Intelligent database systems

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/220688503 Intelligent Database Systems Book · January 2001 Source: DBLP CITATIONS READS 65 1,368 3 authors, including: Elisa Bertino Barbara Catania 1,100 PUBLICATIONS 21,965 CITATIONS 111 PUBLICATIONS 1,446 CITATIONS Purdue University SEE PROFILE Università degli Studi di Genova SEE PROFILE Some of the authors of this publication are also working on these related projects: resource allocation and intrusion detection for Internet of Things View project All content following this page was uploaded by Gian Piero Zarri on 24 December 2013. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. Intelligent Database Systems Barbara Catania University of Genova (Italy) IIWAS 2001 - Linz, Austria Barbara Catania Pag. 1 IIWAS 2001 - Linz, Austria Outline • Introduction to Intelligent Database Systems (IDBs) • Fundamental IDB approaches • IDBs and their role in Web applications • An IDB approach for metadata representation and retrieval • Conclusions • Bibliography Barbara Catania Pag. 2 IIWAS 2001 - Linz, Austria Introduction Barbara Catania Pag. 3 IIWAS 2001 - Linz, Austria What is an IDB? DB technology: • limited modeling capabilities • new data management applications AI techniques: • often toy systems • no persistent management of data DB techniques can aid an AI system to deal with large amount of IDB Technology information Barbara Catania Pag. 4 AI techniques can provide semantic support to a DB system IIWAS 2001 - Linz, Austria Characteristics of IDBs • Architecture based (at least implicitly) on an organization in the Expert Systems (ESs) style – Fact DataBase (FDB) + Rule Base (RLB) • Use of AI techniques – Knowledge representation techniques • semantic data representation – Inference techniques • improved reasoning about data – Intelligent user interfaces • help users to make requests and receive replies • persistency of the FDB Barbara Catania Pag. 5 IIWAS 2001 - Linz, Austria A traditional taxonomy of IDBs Efforts originated in a DB context Static extensions Dynamic extensions extending the expressive power of traditional DB data models introducing some form of reasoning inside DBMSs Efforts originated in a AI context Basic solutions Advanced solutions coupling knowledge-based systems and DBMSs attempt to use AI systems to deal directly with large amount of information Barbara Catania Pag. 6 IIWAS 2001 - Linz, Austria Fundamental IDB approaches Barbara Catania Pag. 7 IIWAS 2001 - Linz, Austria Efforts originating in a DB context Relational model Dynamic extensions OODBMS Active DBMS 75 80 85 90 95 00 UML Semantic models Static extensions Barbara Catania Nested model OMT Hyper-semantic models Pag. 8 IIWAS 2001 - Linz, Austria Static extensions Extensions of the relational model Conceptual models semantic data models Nested relational model Hyper-semantic data models OO data models relational functional Semantic networks KL-ONE - AI Barbara Catania Pag. 9 IIWAS 2001 - Linz, Austria Dynamic extensions Introducing programming languages constructs Introducing active rules Active DBMS OODBMS Frame-based systems - AI semantic data models Barbara Catania Pag. 10 Production rules - AI IIWAS 2001 - Linz, Austria Efforts originating in an AI context • Based on the notion of Knowledge Based system (KBS) • KBSs typically contain: – explicitly represented rules RLB – simple facts FDB – components which can make inferences over the Knowledge Base KB = RLB + FDB • the information dealt with by the KBS consists therefore of: – explicitly stored facts and rules – derived facts Barbara Catania Pag. 11 IIWAS 2001 - Linz, Austria Knowledge Based System types • Pure rule-based representations supporting inference by resolution – systems developed in a logic programming context – ES shells based