semantic web

Category: Education

Presentation Description

No description available.


Presentation Transcript

Usability of Semantic Web for Enhancing Digital Living Experience :

By T.Seeta Rama Rao (10C61D5826) Usability of Semantic Web for Enhancing Digital Living Experience


Abstract The number of different types of devices on the home network is expanding rapidly. So every year new devices are hitting the market and reaches the consumers, there are challenges ensuring compatibility among these devices and providing a comprehensive user interface that supports consumers managing their digital content across several devices.

PowerPoint Presentation:

The Digital Living Network Alliance (DLNA) has specified an architecture that enables interoperability between the various devices and allows a user to enjoy the desired content across several devices. To complement DLNA, This paper investigate Ontological representation from semantic web technology to model the interaction between multiple home devices and to provide an enriched and more comprehensive metadata based multimedia search . So that it removes the user burden of searching each device individually. Development of a prototype incorporating these ideas has begun, using a well known semantic web toolkit Jena.


Contents Introduction Semantic web and web ontology languages. Multiple device task and its Representation.a ) Ontological Representation of multiple device Interaction. Application of semantic web in DLNA to enhance User Experience. a) Architecture of DLNA b) Use Cases Prototyping Design of Dynamic User Interface Conclusion References


INTRODUCTION As we have many number of new devices Hitting the market every year brings an expanding array of connected consumer electronics devices, intended for deployment on the home network . It brings a number of problems as follows (1) Achieving a consistent, powerful and engaging user experience among the expanding array of connected or related devices, (2) Achieving a unified point of control for this expanding array of devices.

PowerPoint Presentation:

For this purpose , the Digital Living Network Alliance (DLNA) defines a framework that provides basic network , service and digital media interoperability for forthcoming CE’s applications. Home contains many devices from many vendors . So achieving compatibility and consistency among devices is at most important. To alleviate this problem in this paper authors proposed and taken the initiative to model the interaction between multiple home devices using Semantic Web and web ontology languages (OWL/RDF). NOTE: Sony established the DLNA in June 2003 as the Digital Home Working Group , changing to its current name 12 months later,

PowerPoint Presentation:

Semantic Web ? Semantic Web is a group of methods and technologies to allow machines to understand the meaning – or "semantics" – of information on the World Wide Web. It Describe things in a way that computers applications can understand it.

Web ontology Languages ? :

Web ontology Languages ? It is a set of markup languages which are designed for use by applications that need to process the content of Information instead of just presenting information to humans. OWL ontologies describe the hierarchical organization of ideas in a domain, in a way that can be parsed and understood by software . OWL has more facilities for expressing meaning and semantics than XML , RDF , and RDFS, and thus OWL goes beyond these languages in its ability to represent machine interpretablecontent on the Web . OWL is part of the W3C recommendations related to the Semantic Web .


II. MULTIPLE DEVICE TASK AND ITS REPRESENTATION Let us consider the following devices Cable set-top boxes (STB), IP STB, Mobile Phone, PDA, Laptop, Desktop PC, Stereo Receiver, TV Monitor, Camera, Refrigerator, Washer/dryer, Door bell, Air conditioner, Lighting, Door lock and Car. With each we identified their different tasks as shown in Fig. 1 and modeled their interaction using ontological technique from semantic web.

PowerPoint Presentation:

We use Resource Description Framework (RDF), a low level building block from semantic web to represent simple concept Taxonomies and ontologies for associated data. Triples: The RDF metadata model is based upon the idea of making statements about resources in the form of subject-predicate- object expressions, called as Triples. Here , subject denotes the resource predicate denotes properties or aspects of the resource Object Expressions expresses a relationship between the subject and the object. For example, Let us consider a mobile phone and represented its different tasks in terms of RDF as shown in Table I.

A. Ontological Representation of Multiple Device Interaction :

A. Ontological Representation of Multiple Device Interaction Using an ontology compliance level a new ontology for multiple device interaction (OWL Lite) was created and also generated the model and RDF/XML schema in the Semantic Works interface from Altova Inc . We have defined three classes 1) ConsumerElectronics 2) MobilePhone 3) Multimedia.

PowerPoint Presentation:

We define MobilePhone as a subclass of ConsumerElectronics, which essentially states that any instance of the MobilePhone class must also be an instance of the Consumer Electronics class. Multimedia class is used to (i) define it as the range of a property called device :hasPlay or device :hasStore (ii)create instances of Multimedia.

PowerPoint Presentation:

Defined MobilePhone class to be the domain of the properties hasStore/hasPlay, and the class Multimedia to be the range of the properties hasStore/hasPlay. The properties hasStore/hasPlay applies to the MobilePhone class and takes values that are instances of the Multimedia class. Properties are created at a global level and then related to different classes.

PowerPoint Presentation:

In our ontology, we deal with two properties to carry hasStore and model# A multimedia can be Audio, Video or Picture. We will create this property as an object property. So that we can relate one resource to another. Here we wish to relate instances of the MobilePhone class to instances of the Multimedia class via the hasStore property. The class (or classes) that the property applies to is called the property’s domain , while the set of values the property can take is called the property’s range.

PowerPoint Presentation:

model# -------------------- It is a literal value (a positive integer ) and also object of MobilePhone class which is the name or number of the model). refer Fig. 2 The class MobilePhone is a subclass of the class ConsumerElectronics , and has two properties: the object property hasStore and the datatype property model#. So far we have created three classes,  ConsumerElectronics ,  MobilePhone ,  Multimedia , and two properties, the object property  hasStore the datatype property  model#.

PowerPoint Presentation:

We have defined both properties to apply to the MobilePhone class (by making this class the domain of the properties). Further, we have defined the range of the hasStore property to be instances of the Multimedia class, and the range of the model# property to be a literal value of the XML Schema data type positive Integer. Now we will first create three instances of the Multimedia class, which will be simple instances like video, picture and audio as shown in Fig. 3.

PowerPoint Presentation:

Next , we define three more instances of the MobilePhone class as shown in Fig. 4. now we add a predicate. MobilePhoneAudio instance with its predicate is shown in Fig. 5.

PowerPoint Presentation:

The instance MobilePhoneAudio, MobilePhonePicture, MobilePhoneVideo has therefore been defined to:  Be an instance of the class MobilePhone,  Its object property are hasStore and hasPlay . both takes the instance Multimedia as its object, we can define model# = SGH609 as its value


III . APPLICATION OF SEMANTIC WEB IN DLNA TO ENHANCE THE USER EXPERIENCE The metadata access can be enhanced by making use of semantic web technology. With the multitude of home electronics devices available, it is becoming difficult for consumers to manage their digital content. The Digital Living Network Alliance (DLNA) Has specified an architecture to enable interoperability between the various devices which helps the user to enjoy the desired content.

PowerPoint Presentation:

However, in a typical home, users (now or in the Future) might store thousands of multimedia items across several devices, making it is tedious to search every single device individually. Additionally, we get incomplete search results in the case of a metadata based content search if metadata or semantics is not available. So our approach is as follows

PowerPoint Presentation:

In this paper, DLNA architecture to address these problems were extended by using a specialized device as a service enabling platform and to hide the complexity of distributed content storage. The Semantic Web provides an enriched and more comprehensive metadata-based multimedia search experience to enhance the user experience.


A. ARCHITECTURE The standard DLNA architecture as shown in Fig 6 consists of several Digital Media Servers (DMSs) and a control point that communicates with these DMSs to search for multimedia content. A user will use a control point to invoke the Content Directory Service (CDS) actions to search for the desired content hosted on that particular DMS He is required to repeat this for all available DMSs until he finds what he wants.

PowerPoint Presentation:

Next, we use a Virtual Media Server (VMS),which acts as a single point of contact for the user, as shown in Fig. 7, to relieve the user from the burden of Contacting multiple DMS’s separately.

PowerPoint Presentation:

We also introduce another new component into the architecture: the Semantic Web Engine (SWE) as shown in Fig. 8. Which uses knowledge Imported from external ontologies and data sources to extrapolate the semantics of the available metadata, thus enabling a smarter content search mechanism.

B. Use Cases:

B. Use Cases The following scenarios helps us for idea of getting the benefit of semantic web engine and DLNA. Scenario 1: Let we search IMDb (Internet Movie Database) and AFI (American Film Institute) as an external ontological data sources to get movie information (such as movie reviews, ratings, background production information etc.) and then search the user’s personal digital media server to see if that particular film is available in his personal collections. Also to extend this, the SWE can get the contact list (including shared DMS information) of the user’s friends from an external ontological data sources (such as Facebook ) and in a similar way each one of the user’s friends could share the movie information they have. As shown in the figure 9

PowerPoint Presentation:

Example 2) :The user searching for songs sung by a specific singer. In addition to the simple CDS:Search () results matching the artist’s name, the VMS could provide additional results with the help of the SWE. Let’s say that the SWE can access an internet RDF audiography database. It can look up this information to find the bands that the given artist has been part of and then list music available in the digital home by those bands as additional search result s.

PowerPoint Presentation:

The SWE derives a URI representing an RDF resource from the search keyword by pretending it with the base URI of the ontology data being used. In our example search for songs sung by Paul McCartney, the SWE uses the URI “ Paul McCartney” as the starting node for the RDF query, assuming the base URI “ audiographyband #”. It then uses the audio graphy band data imported from external data sources along with the Content metadata requested from each of the DMSs to perform the RDF query and extract additional matches for the user’s search request. The songs matching the search query are listed in Table II .

PowerPoint Presentation:

We can observe that in addition to the songs marked by Paul McCartney, other songs from other bands“Beatles ”, “Wings” where Paul McCartney as in, have also been included in the results. Thus the search results were enriched and enhanced with the use of external ontological data.


IV. PROTOTYPING We have developed the Semantic Web Engine using Jena 2.5.3 [2] an open source semantic web toolkit. The Demonstration software includes a RDF vocabulary of movie and music which has been generated using Semantic Works as shown in Fig10.

