A New Approach for Knowledge Management and ... - Wseas

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria LunesuA New Approach for Knowledge Management and Optimization usingan Open Source RepositoryGIULIO CONCAS, FILIPPO EROS PANI, MARIA ILARIA LUNESUDIEE, Department of Electric and Electronic Engineering, Agile GroupUniversity of CagliariPiazza d’Armi, 09123 CagliariITALY{concas, filippo.pani, ilaria.lunesu}@diee.unica.it http://agile.diee.unica.itAbstract: The Institutional Repositories (IRs) based on Open Archives represent one of the main free accesstools for the results of scientific research, and their diffusion is continuously growing. In the context of the“Analytic Sound Archive of Sardinia” project, that aims to create an institutional archive with a linguisticallyannotated electronic corpus, this work proposes a new approach for management of knowledge using the toolDSpace (an open source software package developed in 2000 in the context of a joint project of theMassachusetts Institute of Technology with Hewlett-Packard): the purpose is to offer an original way toassociate linguistic annotations (information associated to specific text portions) to the corpus by treating themas metadata, so as to insert and manage them in the archive of choice after formalizing them in XML. Theformalization level of this approach allows for effective text retrievals through a metadata schema and easy,quick corpus interrogations, by formalizing linguistic annotation as a structured metadata schema. There is,thus, the need to have an efficient tool that could classify and store the vast amount of knowledge contained inan electronic corpus of spoken texts of Sardinian language, linguistically annotated at various levels, and thatcould allow a high usability in terms of ease of reference as well as ease of query and communication.Key-Words: Knowledge Management, Open Archive, Institutional Repository, Multimedia content, Metadata.1 IntroductionThe use of User Generated Content and publishingtools facilitate the circulation of resources on theInternet. This amount of information needsapproaches to gather information in an organized,reliable manner, describe it, store it, and retrieve it.All this content needs a minimum level ofinteroperability for tools and description parameters.In the field of scientific communication, that is whathappened with the creation of knowledgemanagement systems like DSpace [1], Eprints [2],etc., which are institutional archives modelled onthe Open Access Initiative that allow structuringinformation and adding standardized metadata to it[3-5]. The problem of content availability andorganization depends on two strictly related issues:availability of appropriate capabilities of indexingand retrieval inside knowledge management tools,and the ability of users to understand and use thesefeatures. Experience suggests that the solution ofthis problem can be found in the use of organizedschemas and relevant standardized metadata, or datatiers, that allow us to describe, classify, andorganize basic information, allowing retrieval anduse [6-9].In this paper we analyze the problem to definethe electronic corpus in an IR, formed by acollection of audio recordings from poetry contestsand singing performances in Sardinian language,stored and annotated on different linguistic levels.An electronic corpus is generally a homogeneouscollection of written or oral texts in digital format,processed with coherent criteria in order to build anempirical basis for language analysis. Its advantageis that it can be annotated by adding linguisticinformation in a specific portion of text.The repository will be compliant with OpenAccess Initiative (OAI), the corpus will be includedin an open IR, being therefore available forSardinian language scholars and everyone whowishes to use it.Linguists and musicologists, creators of thecorpus, needed to study and research the documentsin it, and they asked for the possibility to save theirwork in an readily available digital archive to store,E-ISSN: 2224-3402 139 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesuindex and manage it for both access andcommunication inside the scientific community.This needs was to save the information in anreadily available digital archive to store, index andmanage it for both access and communication insidethe scientific community. The purpose of this studyis to offer an original way to associate linguisticannotations (information associated to specific textportions) to the corpus by treating them as metadata,using an Open Source repository. In particular, anapplication profile has been created for the DublinCore (DC) metadata schema, which is suitable to thenature of the audio recordings in the Analytic SoundArchive of Sardinia (ASAS). In the second sectionof this paper we recall some aspects about theKnowledge Management. In the third we mentionsome typical use of Institutional Repositories , in thefourth, we present our proposed approach forknowledge formalization and management, in thefifth the case study about the use of DSpace and thesixth section includes the conclusion and reasoningabout the future evolution of the project.2 Knowledge ManagementOrganization and availability of contents in