Log in / Register
Home arrow Computer Science arrow Applied Big Data Analytics in Operations Management


Data management reasserts IT role for managing critical information that is business data, its quality and reliability standards. Data present in various shapes, sizes and forms, represents valuable information that needs to be standardized and stored in a meaningful way within a central repository accessible by all business units. These shapes can be of type unstructured, structured or semi structured. SSDM provides a way to manage and store semi unstructured data in meaning full way. Many big enterprises, production houses and service companies monitor social media sources like Facebook, Twitter and Google+ etc. to get customer reviews and feedback. Technology giants companies rely on semantic techniques to manage their data in the expectancy of knowledge. SSDM try to simulate semantic architectures on relatively limited number of data.

As methodology provides theoretical techniques and step to be performed in implementing the idea so SSDM is divided into following steps that are developing core ontologies on identified RDF data sets, enrich the core ontology by adding semi structured or structure data, add rules and relations and finally executes semantic query techniques. To implement this work we used semantic based knowledge management technique (OWL). As with the advances of semantic web and semantic technology there is need to store and manage data in semantic form. Semantic technology gives storage method of knowledge instead of data, and user can understand data and reason data at execution time.


Ontology is a type of linked data structure and act as building block of semantic web. Ontology can be defined as “human intelligible and system understandable data storage method in the form of subject, relationship and object form. RDFS (Resource Descriptor Framework Schema) is first language that is recommended by w3c in 2004 for developing ontology. RDFS was built on RDF and extends features of RDF like class and properties. OWL (Ontology Web Language) is latest recommendation of ontology development. Ontology structure can be defined with quintuple as follow:

  • • Represents ontology.
  • • CL represents set of classes or concept.
  • • I represent instance or individual of ontology.
  • • CD is concept of ontology and can be super or equivalent concept.
  • • P is properties of an ontology it can be class property or data property.
  • • AD act as definition of property and it includes domain value and range value.

Ontology can be visualized using graph structure where each class is shown by a node and their relationship is indicated by a directed line. Steps to develop ontology are:

  • • Define concepts called entities,
  • • Finding relation between entities,
  • • Organize entities, and
  • • Define the attributes, values they can take and their properties and define instances.

An example ontology graph is shown in figure below where node person, item, thing etc are classes and directed line represent relationship. In the Figure 3 a rectangular box explains person is class, its super class is Item and person and item have subclass relationship that is denoted by person SubclassOf Thing. These classes may have much relationship with other classes or with subclasses. Ontology has rdf: class and rdf: subclassOf default relationship for classes

OWL provides features like cardinality constraints, disjoint classes, equality, reasoning support etc. OWL-Lite, OWL-DL and OWL-Full are type of OWL out of these three only OWL-DL is widely. SPARQL, RDQL etc are well known query languages for ontology.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science