Fig. 10. A snapshot of the RDF vocabulary for Movie and Music :

Fig. 10. A snapshot of the RDF vocabulary for Movie and Music

PowerPoint Presentation:

This vocabulary has been converted to a java class file using the schemagen utility from Jena API which helps to get access to different properties of RDF graph based model.


A sample of this class file is shown in Fig. 11. We assume there are two Digital Media Servers, one for the movie Database and another for the music database, both of which host Data in RDF metadata format as shown in Fig. 12 and Fig. 13. We use SPARQL [4], a query language which can select RDF triples from a triple set. Applications based on Jena API can search the movie-database and music-database individually by specifying a SPARQL query or can make a composite search on both databases at the same instant as shown in Fig.14. The output of this query is also shown at the end of the Fig. 14.


V. DESIGN OF DYNAMIC USER INTERFACE Tools for querying data have traditionally been text based , although graphical interfaces have been pursued. An Analysis of the text based queries had revealed a construction pattern that could be satisfied with a simple visual Interface. The pattern consisted of finding intersections of groups of items, where the items could be popular music and Movie information , and where the items of the intersection would be composers, singers and musicians.

PowerPoint Presentation:

These groups of items were often constrained by several other Parameters. This approach could not scale to a professional user who has to make hundreds of queries in a database. Therefore, we set out to build a Query UI that did not require the human user to become an expert of the system. The ongoing challenge is to allow a novice user to navigate the metadata and formulate the queries without much technical intervention.RDF vocabulary metadata navigation would be the first challenge for our users.

PowerPoint Presentation:

This vocabulary consists of properties about the data. For our prototype application, there are three Vocabularies such as  movie vocabulary  music vocabulary  iTunes vocabulary. Each vocabulary has its own nuances, which must be respected while doing queries, such as movie vocabularies with categories that can be used themselves for querying.

PowerPoint Presentation:

For example, if a user enjoyed and owned music by the violinist Joshua Bell and was curious to find other music by him, today they could type “Joshua Bell” as a query in and receive a long list of search results. Buried in the midst's of mostly CD titles is a DVD, The Red Violin, a movie in which Bell did not appear, but performed the haunting violin solos. The user could also do a search on Joshua Bell in and find The Red Violin among the search result items, but will not find any CD titles by Bell. In neither of these search examples will the user find: ∙ Music the user already owns by Bell ∙ Music the user’s friends own by Bell ∙ Joshua Bell music currently on sale by a a local commercial entity

PowerPoint Presentation:

The approach described in this paper, with extensions just slightly beyond the range of our initial work, could Provide such user discoveries. From the user’s perspective, the search would be conducted on the VMS, which would coordinate the necessary component searches among a number of DMS’s.

PowerPoint Presentation:

Simplicity in the interface is important for the reasons previously mentioned in this paper. A simple interface would make it possible for the user to manage a larger and wider range of digital media across devices. A sense of fun in the interface would encourage inquisitiveness, which in turn would drive down the level of user burden on performing searches. Our theory is that a sense of Putting aside issues of trust and security which are beyond the scope of this paper , we can consider under what circumstances the user needs to be made aware of the details of the search. Often the underlying details of the system would make no difference to the user. Users simply want their services to work and work well.

PowerPoint Presentation:

The user would be spared dealing with the complexities of the DMS searches. Our goals are to make the interface, simple, fun, and informative, but informative only to the level necessary for the user. More fun more searches he find more desired results he gets.

PowerPoint Presentation:

Similarly, when a user performs a search with the system proposed in this paper, They should be able to specify which general sources to search (i.e. which VMS). However. VMS access to specific DMS’s may be added or removed without the user’s knowledge. It is the job of the VMS to return the best results possible given the Resources available at the time. The user should not be burdened With these specifics, unless of course they ask for them or if their quality of service is effected. Unless the search represents a cost of time (for what ever reason the information requested is not immediately available) or a cost of money (information requested is available from a commercial source for a fee), the user need not be burdened.

Last Note::

Last Note: As the interface design work moves forward, these Are the guiding principles we employ to help us envision and implement this work. Given the early stage of the project, specifics on the design will be l eft for a later publication.


VI. CONCLUSION This paper presents an architecture for a digital home, which Includes a Virtual Media Server (VMS) and a Semantic Web Engine (SWE). A description to the basic communication between the different components in the described architecture and also present some examples where the presence of the SWE can enrich the quality of metadata-based multimedia search in the digital home.


REFERENCES [1] Digital Living Network Alliance; . [2] Jena Semantic Web Toolkit: [3] SemanticWorks2007: [4] SPARQL tutorial: [5] S. N. Murphy, V. Gainer, H. C. Chueh , “A Visual Interface Designed for Novice Users to find Research Patient Cohorts in a Large Biomedical Database”, AMIA Annu Symp Proc. 2003; 2003: 489-493. [6] Sadhna Ahuja , Tao Wu & Ora Lassila , “Using the Semantic Web to Enhance the Digital Living Experience”, IEEE CCNC 2006, Las Vegas (NV), January 2006 [7] Duane Degler , Scott Henninger and Lisa Battle, “Semantic Web HCI: Discussing Research Implications”, SIGCHI 2007