only on the production rule paradigm • Pure frame- or object-based representations supporting inference by inheritance – frame systems – terminological (description logic) systems – KESE: hybrid systems, commercially available, supporting alternative inference methods and representation schemes (SPOKE, KEE) Barbara Catania Pag. 12 IIWAS 2001 - Linz, Austria An overall view KBS type Logic programming systems Expert Systems (ES) Inheritance-based systems Formalism logical clauses production rules ontology of concepts, inheritance inheritance hierarchies Frame-based systems structured concepts Terminological systems terminological knowledge, assertional classification knowledge, based on description logic KESE Barbara Catania Reasoning resolution principle procedural, logic various approaches Pag. 13 inheritance various approaches IIWAS 2001 - Linz, Austria Tradeoff • Tradeoff between: computational complexity expressive power, and completeness – sound, complete, tractable but limited expressive power • KRYPTION, CLASSIC – sound, complete, intractable • KRISL – sound, higher expressive power, intractable and incomplete (more efficient) • BACK,NIKL, LOOM Barbara Catania Pag. 14 IIWAS 2001 - Linz, Austria KBS and DBMS • Conventional KBSs are inadequate for supporting new data/knowledge-intensive applications • Problems: – KBSs usually deal with knowledge bases of small size, in volatile memory – KBSs provide only limited DBMS services • Possible solution: coupling KBSs with DBMSs – Coupling of logic programming systems with DBMSs • Deductive databases – coupling of ES shells and KESEs wih DBMSs • five classes of approaches Barbara Catania Pag. 15 IIWAS 2001 - Linz, Austria Deductive databases • Intensional database (IDB), containing logic formulas – IDB: parent(X,Y) <- father(X,Y) parent(X,Y) <- father(X,Z), parent(Z,Y) • Extensional database (EDB), containing base relations – EDB: father(ann,john)father(john, mark) • Through logic inference mechanisms, derive, from base relations, information not explicitly stored in the EDB – father(ann,john), father(john, mark), – parent(ann,john), parent(john, mark), parent(ann,mark) • Language typically used for IDB: – Datalog (restriction of Prolog, set-oriented) • formal theoretical foundation Barbara Catania Pag. 16 IIWAS 2001 - Linz, Austria Barbara Catania Pag. 17 IIWAS 2001 - Linz, Austria Coupling ES shells and KESEs with DBMSs • No theoretical foundations • mismatch between ES shell/KESEs and DBMS • semantic, impedence, and granularity mismatch • most proposals for KESEs does not give rise to real IDBs – useful for hystorical motivations – they represent the basic approaches of IDB architecture • DB used to store AI objects: – AI objects are translated into and out of DB objects – AI objects are stored in their native format in the DB (for example, as LOB) Barbara Catania Pag. 18 IIWAS 2001 - Linz, Austria Coupling ES shells and KESEs with DBMSs • Full-bridge solution • often flat file as intermediate medium • the control of the interactions and the processing can be located on the central bridge or distributed • Such architecture does not scale up well • Examples: – DIFEAD (ESs, rel. DBMSs, intermediate data dictionary) – KADBASE (ESs, rel. DBMSs, distributed) – Europe-Bruke approach (BACK, rel. DBMSs) Barbara Catania Pag. 19 IIWAS 2001 - Linz, Austria Coupling ES shells and KESEs with DBMSs • Extension of a KB with components proper to a DBMS • used mainly for KBs based on the logical approach • adopted by the vendors of the main ES tools to provide their systems with some elementary possibilities of extracting information from a database • Examples: ROCK, KBMS, SPOKE • no standard approach exists for realizing the access functions Barbara Catania Pag. 20 IIWAS 2001 - Linz, Austria Coupling ES shells and KESEs with