KMSsbasically depend on two factors: one is whetherKMSs have effective tools for information indexingand retrieval; the other is how. The solution to thisissue was found in the experience of the library andarchive industry, which have been dealing with theissues related to organization and collection ofinformation since way before the digital revolution.This experience suggested using metainformation,i.e. data used to describe and classify information,as a possible solution. The tools used to enter andmanage contents on the Internet must allow forentering and retrieving organized and relevantmetainformation, as metadata.2.1 MetadataMetadata have thus a fundamental role in organizingand managing digital resources, especially whenthere is a great quantity of available information thatmust be indexed and catalogued to facilitate searchand retrieval [10-12]. The selection of whichmetadata to use in describing a resource depends ona thorough observation of the characteristics,properties, common features, and differences in theinformational environment the source belongs to.A metadata schema is a set of structuredmetadata, developed for specific purposes in orderto establish a standard of metadata structure andterminology, and to associate different types ofmetadata. Every metadata schema includes adefinite number of elements, called metadataelements, each with its own meaning and purpose,i.e. describing the information resource [6-7].However, since standardization is the purpose, it isalways advisable to use largely used metadataschemas rather than creating new ones. Applicationprofiles are made of metadata sets derived fromdifferent schemas, and are aimed to create tools forparticular applications while keepinginteroperability with the original base schema. Thisprocedure and the application of common rules canmake different systems interoperable, like those inlibraries, museums and archives, making them ableto share a part of common metadata [13-14].2.2 The Dublin Core StandardA support to content management is offered by theDC metadata schema, which easily pairs up withother metadata schemas in the OAI architecture,improving granularity and refinement of theirstructures [15-17]. The rapid spreading of DC asmetadata schema was doubtlessly favoured by itsremarkable simplicity, thanks to which it couldadapt to many kinds of resources and usageenvironments. It is important, for a semantic modelused in resource discovery not to be dependent onthe format of the resource it needs to describe.In the latest years, DC was increasingly used inmany fields to describe, organize, manage, resourcesin possession of institutions and internationalorganizations, and also to support and provide addedvalue services, assuring a base format foraggregation and exchange of metadata collections,such as in the Open Archive Initiative, or asindispensable search tools in portals. The use of astandardized general classification system allows formetadata in such collections to be combined and forknowledge inside each collection to be shared [8-9].2.3 Linguistic Annotations and CorpusThe linguistics corpus studies great quantities oflinguistic productions, either spoken or written, byobserving their characteristics: lexicon, syntax,collocations, phonic chain, morphologic structures,etc. A corpus is any complete and orderly collectionE-ISSN: 2224-3402 140 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesuof written texts, by one or more authors, on a certaintopic, or, linguistically speaking, the sample of alanguage as examined in the description of the samelanguage. In order to exploit the wealth ofinformation stored in a corpus as linguistic data, thecorpus must be enriched with additionalinformation: linguistic annotations, i.e. the adding oflinguistic or metalinguistic information to differentportions of a text [18].3 Institutional RepositoriesSince the Nineties of the last century, a newphenomenon has affected the process ofscientific communication and knowledgesharing: the appearance and spread of digitalrepositories of scientific contributions in orderto make the movement of information more“agile”. In 1991, Paul Ginsparg, at Los AlamosNational Laboratory (USA), paves the way to thearXiv, a repository of works on Physics andMathematics [19]. In June 1994 Stevan Harnad senta “subversive” proposal to the mailing list of theVirginia Polytechnic Institute: they ought to sharetheir ideas through the contributions of selfarchivingon the internet, in order to communicatetheir results more effectively. A new kind of openarchive begins to emerge: the IR, supported andmanaged by an institution, such as an university,which incorporates the contributions of itsresearchers. In October 1999 a group of researchersand librarians in Santa Fe (USA) marked the turningpoint: the rise of the OAI, essential to themanagement of technical aspects such as protocolsand data exchange standards, localization andsubsequent retrieval of scientific contribution, andsoftware such as operating tools and for indexing.At the beginning of the new century, when thearchives have already opened and operational, theexpression “Open Access” is used for the first timein a public document: Budapest Open AccessInitiative Manifesto (2002). It suggested for the firsttime to adopt both strategies, called“complementary”, to encourage the spread of theopen access system: the "self-archiving", i.e.archiving in institutional and disciplinary "openelectronic archives", of articles by researcher and"open access journals", the new generation ofscientific open access journals.During an industrial project aimed to the creationof the Analytic Sound Archive of Sardinia, the ideato create an Institutional Archive to solve theproblems of organization and availability ofinformation came forth. There was, thus, the need tohave an efficient tool that could classify and storethe vast amount of knowledge contained in anelectronic corpus of Sardinian language, and thatcould, at the same time, allow a high usability interms of ease of reference as well as ease of queryand communication.3.1 Formalization of KnowledgeIn this context, knowledge is represented not onlyby the texts of the corpus, but especially by the metalanguage and linguistic annotations that enhancethem.For each audio clip, a set of metainformationdescribing the content is needed in order to enablethe search and retrieval of data by local author, title,date of recording, to more particular features likelinguistic variety or singing type. Each audio clip isalso enhanced by a set of linguistic annotations. Theinsertion of the audio clips in the chosen KMSrequired the formalization of all the associatedmetainformation in the form of a structured set ofmetadata.Linguists and musicologists working on theSardinian Linguistic Sound Archive chose a list ofpossible annotation levels (syllable, tone,morpheme, syntagm, accents, etc.), useful for bothlinguistic and musical analysis of audio recordings.4 Proposed approachOur proposed approach for knowledge formalizationand management, gathered in an annotatedelectronic corpus in an IR based on the OAI model,will be described below.4.1 Formalization of Metadata SchemasIn order to manage and organize the informationthat makes up the corpus, KMSs associate organizedand relevant information to a text when it is entered.Metadata schemas mirror the complex nature of dataand are often strongly structured and hierarchical,including many kinds of metadata, with manydifferent functions.Building an effective system of structuredmetadata means creating a conceptual model toformalize and model the essential semanticcharacteristics of a knowledge domain.After designing the conceptual model of theknowledge domain, a top-down approach can beused for structuring the metadata schema. If theknowledge domain is made of an electronic corpusE-ISSN: 2224-3402 141 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesuand its objects are its texts, essential metadata(author, title, language, publishing date, etc.) mustbe deducted and formalized from their semanticcharacteristics. Some of those metadata may befurther specified according to a hierarchicalstructure: for example, the metadata “author” maybefurther refined as “main author”, “illustrator” and,"curator".4.2 Formalization of Linguistic AnnotationsThe information in the corpus are organized asinformal annotations and the most efficient way touse their information is formalizing this annotationthrough metadata schemas. In this way, not onlyannotations can be associated to their texts, but theycan also be used as search parameters for findingtexts.Linguistic annotations created with specialsoftware, like PRAAT [20] for audio files, aregenerally stored in a semi-structured manner. Infact, each annotation is distinctly represented insidethe file, according to a defined, repetitive structurewhere the annotation texts is paired with the instantor the time interval it refers to. Moreover, thebelonging of each annotation to a certain linguisticlevel is clearly stated in the file. The formalizationof annotations in a metadata schema can beachieved using a bottom-up or inductive reasoning.Starting with the analysis of the structure of eachannotations in the file and applying inductive logic,a "category" is abstracted from every linguisticlevel. This formalization allows for easily codingand representing of annotations though markuplanguages like XML, because their structure can bedescribed with tags or markers, for metadata andtheir qualifiers, inside which a linguistic label isfound. All annotations in the same linguistic level,e.g. phonetics, can be formalized in the XML asdifferent occurrences of the same metadata called"phoneme", whose value can be made up of twoterms: linguistic label and eventually time interval.4.3 Choosing a Metadata SchemaThe use of both a deductive and an inductiveapproach allows metainformation and linguisticannotations to be formalized in a single structuredmetadata schema. Entering metadata in a knowledgemanagement system requires the selection of anoperational criterion based on the particular needsthe system has to work with.Most archives use Qualified Dublin Core as mainschema for indexing and displaying metadata andSimple Dublin Core to show them through the OAI-PMH standard. There are four main criteria forchoosing a metadata schema, with differentapproaches in metadata organization: 1) mapping ofnative metadata on existing DC elements; 2)mapping of native metadata on DC elements andcreation of new customized qualifiers for DCelements; 3) creation of a customized metadataschema, identical to the native metadata set; 4)creation of DC metadata records as abstraction ofnative metadata records and entering of the latter asattachments to the resource. Out of the criteriamentioned above, the first one is the leastsatisfactory for preservation and reuse of descriptivemetadata of resources, while the third one is themost preserving of the integrity and granularity oforiginal metadata but needs great efforts for thecreation of a customized metadata schema, togetherwith high maintenance costs for the archive. Thesecond and fourth criteria combine preservation andgranularity needs with archive management costsbetter than the other two. Choosing between themdepends solely upon the particular requirements ofthe archive.Once the decision on which criterion to use issettled, the archive must be configured so that it iscompatible with the approach of choice for metadatamanagement. In particular, if the second criterion isadopted, the DC schema must be updated with new,customized qualifiers; if the third criterion ischosen, the entire metadata schema created ad hocmust be entered into the system. In this way,customized metadata and qualifiers can be used todescribe texts of the corpus inside the archive.Generally, metadata schemas can be configuredthrough the user interface of the archive. However,schemas rich in elements and qualifiers are betterconfigured with the import tools provided bymanagement systems, after having encoded themwith the XML markup language. XML is used byarchives to manage the import-export of metadata.Compilation of metadata records associated totexts in the corpus may be usually done with either auser interface or with batch import tools. Instead,when big quantities of metadata need to beassociated to one resource, like with linguisticannotations, there are specific batch import toolsthat require the specification of all metadata asattribute-value pairs, coded in an XML file.E-ISSN: 2224-3402 142 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesu5 Case studyThe ASAS project (http://asas.flosslab.it) aims tocreate an IR with an annotated spoken languageelectronic corpus that could become a platform forthe preservation, study, communication andappreciation of oral traditions of the Sardinianlanguage, especially improvised poetry.5.1 Annotations through PRAATThe electronic corpus was annotated by linguistsand musicologists through the PRAAT software[19], which, besides performing spoken languageanalysis, allows for multilevel segmentation andlinguistic annotations of audio files. The softwarehas a graphic interface with waveforms and voicespectrum that make annotators' work easier andmake visible those acoustic phenomena that can befound by an accurate spectrum analysis, followed byannotation levels. Linguists and musicologistsworking on the Sardinian Linguistic Sound Archivechose a list of possible annotation levels (syllable,tone, morpheme, syntagm, accents, etc.), useful forboth linguistic and musical analysis of audiorecordings.5.2 Metainformation Associated to AudioRecordingsMusicologists and Linguists, other than withannotations, wanted to complete every audiorecording by describing it with a number ofinformation, chosen among the most relevantfeatures of the recordings. The information could beused to manage recordings in the archive, becauseby describing them they allow for selection andorganization, facilitating efficient retrieval andusage. Metainformation range from somethingclosely related to cataloguing, like author, title,object, recording date, etc., up to more technicalinformation like the different singing types, speechtypes, accompaniment or instruments. Linguists andmusicologists selected 38 metainformationsassociated to audio recordings: title, author, object,description, format, etc.characteristics of texts. Through a continuousdialogue with the scholars, audio recordings wereanalysed for their essential and basic properties,needed to organize and retrieve texts in the corpus.Twelve general metadata were found: title, author,publisher, object, contributor, date, place, occasion,document accessibility, language, description andformat. Those metadata outlined the necessaryinformation to describe spoken texts in the corpus,conveying in particular singing or speech type, theoccasion in which the audio was recorded, and thelinguistic variety it belongs to.The top-down approach proceeds to furtherspecialize the metadata. More specific, or qualified,metadata are represented by adding a qualifier to thename of the more general metadata and using thecommon syntax metadata.qualifier.Lastly, "relational" metadata are defined as well,in order to define a certain relation among two ormore different objects belonging to the corpus. Aninclusion relation must be specified in order todescribe the belonging of one or more objects to thesame recording set, for example different songs in asinging contest.5.4 Formalization of Linguistic Annotations:Bottom-Up ApproachThe formalization of annotations in a metadataschema can be achieved using a bottom-up orinductive reasoning, as explained in the previoussection. The structure of annotations is analysedwith the PRAAT software. Annotations areorganized with a precise structure: each annotationis made of a time interval and a text label or by aninstant and a marker with its text. All annotations inthe same linguistic category are collected in thesame tier (or annotation level), which can beconsidered as the category they belong to, giving itsname to the corresponding metadata. In this way, arepeatable metadata is found in each annotationlevel of the TextGrid (the text file where PRAATstores all Tier with their own segmentations andannotations) and each annotation can be representedas multiple occurrences of that metadata.5.3 Formalization of SemanticCharacteristics: Top-Down ApproachAfter designing the conceptual model of theknowledge domain, a top-down or deductiveapproach can be used for formalizing the semantic5.5 Choosing a Metadata Schema for KMSEnteringDepending on the interoperability needs that mustbe met, importing the metadata schema that was justcreated into the knowledge management systemE-ISSN: 2224-3402 143 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesumay not be appropriate or convenient. Most archivesuse Qualified DC as main schema for indexing anddisplaying metadata and Simple DC to show themthrough the OAI-PMH standard. Therefore, theadoption of Dublin Core must be thoroughlyevaluated when an archive is needed to be compliantwith the interoperability principles required by OAI.Our of the four criteria listed in section 4.3, the mostsuitable technique for the case study is an hybridmodel between the second (mapping of nativemetadata on DC elements and creation of newcustomized qualifiers for DC elements) and the thirdone (creation of a customized metadata schema,identical to the native metadata set). The thirdcriterion is more convenient for linguisticannotations, so that a dedicated metadata schemacan be created to preserve their granularity; whilethe second criterion is best suited for all othermetadata, because it combines the advantages ofgranularity as provided by qualifiers tointeroperability provided by DC metadata.5.6 Application Profile for the AnalyticalSound Archive of SardiniaIn creating a specific application profile for theASAS, a "conservative" approach was used towardsthe original Qualified DC elements and qualifiers inorder to use as many of them as possible for theformalization of descriptive and relational metadata.A special schema, identified by the prefix "asas",was created instead for annotations. Its metadatawere entered into the DC application profile asoutlined below.Metainformationor ASASAnnotationTitleAuthorPublisherObjectDescriptionContributorAnnotatorLocationDateOccasionSourceDocumentAccessibilityDC Application Profile Metadatadc.titledc.creatordc.publisherdc.typedc.type.categorydc.contributordc.contributor.annotatoredc.coverage.spatialdc.date.createddc.subjectdc.relation.isbasedondc.rightsPerformerdc.contributor.sperakerPerformerPerformer's Age dc.description.speakerPerformerPerformer's Placeof OriginLanguageSourceCompletenessSource No.Source SectionNo.Document TypeFormatAcquisitionMethodReading TypeInterview TypeMonody TypeUnison /HeterophonyAccompanimentTypedc.description.speakerPerformerdc.languagedc.description.integritàdc.relation.ispartofseriesdc.relation.ispartofseriesdc.format.audioVideodc.format.mediumdc.format.modoAcquisizionedc.type.letturadc.type.intervistadc.type.monodiadc.type.unisonoEterofoniaPolyphony Type dc.type.polifoniaInstrumentalInstrumentSinging TypeOtherSyllableToneMorphemePhoneWordPart of SpeechSyntagmSentenceInformationStructureTurnPerfdc.type.monodiaAccompagnamentodc.type.strumentaledc.type.strumentodc.type.tipoCantodc.descriptionasas.annotazione.sillabaasas.annotazione.toniasas.annotazione.morfemaasas.annotazione.fonoasas.annotazione.parolaasas.annotazione.posasas.annotazione.sintagmaasas.annotazione.fraseasas.annotazione.strutturaInformativaasas.annotazione.turnPerfMusical Syllable asas.annotazione.sillabaMusicaleMetric Segmentasas.annotazione.segmentoMetricoMusical Segment asas.annotazione.segmentoMusicaleTonal CentreNotationOrnamentationAccentsMelismaticSyllableADD1asas.annotazione.centroTonaleasas.annotazione.notazioneasas.annotazione.ornamentazioneasas.annotazione.accentiasas.annotazione.sillabaMelismaticaasas.annotazione.annotazioneLiberaTable 1: Application profile for the ASASE-ISSN: 2224-3402 144 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria LunesuThe next step is to enter metadata in theknowledge management system: oncemetainformation have been organized andstructured, the KMS is configured so that it can beadapted to the selected metadata schema.5.7 Choice and Customization of the KMSDSpace, an open source software packagedeveloped in 2000 in the context of a joint project ofthe Massachussetts Institute of Technology withHewlett-Packard, provides all the necessary toolsfor creation and management of an IR based on theOpen Access model [1]. Such an IR can collect,store, index, preserve and make accessible theinformation output created by universities andresearch institutes in a digital format.DSpace is designed as a central storage facilityable to collect all kinds of content from thecommunity relating to the institution through a userinterface as simple and intuitive as possible. It cancollect various types of digital resources includingtext, images, video, audio, articles and preprints,technical reports, working papers, datasets, andlearning objects directly from the creators.DSpace was chosen to realize the AnalyticSound Archive of Sardinia as it fulfills all therequirements asked by linguists and musicologists.It is in fact completely customizable, supportsnatively Qualified DC metadata schema and iscompatible with OAI with the support of OAI-PMH.The proposed approach allows to insert the corpusand the associated knowledge inside of DSpace,ensuring the maintenance of its structure and theability to interrogate and update it easily by addingor modifying its contents. Each text of the corpusis inserted into a DSpace item so that it can beuniquely associated with all of the metadata neededfor the linguistic analysis. The audio file containsthe registrations and the original files with theannotations are loaded inside of the item as abitstream, while the metadata are stored in thesystem database.The first step consisted in the insertion of thecustomization of new qualifiers for the Dublin Coredescriptive metadata representation and a newscheme called "asas" for the representation of theannotations. When inserting the corpus into DSpaceit was decided to create a specific item for each ofaudio clip. It was therefore necessary to set therelease wizard offered by DSpace by changing thespecific XML file responsible for entry forms(input-forms.xml). The descriptive metadata,identified by researchers, such as title, author, typeof song, instrument, etc., and all metadatacorresponding to linguistic annotations (phono,morpheme, word, etc.), was associated to each item,together with the original file containing the audiorecording and the original file of annotations.Figure 1: Customization of DSpace metadata'sRegisterAfter the insertion of metadata, the interface wascustomized by replacing the standard formsprovided by DSpace using modules specficallydesigned to allow the creation of items and therelease of DC metadata according to the specificneeds of the project. The metadata on theannotations were inserted instead using directimport because the high number of occurrences foreach item made it difficult to enter them manually,as shown by Hillman and Westbrooks [21].Finally, we proceeded to customize the searchinterface of DSpace in order to adapt it to newmetadata and to the particular needs of the AnalyticSound Archive of Sardinia. In essence, all metadatacorresponding to linguistic annotations needed to beindexed in DSpace’s search engine so that we couldfind a certain audio clip even through the search ofan associated record. Furthermore, some descriptivemetadata such as location, type of performer andcontribution were indexed to allow effectivesearching that exploited the granularity of themetadata.5.7.1 Metadata SchemasThe metadata are stored and managed by DSpacethrough a special tool, the Metadata Registry, wherethe Qualifed Dublin Core schema is configured bydefault. It can nevertheless be changed, and newcustomized schemas can be added. The systemoffers two ways to configure the register: one is thegraphic interface named Manakin, and the other canE-ISSN: 2224-3402 145 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesube used by the terminal. Each of them has a specificpurpose. The first method allows an authorized userto act on the diagrams through an easy and intuitiveweb interface. Once you create a schema, themetadata can be added one at a time, with anyqualifiers and related notes. This feature is crucialfor the updating and maintenance of the system as itcan make adjustments quickly and easily withoutthe intervention of a computer expert. Likewise, youcan choose the second solution where using aspecific command, the metadata schema expressedin XML can be imported in the register according toa specific syntax.During the creation of the Sound Archive,metadata was to be gradually defined and refined bylinguists and musicologists, so it was decided toinsert and manage it through the DSpace webinterface. We obtained a customized Qualified DCschema with new specific qualifiers and a newschema “asas” with the metadata to use for theinsertion of linguistic annotations, as you can see inthe Figure 2.Figure 2: The new schema “asas”Once we reached the final version of the scheme,however, it was decided to formalize the metadata inXML so that should the system be reinstalled ortransferred, the register could be quickly configuredvia the import metadata command.5.7.2 Insertion of Metadata in the KMSDSpace is preset to use, during the phase of theinsertion of an item, the Qualified Dublin Coremetadata schema but at the same time allows tocustomize it, or to create new metadata schemasaccording to the users and their requirements. Thefirst step in order to submit the metadata in DSpacewas to organize and structure them as metadataschemas. The second step was, instead, to configurethe knowledge management system so that it couldbe adapted to the particular chosen metadataschema.Figure 3: Interface personalization input metadataThe Manakin graphic interface offers a tool tomanage the Metadata Register that allows tocustomize the preconfigured Dublin Core schema ina rapid and intuitive way. In the Metadata Registercan be entered all new qualifiers with their relatedcomments to obtain the desired application profile,which coexist in the Qualified Dublin Core standardwith its qualifiers, qualifiers and made for thespecific project, Analytical Sound Archive ofSardinia. In DSpace this entity is called “item” andis "built" with a wizard that, using the preconfiguredmodules, allows to specify the values ofthe meta-information to be formalized as metadata.These modules are ready for the insertion of themetadata belonging to the Qualified Dublin Coreschema, but in the case study they were fullycustomized to fit the new specific qualifiers for thearchive.5.7.3 Research ModulesDSpace offers a powerful search interface that isconfigured to use the main DC metadata (like title,author, language) or the free full-text research asparameter for the search. The researchers asked forspecific search criteria, besides the standard ones, soyou can search for audio clips by place of recording,the participation to a particular event, but also theannotations they contain. The last criterion, inparticular, is the most important function for thestudy of the language corpus as it allows to performstatistical analysis on the text easily and quickly,without the need to use specific and complicatedsoftware interrogation. The search indexes weremodified so that all needed metadata were selectedas criteria in the search interface and informationlike place of recording, performers’ information, thenumber indicating the event place and all metadataE-ISSN: 2224-3402 146 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesucorresponding to the levels of annotation werespecifically added.6 CONCLUSIONSThe purpose of this work was to offer a newapproach to formalization and management ofknowledge represented by a set of audio recordingsbelonging to a corpus plus the linguistic informationadded to the same corpus with annotations. Theapproach was applied to formalize knowledge in theASAS, a joint project by linguists and musicologistsat University of Cagliari. The project aimed topresent a study on improvised poetry in Sardinianlanguage, using an electronic corpus they createdand annotated. In order to make the resourcesopenly accessible through the Internet, as per ouraim, we used DSpace as a specific tool to organizeand manage a big quantity of information comingfrom the audio recording. We have chosen this toolbecause after some investigations we found thatmore Universities and research Institutes (i. e.Brunel University, Cornell University andMassachusetts Institute of Technology) use DSpace,in fact it is a very efficient tool easy to use,customizable and flexible to allow the management,the classification and the storage of a vast amount ofknowledge contained in an electronic corpus ofSardinian language, and that could, at the sametime, allow a high usability in terms of ease ofreference as well as ease of query andcommunication. The formalization of a structuredmetadata schema was reached through the creationof an application profile for the Qualified DublinCore metadata schema, where customized qualifierswere added to the standard elements and qualifiers.Metadata in non-standard schemas could then bebetter represented. Linguistic annotations wereformalized as well through a metadata schema.Corpus interrogation was thus made easier andquicker, since it used the knowledge managementsystem's search tool.This work leaves space for future research onways to improve the service. A dedicated website orthe integration of this system in an institutionalportal through an exploration interface would beparticularly interesting. Another feature that couldbe implemented may be a virtual map whererecordings can be explored by geographic location.In future works we want to use the sameapproach for knowledge formalization andmanagement as the one represented in a set ofscientific papers [22-36] on different topics ofSoftware Engineering. Knowledge in papers isusually limited to keyword management; however,we think this approach could be effectively used forother types of knowledge as well.References:[1] DSpace, http://www.dspace.org[2] EPrints, http://www.eprints.org[3] Tansley R., Bass, M., Stuve, D., Branschofsky,M., Chudnov, D., McClellan, G. and Smith, M.,The DSpace Institutional Digital RepositorySystem: Current Functionality, JCDL '03Proceedings of the 3rd ACM/IEEE-CS jointconference on Digital libraries, 2003.[4] Lynch, C., Institutional repositories: essentialinfrastructure for scholarship in the digital age,Association of Research Libraries: a bimonthlyreport, no. 226, 2003.[5] Swan, A. and Carr, L., Institutions, theirrepositories and the Web. Serials Review,2008, p. 31. ttp://eprints.ecs.soton.ac.uk/14965[6] Heery, R. and Patel, M., Application profiles:mixing and matching metadata schemas,Ariadne, 2000.[7] Lagoze, C. and Van de Sompel, H., The makingof the Open Archives Initiative protocol formetadata harvesting, Library Hi Tech, 2003.[8] Lunesu, M. I., Pani, F. E. and Concas, G., Anapproach to manage semantic informationsfrom UGC, International Conference onKnowledge Engineering and OntologyDevelopment (KEOD), 2011.