, discovery, and extraction, generally covered under the name of knowledge management.

The process of knowledge management for industrial data can be carried out either manually or by some automated way. However, organizing the unstructured industrial data manually requires excessive resources in terms of man-power, time, equipment, money etc. Contrarily, automatic techniques help in better and efficient utilization of the available resources in lesser amount of time. Therefore, unique analytical tools are required to assert intelligence from industrial data by automatically understanding and analyzing it.

Knowledge generation covers the discovery and delivery of quality information to the user, eliminating the irrelevant ones and focusing on the relevant pieces of information. Higher the relevancy of extracted knowledge, the underlying knowledge management system will be more efficient and reliable. In industry, the knowledge management efforts help in achieving the business objectives such as improved performance, competitive advantage, innovation, the sharing of lessons learned etc. It follows the path from raw data to fine knowledge generation and then its application to diverse industrial domains, as shown in Figure 1.

Figure 1.

Process of knowledge generation and usage


The process of knowledge management starts by taking the raw facts and leads towards information. Information is basically understanding of data pieces within a given context. It deal with what, who, where, when etc. Whereas knowledge is the manipulation of information in such a way that, it can deal with the “how” part. It comprises of all the strategies, practices or approaches that are incorporated to perform a given undertaking. Thus, a collection of data is not information and a collection of information is not knowledge (Knowledge Management, 2010).

In this chapter we discuss how the use of knowledge management systems helps to handle the available resources more efficiently in industry. How the available industrial unstructured data can be transformed into organized information to assert valuable knowledge from it and how this redounds to an increase of productivity, specifically in industry. In this regard, the key focus is on ontologies and semantic technologies for industrial data manipulation. The overall objectives of this chapter are to familiarize the practitioners, readers and researchers with the following.

Types of Data

Generally, data is divided into three following types:

  • Structured data

  • Semi-structured data

  • Unstructured data

We will discuss each of these one by one in the following section.

Key Terms in this Chapter

Knowledge Management (KM): KM deals with the creation, manipulation, organization and maintenance of the knowledge.

Open Source Tools: Open source tools are freely available softwares under open source license. These softwares can be used with the provided functionality or may be changed, enhanced or modified according to the needs, free of charge.

Knowledge: Knowledge is the familiarity or awareness about a domain, which is required to work in that domain.

Semantic Technology: A specification to represent unstructured text data in such a way that it can be processed interlinked and shared by the machines.

Ontological Database: A database containing organized knowledge instances extracted from the text by applying domain ontology along with the defined rules.

W3C: W3C is an international community that defines the standards for semantic web. Aim of W3C is to transform the web data into a useful, sharable and machine understandable format so that the machines can interpret this data to deduce valuable knowledge.

Metadata: Metadata is the additional data attached with the given data, to make it self-explanatory and meaningful. Basically, meta-data helps in identifying the meanings of data so that machines can operate on this data to deduce its meanings.

Ontology: A data structures used to define concepts, relationships, properties, restrictions and, conditions that are applicable on that concept. Semantic technologies perform mapping and reasoning with data using the defined ontologies.

Knowledge Engineering (KE): KE is an artificial intelligence domain that covers technological details i.e. procedure, methods and techniques required for KM.

Seed Domain Ontology: It is a domain specific ontology containing the core concepts of the underlying domain. This ontology can be populated and enhanced with the passage of time to incorporate new data about the domain and also to enrich it.

Complete Chapter List

Search this Book: