EXPLOITING INTELLIGENT CONTENT VIA AXMEDIS/MPEG-21 FOR MODELLING AND DISTRIBUTING NEWS

PIERFRANCESCO BELLINI, IVAN BRUNO, PAOLO NESI
2011 International journal of software engineering and knowledge engineering  
The content technology needs to attain forms with more intelligence, flexibility and complete features than those being currently on the market or proposed by standards. In this paper, an analysis of the state of the art about intelligent and complex content models is presented. The analysis allowed identifying a number of topics and features which models and formats should evolve according to. The work has been used to extend AXMEDIS content model and format which in turn is grounded on
more » ... , SMIL, HTML, and other standards. The Extended AXMEDIS format presents a set of new features among them: semantic descriptors, extended annotations, intelligent behavioral and semantic computing capabilities. The newly obtained format has been compared against NewsML which is one of the most widespread formats for news production and distribution. The management of news has some peculiarities such as container, production tools and players, that may take advantages of the intelligent content features and applications. Moreover, news have to be massively processed for ingestion and repurposing, and present relevant requirements on right control. Also these features may be satisfied by AXMEDIS tools . To this end, a comparative analysis of processing and modeling NewsML with AXMEDIS tools and format has been performed and reported to verify the usage. In addition, AXMEDIS format can be profitably used for a range of innovative applications of intelligent content. MARC [11], etc. and among the identification codes, ISBN, ISAN, ISRC, ISMN, [21], etc., but also classical URI. More recently, a certain number of formats has been presented with the aim of providing more advanced experiences to final users enforcing some degree of intelligence into the object format among them: ACEMEDIA [10], [23], X-MEDIA [24], [25], AXMEDIS [26], [27], [28], EMMO [29], and KCO [30] . ACEMEDIA defined a content format to enable creating personalized content collections. X-MEDIA content model is mainly focused on semantic aspects that can be managed by ontologies and RDF. X-Media is mainly oriented towards knowledge management and sharing with limited application to text and image contents and it has related content objects with very limited autonomy of work that are not proactive with the user. AXMEDIS is an extended version of MPEG-21 supporting DRM and proposing content packing with presentation capabilities in HTML, FLASH and SMIL [12] , including behavioral capabilities and semantics descriptors. EMMOs (Enhanced Multimedia Meta Objects) encapsulates relationships among multimedia objects and maps them into a navigable structure. An EMMO contains media objects, semantic aspect, associations, conceptual graphs, functional aspect. KCO, Knowledge Content Objects, is not a package, and it is based on the DOLCE foundational ontology and has semantic aspects to describe the properties of KCOs, including raw content or media item, metadata and knowledge specific to the content object and knowledge about the topics of the content (its meaning). The semantic information in a KCO includes: content description; propositional description (Semantic Description and Content Classification); presentational description; community description (the purpose); business description (the trade, price...); trust and security description, self description (the structure). Most of these last models present descriptors that may be used for powerful semantic classification. Some of them present also capabilities to formalize content behavior, for example in Java and/or JavaScript. Among the formats mentioned, the AXMEDIS implementation of the MPEG-21 file format and MXF [3] supports the direct play. Only the MPEG-21 also supports a range of business and transaction models via DRM (Digital Rights Management) solutions [9], [10] and with a set of technological protection supports.
doi:10.1142/s0218194011005141 fatcat:fn5rnttynjdiven7i2xp4a22iy