[9] Lunesu, M. I., Pani, F. E. and Concas, G.,Using a standards-based approach for amultimedia knowledge-base, InternationalConference on Knowledge Management andInformation Sharing (KMIS), 2011.[10] Chopey, M. A., Planning and Implementing aMetadata-Driven Digital Repository, HaworthPress Inc., 2005.[11] Dunsire, G., Collecting metadata frominstitutional repositories, OCLC Systems &Services, Vol. 24, No. 1, 2008, pp. 51-58.[12] Solodovnik, I., Metadata issues in DigitalLibraries: key concepts and perspectives,Italian Journal of Library and InformationScience, Vol. 2, No. 2, 2011.[13] Pani, F. E., Lunesu, M. I., and Concas, G.,Optimization of Knowledge Availability in anInstitutional Repository, InternationalConference on Knowledge Engineering andOntology Development (KEOD), 2012.E-ISSN: 2224-3402 147 Issue 5, Volume 10, May 2013

WSEAS TRANSACTIONS onINFORMATION SCIENCE and APPLICATIONSGiulio Concas, Filippo Eros Pani, Maria Ilaria Lunesu[14] Pani, F. E., Lunesu, M. I., and Concas, G.,Knowledge Formalization and Management inKMS, International Conference on KnowledgeManagement and Information Sharing (KMIS),2012.[15] Hutt, A. and Riley, J., Semantics and Syntax ofDublin Core Usage in Open Archives InitiativeData, Joint Conference on Digital Libraries,ACM Press, 2005.[16] Jackson, A. S., Han, M. J., Groetsch, K. andMustafoff, M. (2008). Dublin Core MetadataHarvested Through OAI-PMH. In Proceedingsof the 5th ACM/IEEE-CS joint conference onDigital libraries.[17] Hillmann, D. I., Using Dublin Core, DublinCore Metadata Initiative Recommendation,2005.[18] Llisterri, J., Text Corpora Working GroupReading Guide, EAGLES (Expert AdvisoryGroup on language Engineering Standards)Document EAG-TCWG-FR-2, CNR, Istituto diLinguistica computazionale, 1996.[19] arXiv (Cornell University Library),http://arxiv.org/[20] PRAAT, http://www.fon.hum.uva.nl/praat/[21] Hillman, D. I. and Westbrooks, E. L., Metadatain practice, American Library Association,2004.[22] Concas, G., Marchesi, M., Murgia, A., Pinna,S., Tonelli, R., Assessing traditional and newmetrics for object-oriented systems. InProceedings of the 2010 ICSE, Workshop onEmerging Trends in Software Metrics(WETSoM ’10), 2010, pp. 24-31.[23] Melis, M., Turnu, I., Cau, A., Concas, G.Evaluating the impact of test-first programmingand pair programming through softwareprocess simulation. In Software ProcessImprovement and Practice 11 (4), 2006, pp.345-360.[24] Concas, G., Marchesi, M., Murgia, A., Tonelli,R., An empirical study of social networksmetrics in object oriented software. Advancesin Software Engineering, Vol. 2010 (ID:729826).[25] Concas, G., Marchesi, M., Murgia, A., Tonelli,R., Turnu, I., On the distribution of bugs in theEclipse system, IEEE Transactions on SoftwareEngineering 37 (6), art. no. 5928349, 2011, pp.872-877.[26] Turnu, I., Concas, G., Marchesi, M., Pinna, S.,Tonelli, R., A modified Yule process to modelthe evolution of some object-oriented systemproperties, Information Sciences 181 (4) ,2011,pp. 883-902.[27] Locci, M., Concas, G., Tonelli, R., Turnu, I.,Three algorithms for analyzing fractal softwarenetworks, WSEAS Transactions on InformationScience and Applications 7 (3) , 2010, pp. 371-380.[28] Tonelli, R., Concas, G., Locci, M., Threeefficient algorithms for implementing thepreferential attachment mechanism in Yule-Simon Stochastic Process, WSEASTransactions on Information Science andApplications 7 (2), 2010, pp. 176-185.[29] Concas, G., Lisci, M., Pinna, S., Porruvecchio,G., Uras, S., Open source communities associal networks: An analysis of some peculiarcharacteristics, Proceedings of the AustralianSoftware Engineering Conference (ASWEC),art. no. 4483227, 2008, pp. 387-391.[30] Turnu, I., Melis, M., Cau, A., Setzu, A., Concas,G., Mannaro, K., Modeling and simulation ofopen source development using an agilepractice, Journal of Systems Architecture 52(11), 2006, pp. 610-618.[31] Destefanis, G., Tonelli, R., Concas, G.,Marchesi, M., An analysis of anti-micropatternseffects on fault-proneness in large Javasystems, ACM Symposium on AppliedComputing, 2012, pp. 1251-1253.[32] Murgia, A., Tonelli, R., Marchesi, M., Concas,G., Counsell, S., McFall, J., Swift, S.,Refactoring and its relationship with fan-in andfan-out: An empirical study, The EuropeanConference on Software Maintenance andReengineering (CSMR), art. no. 6178854, 2012,pp. 63-72.[33] Concas, G., Marchesi, M., Murgia, A., Pinna,S., Tonelli, R Assessing traditional and newmetrics for object-oriented systems.International Conference on SoftwareEngineering, 2010, pp. 24-31.[34] Turnu, I., Marchesi, M., Tonelli, R. Entropy ofthe degree distribution and object-orientedsoftware quality, 3rd International Workshopon Emerging Trends in Software Metrics(WETSoM), art. no. 6226997, 2012, pp. 77-82[35] Concas, G., Marchesi, M., Murgia, A., Tonelli,R., Turnu, I., On the distribution of bugs in theEclipse system, IEEE Transactions on SoftwareEngineering 37 (6), art. no. 5928349, 2011, pp.872-877.[36] Concas, G., Marchesi, M., Destefanis, G.,Tonelli, R. An empirical study of softwaremetrics for assessing the phases of an agileproject, International Journal of SoftwareEngineering and Knowledge Engineering 22,2012, pp.525-54.E-ISSN: 2224-3402 148 Issue 5, Volume 10, May 2013