DBMSs • Extension of a DBMS with components proper to a KBS • Two possible interactions: – explicit access procedure: an explicit call to the KBS is inserted in the application program – implicit access procedure: the access to the inference engine is through the same query interface used to access data • Similar to rule based systems and OODBMSs Barbara Catania Pag. 21 IIWAS 2001 - Linz, Austria Coupling ES shells and KESEs with DBMSs • DB and KBS systems are strongly integrated – only one environment – no semantic mismatch problems • Architecture d): – construction of a DB system after (or during) the set up of the KBS – integral approaches • Architecture e): – the DBMS technology is more stable and mature than the KBS technology, and the installed base of DBs is definitely larger than the KBSs base – DBs are probably a better place for incorporating ES functionalities than vice versa – Examples: ARCHES, KBase Barbara Catania Pag. 22 IIWAS 2001 - Linz, Austria Integral approach • Example of type d) architecture: – – – – only one pure KBS environment data model is some sort of AI knowledge representation language all sort of inference techniques are used KBS environment should be able to support DBMS services • attempt to use some kind of AI system to deal directly, in a DBMS style, with large quantities of persistent information • no theoretical foundation • Example: TELOS, CYC, NKRL, lexical approaches (WordNet) • Limitations: – great variety of knowledge representation models – complexity of the used formalisms – lack in supporting DBMS functionalities Barbara Catania Pag. 23 IIWAS 2001 - Linz, Austria IDBs and their role in Web applications Barbara Catania Pag. 24 IIWAS 2001 - Linz, Austria Some applications • Metadata representation • Integration of heterogeneous sources • Web application design Barbara Catania Pag. 25 IIWAS 2001 - Linz, Austria Metadata representation: problem ! ! Barbara Catania Pag. 26 IIWAS 2001 - Linz, Austria Metadata representation: an IDB approach " #$ % • General solution for the mixed media access problem & #& % ' ( $ Barbara Catania ) $ – – – – texts images pictures … • support similarity-based indexing – similar caption = similar documents Pag. 27 IIWAS 2001 - Linz, Austria Metadata representation: examples • Solutions based on the illustrated approach have been proposed, among the others, in: – CYC – NKRL (see later) • a solution based on TELOS has also been proposed to construct and manage an API for a metadata repository Barbara Catania Pag. 28 IIWAS 2001 - Linz, Austria Integration: problem Application layer (client side) Mediation layers/ Metasearcher layers Foundation layers (server side) • Problems: – how is it possible to represent a global domain model? – how is it possible to represent the local knowledge? – how is it possible to map global queries into local queries and merging results? Barbara Catania Pag. 29 IIWAS 2001 - Linz, Austria Integration: an IDB approach • Usage of knowledge representation languages for representing: – domain model – heterogeneous sources – query mapping • important role played by ontologies • advantages: – clear formal and declarative foundation – powerful reasoning facilities Barbara Catania Pag. 30 IIWAS 2001 - Linz, Austria Integration: examples • Carnot project (MCC): – integration of heterogeneous sources using a set of articulation axioms that describe how to map SQL queries and domain concepts – articulation axioms built in CYC • SIMS (University of Southern California): – LOOM is used both to represent the global domain model and the local heterogeneous sources characteristics • TSIMMIS (Stanford University): – inheritance-based language (OEM) to describe sources – a logic OO-language is used to specify mediators as views upon OEM sources (LOREL) • Garlic (IBM) – ODMG as model