Ontology development for flexible content

Abstract

Competitive environment in converging digital media industry is changing rapidly. This calls for tighter integration of creativity, business processes, and technologies. Media companies need flexible methods to manage electronic content production and delivery, and metadata is a key enabler in making this goal a reality. However, metadata is useful only if its nature is understood clearly and its structure and usage are well-defined. The management of common vocabularies is one of the key issues in blocking electronic publishing as well as other automated business interactions in the electronic world. For this purpose, an ontology, consisting of conceptual models that map the content domain into a limited set of meaningful concepts, is needed. The result is a high-level semantic description captured in document metadata. Moreover, since the metadata is defined in multiple dimensions that each have their own conceptual model, this explicit typing of descriptive data can be used directly in applications and services that utilize semantic metadata, such as information filtering and information augmentation. This paper introduces an ontology development framework rooted at the core business processes of electronic publishing that can be used when defining semantic metadata structures for electronic content. The framework underlines the different nature of ontology development and metadata publishing, and how these two processes influence each other. This paper discusses also the application of the ontology development framework in practice. The framework has been created in the SmartPush project, where media companies explore new business opportunities for electronic publishing and delivery.

Visualizing the Semantic Web

“…Uschold and Gruninger, 1996;Jokela et al , 2000). For the first set of documents, both stages were executed using methods drawn from the ontology literature (Le.…”

“…The first stage was to examine the documents' audience, as suggested in Jokela et al (2000), and to gather information about requirements. The first stage was to examine the documents' audience, as suggested in Jokela et al (2000), and to gather information about requirements.…”

<title>Personal video retrieval and browsing for mobile users</title>

“…Tools have been created that facilitate manual description of video content during production, while low-and mid-level metadata can be created automatically by audio and visual content analysis tools [2], [3], [4]. Ontologies, in combination with personalization have been proposed to perform intelligent queries and video summarization from metadata [5], [6]. However, all these efforts so far have been mainly concerned with professional, rigidly structured video material -especially news and sport programs [7], [8], [9] -whereas home video lacks professional quality and structure.…”

The role of structured content in a personalized news service

“…Consequently, we did not concentrate on analyzing the descriptiveness of the metadata. Although we have discussed the different components affecting semantic metadata quality in our previous work [6], the development of suitable methods to measure metadata quality should be addressed in the future research.…”

“…The final version of the domain ontology was refined together with the content matter expert based on the experiences gathered from the testing with the actual news material. Details on the principles and methods applicable to building ontologies for content can be found in our previous work [6].…”