for sources and programming interfaces Barbara Catania Pag. 31 IIWAS 2001 - Linz, Austria Web application design: problem • Web applications are characterized by three main design dimensions: – structure – navigation – presentation • Problems: – which models can be used to support the development of Web applications in all the lifecycle steps? Barbara Catania Pag. 32 IIWAS 2001 - Linz, Austria Web application design: approach • Conceptual level: – Structural modeling: • semantic/hypersemantic data models • OO models – Navigation: • techniques proposed for the more general problem of human-computer interaction specification – first-order logic, Petri Nets, finite state machines, ... – Presentation: • software tools and formal methods • Design level: – structured or semi-structured data models Barbara Catania Pag. 33 IIWAS 2001 - Linz, Austria Web application design: examples • WebML [Politecnic of Milano, Italy] • Araneus [University of Rome, Italy] • Strudel [At&T, INRIA, Univ. Washington] Barbara Catania Pag. 34 IIWAS 2001 - Linz, Austria An IDB approach for metadata representation and retrieval Joint work with E. Bertino and G.P. Zarri Barbara Catania Pag. 35 IIWAS 2001 - Linz, Austria To better explain ... • A specific problem concerning Web applications • a concrete approach – example of an integral approach (NKRL) – example of KESE (CONCERTO) – example of type b) architecture (Knowledge Manager) – important problems (standardization, DBMS facilities) Barbara Catania Pag. 36 IIWAS 2001 - Linz, Austria Metadata • Machine-understandable knowledge that describes the properties and the relationships of Internet resources • To be used to get information about the structure and the contents of these resources • Different classes: – Structure-based metadata: external characteristics of the support (color, shape, texture, motion, etc.) – Content-specific metadata: representing the meaning of documents • keywords • conceptual annotations Barbara Catania Pag. 37 IIWAS 2001 - Linz, Austria Conceptual annotations • Structured information, describing in depth the semantic meaning of a document • several proposals: – UNTANGLE, MIHMA, Information Manifold, Ontobroker • often based on description logic • limitations: – often unable to describe complex events – not always adequate to describe actions, facts, events – automatic extraction quite difficult • alternative approach: NKRL [Zarri, ‘94-’00] Barbara Catania Pag. 38 IIWAS 2001 - Linz, Austria Why NKRL ? • NKRL: Narrative Knowledge Representation Language • Ability to represent, through ontologies, both: – the important notions of a given application domains (concepts) – mutual relationships between concepts (facts, events) • ability to (partially) automatically extract conceptual annotations in NKRL by using tools developed in two European projects: – NOMOS (Esprit P5330) – COBALT (LRE P61011) • the proposed solution fluctuates between: – very simple, low-level rule-based techniques making use of elementary semantic categories like those included in WordNet – very complex inference-intensive applications of CYC Barbara Catania Pag. 39 IIWAS 2001 - Linz, Austria The proposed approach " #$ % • General solution for the mixed media access problem & #& % ' $ &(* – – – – texts images pictures … • support similarity-based indexing – similar caption = similar documents Barbara Catania Pag. 40 IIWAS 2001 - Linz, Austria NKRL On March 1st, 2000, Barbara will go to London city Human_being {moving template} MOVE subj barbara obj london [begin] date: 01/03/2000 london barbara Barbara Catania MOVE subj human_being obj city [begin] Pag. 41 IIWAS 2001 - Linz, Austria Definitional component • Supplies the tools for representing the important notions (concepts) of a given domain • a concept is a frame-based structure composed of – OID – symbolic label like physical_entity, human_being, city, etc. – a set of characteristics features • concepts are represented by using an ontology of terms, called HCLASS • general concepts belonging to the upper levels of are represented inside a catalogue and are assumed to be invariable • similarities with terminological languages Barbara Catania Pag. 42 IIWAS 2001 - Linz, Austria Enumerative component • It is composed of all the instances of sortal concepts, called individuals • non sortal concepts does not admit direct instances • similarly to concepts, individuals are represented as frame based structures • Example: – chair27 – paris_ – lucy_ Barbara Catania Pag. 43 IIWAS 2001 - Linz, Austria Descriptive component • It contains the description of the events proper to a given domain • supplies the tools used to produce the formal representations (predicative templates) of general classes of narrative events, like ‘moving a generic object’, ‘formulate a need’, ‘be present somewhere’ • Templates are structured into an inheritance hierarchy, HTEMP, corresponding to a taxonomy of events • Basic templates (more than 150) are described in a catalogue • By means of proper specialization operations, it is possible to obtain from the basic templates the derived templates needed to implement a particular application Barbara Catania Pag. 44 IIWAS 2001 - Linz, Austria Descriptive component • Templates are characterized by a threefold format: (Pi (R1 a1 )(R2 a2 ) … (Rn an )) – Pi denotes the symbolic label identifying the template (class of events) – Rk , k = 1,…, n, denote generic roles – ak , k = 1,…, n, denote the arguments associated with the roles (concepts, instances, pred. occ.) • Predicates: BEHAVE, EXIST, EXPERIENCE, MOVE, OWN, PRODUCE, RECEIVE • Roles: SUBJ(ect), OBJ(ect), SOURCE, DEST(ination), MODAL(ity), TOPIC, CONTEXT Barbara Catania Pag. 45 IIWAS 2001 - Linz, Austria Factual component • Concerns the instances (predicative occurrences) of the predicative templates – representation of single, specific events • Examples: – Tomorrow, I will move the wardrobe – Lucy was looking for a taxi Barbara Catania Pag. 46 IIWAS 2001 - Linz, Austria Example Milan, October 15, 1993. The financial daily Il Sole 24 Ore reported Mediobanca had called a special board meeting concerning plans for capital increase. c1) MOVE SUBJ (SPECIF sole_24_ore financial_daily): (milan_) OBJ #c2 date-1: 15_october_93 date-2: c2) PRODUCE Barbara Catania SUBJ mediobanca_ OBJ (SPECIF summoning_ (SPECIF board_meeting_1 mediobanca_ special_)) TOPIC (SPECIF plan_1 (SPECIF cardinality_ several_) capital_increase_1) date-1: circa_15_october_93 date-2: Pag. 47 IIWAS 2001 - Linz, Austria Advanced representation facilities • Structured arguments built up making use of a specialized sublanguage (AECS), including four expansion operators: – – – – disjunctive (ALTERNative = A) distributive (ENUMeration = E) collective (COORDination = C) attributive (SPECIFication = S) • ability to bind predicative occurrences together – binding occurrences Barbara Catania Pag. 48 IIWAS 2001 - Linz, Austria Application to multimedia documents & 0 + % ,-./0 /"12 #/3,+.4 5 #/3,+.4 5 %%6 # !78' 5$ 9 5 . ; ! Barbara Catania $ 5 5 % &(* 7:6 5 6 < Pag. 49 IIWAS 2001 - Linz, Austria Queries in NKRL • Query are expressed through search patterns • It must be possible to specify: – perfect match (identical structure) – perfect match apart from cardinality (identical structure apart from the cardinality of AECS lists) – subsumed match (information globally congruent from a semantic point of view - e.g., additional SPEFIC lists are possible-) • automatic transformation of queries into similar queries Barbara Catania Pag. 50 IIWAS 2001 - Linz, Austria Example Which was the theme of the recent board meeting called out by Mediobanca ? ((?w :predicate IS-OCCURRENCE PRODUCE :SUBJ mediobanca_ :OBJ (SPECIF ?x (SPECIF ?y mediobanca_)) :TOPIC ?z) (1_october_93, 20_october_93) ((?x IS-A (:OR assembly_ adjournment_ dissolution_)) (?y IS-A board_meeting) (?z IS-A planning_activity))) Barbara Catania Pag. 51 IIWAS 2001 - Linz, Austria The CONCERTO Esprit Project • The previous ideas have been implemented in the context of the CONCERTO Esprit Project • only textual, possibly semi-structured (HTML, XML) documents • the architecture can be extended to deal with multimedia documents Barbara Catania Pag. 52 IIWAS 2001 - Linz, Austria The CONCERTO KESE architecture ' = > $ ?$ 8 0 ( ! 0 ! + ! -! 1/, * $ $ $ 1 8 ' = . : + ' $ + $ ' * $ 8 * $ . : , $ + * $ 0 $ * $ ( ) ! , + ' ) $ , 1 . :. . : ! " . Barbara Catania : Pag. 53 IIWAS 2001 - Linz, Austria The KM architecture KMIL document through HTTP protocol Web server Knowledge Manager Servlet engine Java API TCP/IP protocol Repository CLIENT Ontologies Example of type b) architecture Barbara Catania n Pag. 54 Documents Conceptual annotations n 1 n IIWAS 2001 - Linz, Austria Technological choices • How to represent conceptual annotations • How to implement the repositories • How to communicate with the Knowledge Manager Barbara Catania Pag. 55 IIWAS 2001 - Linz, Austria Ontologies and Conc. Ann. Representation • Ontologies: – linearization of the hierarchies in a set of tables • Conceptual annotations: – Traditional implementation: three-layered approach: • Common Lisp + a frame/object oriented environment + NKRL – To increase the standardization: • Java + RDF (Resource Description Format) • implemented in XML Barbara Catania Pag. 56 IIWAS 2001 - Linz, Austria RDF • RDF (W3C): proposal for defining and processing WWW metadata • model based on directed labelled graphs – nodes represent Web resources – described by using attributes – edges represent relationships between resources • no predefined vocabulary (ontologies, keywords,…) exists • model implemented in XML Barbara Catania Pag. 57 IIWAS 2001 - Linz, Austria Problems mapping NKRL in RDF • RDF structures: dyadic – two resouces are linked by a binary conceptual relation under the form of a property • NKRL structures: threefold relationship – symbolic label – predicate – one or more roles and fillers • NKRL structures have been transformed in dyadic structures and mapped in RDF Barbara Catania Pag. 58 IIWAS 2001 - Linz, Austria An example of RDF representation <?xml version=1.0 ?> <!DOCTYPE DOCUMENTS SYSTEM CA_RDF.dtd> <CONCEPTUAL_ANNOTATION> <rdf:RDF xmlns:rdf=http://www.w3.org/1999/02/22-rdf-syntax-ns# xmlns:ca=http://projects.pira.co.uk/concerto#> <rdf:Description about=occ11824> <rdf:type resource=ca:Occurrence/> <ca:instanceOf>Template43</ca:instanceOf> <ca:predicateName>Move</ca:predicateName> <ca:subject rdf:ID=Subj43 rdf:parseType=Resource> <ca:filler>barbara_</ca:filler> </ca:subject> <ca:object rdf:ID=Obj43 rdf:parseType=Resource> <ca:filler>london<ca:filler> </ca:object> <ca:listOfModulators> <rdf:Seq><rdf:li>begin</rdf:li></rdf:Seq> </ca:listOfModulators> <ca:date1>01/03/2000</ca:date1> </rdf:Description> IIWAS 2001 - Linz, Austria Pag. 59 </rdf:RDF> Barbara Catania </CONCEPTUAL_ANNOTATION> Repository implementation • Conceptual annotations are represented in XML • Two possible usages of XML documents: – Data Centric: such documents represent the tool by which traditional data are transferred over the Web • XML as a vehicle for data transfer • Example:sales orders, flights scheduling,... – Document Centric: the information is represented by the document itself • XML as a model for data representation • Example: books, textual documents, metadata • In CONCERTO: – document centric XML documents Barbara Catania Pag. 60 IIWAS 2001 - Linz, Austria XML and DBMS • Two categories of DBMS: – XML-Native DBMS: architecture designed for totally supporting management of XML documents • not yet very robust • useful for Document Centric documents • Example: eXcelon (Object Design Inc.) – XML-Enabled DBMS: all DBMS that extend their architecture with functionalities proper to the management of XML documents • Object-Relational (DB2, Oracle8i,…), relational (Microsoft SQL Server) • useful for Data Centric and partially for Document Centric documents Barbara Catania Pag. 61 IIWAS 2001 - Linz, Austria Communication protocol • Standard communication protocol – Knowledge Manager Interface Language (KMIL) for interacting with the KM – implemented in XML • The Knowledge Manager can be hosted on a generic machine, becoming independent from the other modules of the architecture Barbara Catania Pag. 62 IIWAS 2001 - Linz, Austria An example of KMIL input <?xml version=1.0?> <!DOCTYPE KMIL-SESSION SYSTEM KmilIn.dtd> <KMIL-SESSION> <KMIL-ACTION serial_number=1> <KMIL-INSERT-PredOcc IdPO=occ11824 Doc=doc132> <TEXT> RDF Text </TEXT> </KMIL-INSERT-PredOcc> </KMIL-ACTION> <KMIL-ACTION serial_number=2> <KMIL-INSERT-PredOcc IdPO=occ11845 Doc=doc133> <TEXT> RDF Text </TEXT> </KMIL-INSERT-PredOcc> </KMIL-ACTION> </KMIL-SESSION> Barbara Catania Pag. 63 IIWAS 2001 - Linz, Austria An example of KMIL output <?xml version=1.0?> <!DOCTYPE KMIL-SESSION SYSTEM KmilOut.dtd> <KMIL-SESSION> <KMIL-ACTION-OUTPUT serial_number=1 action_status = OK> </KMIL-ACTION> <KMIL-ACTION serial_number=2 action_status = ERROR> <ERROR-CODE code = KMIL-ERR-08/> </KMIL-ACTION> </KMIL-SESSION> Barbara Catania Pag. 64 IIWAS 2001 - Linz, Austria On-going work • Efficient management of conceptual annotations on persistent storage – clustering – optimization/indexing – security • Strongly related to XML document management – initial work on clustering and caching Barbara Catania Pag. 65 IIWAS 2001 - Linz, Austria Conclusions Barbara Catania Pag. 66 IIWAS 2001 - Linz, Austria Ideal IDBs • Like a DBMS: – – – – – persistent storage management support of query and update languages support of indexing and query optimization techniques concurrency control and recovery security • Like advanced data models: – like nested models: non-atomic attribute values – like semantic data models and OODBMS: abstraction, inheritance – like hyper-semantic data models: no real distinction between data and knowledge – like active DB: reaction Barbara Catania Pag. 67 IIWAS 2001 - Linz, Austria Ideal IDBs • Like more advanced KESEs: – support of various inference techniques: deductive, abductive, nonmonotonic, probabilistic, analogical • Like any state-of-the-art DBMS, ES shell or KESE: – sophisticated user interfaces as well as knowledge and application engineering tools • Like (some) advanced systems in the AI style: – uniform, high-level type of representation (frames, objects, semantic networks, hybrid representation schemata …) in both the Rule Base and the Fact DataBase Barbara Catania Pag. 68 IIWAS 2001 - Linz, Austria Recent trends • The techniques analyzed before are now at the basis of several research directions • macroscopic directions: – – – – Advanced data models advanced reasoning advanced architectures advanced index and retrieval techniques Barbara Catania Pag. 69 IIWAS 2001 - Linz, Austria An extended taxonomy of IDBs • Advanced data models • advanced architectures – ... – heterogeneous systems – temporal DBMS – cooperative systems – semistructured and unstructured (multi-agent systems) data representation – Ontologies • advanced indexing • advanced reasoning techniques – ... – – temporal DBMS – query languages for – semistructured and unstructured data – data mining Barbara Catania Pag. 70 indexing and retrieving advanced data Internet indexing and retrieval techniques IIWAS 2001 - Linz, Austria Bibliography Barbara Catania Pag. 71 IIWAS 2001 - Linz, Austria Efforts in the DB context BERTINO, E., MARTINO, L. (1993). Object-Oriented Database Systems Concepts and Architectures, Addison-Wesley. BOOCH, G., RUMBAUGH, J., AND JACOBSON, I. (1998). The Unified Modeling Language User Guide. Addison-Wesley, 1998. BUNEMAN, P. (1997). Semi-Structured Data. Tutorial at ACM SIGARTSIGACT-SIGMOD Int. Conf. on Principles of Database Theory, 1997. CERI, S., GOTTLOB, G., AND TANCA, L. (1990) Logic Programming and Databases, Springer-Verlag, Berlin (Germany), 1990. CHEN, M.S., HAN, J., AND YU, P.S. (1996). Data Mining: An Overview from a Database Perspective. IEEE Transactions on Knowledge and Data Engineering, 8(6):866-883, 1996. GYSSENS, M. AND GUCHT, D.V. (1991). A Comparison between Algebraic Query Languages for Flat and Nested Databases. Theoretical Computer Science, 87:263-286, 1991. Barbara Catania Pag. 72 IIWAS 2001 - Linz, Austria Efforts in the DB context HULL, R. AND KING, R. (1987) Semantic Database Modeling: Survey, Applications, and Research Issues, ACM Computing Surveys, 19, 201-260. PECKHAM, J. AND MARYANSKI, F. (1988). Semantic Data Models. ACM Computing Surveys, 20(3):153-189, 1988. POTTER, W.D., TRUEBLOOD, R.P., AND EASTMAN, C.M. (1989). HyperSemantic Data Modelling. Data & Knowledge Engineering, 4:69-90, 1989. POTTER, W.D., and TRUEBLOOD, R.P. (1988) Traditional, Semantic and Hyper-semantic Approaches to Data Modeling, IEEE Computer, 21, n° 6, 53-64. RUMBAUGH, J. ET AL. (1991). Object Oriented Modeling and Design, Englewood Cliffs, N.J., Prentice Hall, 1991. WIDOM, J. AND CERI, S. (1996). Introduction to Active Database Systems. In Active Database Systems: Triggers and Rules for Advanced Systems (J.Widom, S. Ceri, eds.), Morgan Kaufmann Publishers, San Francisco (Calif.). Barbara Catania Pag. 73 IIWAS 2001 - Linz, Austria Efforts in the AI context BRACHMAN, R.J., AND LEVESQUE, H.J. (1985) A Fundamental Tradeoff in Knowledge Representation and Reasoning, in Readings in Knowledge Representation and Reasoning, Brachman, R.J., and Levesque, H.J., eds. San Francisco: Morgan Kaufmann BREWKA, G. (1987) The Logic of Inheritance in Frame Systems, in Proceedings of the Tenth International Joint Conference on Artificial Intelligence - IJCAI/87. San Francisco: Morgan Kaufmann. CHANDRASEKARAN, B., JOSEPHSON, J.R., AND BENJAMINS., V.R. (1999) Ontologies: What are they? Why do we need them? IEEE Intelligent Systems and Their Applications, 14(1):20-26, 1999. Special Issue on Ontologies. ECAI. Proc. of the European Conference on Artificial Intelligence. FORGY, C.L. (1995) The OPS Languages : An Historical Overview, PC AI 9(5), 16-21. Barbara Catania Pag. 74 IIWAS 2001 - Linz, Austria Efforts in the AI context FRIDMAN NOY, N., AND HAFNER, C.D. (1997) The State of the Art in Ontology Design - A Survey and Comparative Review, AI Magazine, 18(3), 53-74. KRAMER, B.M., CHAUDHRI, V.K., KOUBARAKIS, M., TOPALOGLOU, T., WANG, H., AND MYLOPOULOS, J. (1991) Implementing TELOS, ACM SIGART Bulletin, 2(3), 77-83. LENAT, D.B., GUHA, R.V., PITTMAN, K., PRATT, D., AND SHEPHERD, M. (1990) CYC : Toward Programs With Common Sense, Communications of the ACM, 33(8), 30-49. MACGREGOR, R. (1991). Inside the LOOM Classifier. SIGART Bulletin, 2(3):70-76, 1991. MILLER, G.A. (1995) WordNet : A Lexical Database for English, Communications of the ACM, 38(11), 39-41. MUGNIER, M.-L., AND CHEIN, M., eds. (1998) Conceptual Structures : Theory, Tools, and Applications (Lecture Notes in Artificial Intelligence 1453). Berlin: Springer-Verlag. Barbara Catania Pag. 75 IIWAS 2001 - Linz, Austria Efforts in the AI context MYLOPOULOS, J., BORGIDA, A., JARKE, M., AND KOUBARAKIS, M. (1990) TELOS: Representing Knowledge About Information Systems, ACM Transactions on Information Systems, 8, 325-362. NEBEL, B., AND VON LUCK, K. (1988) Hybrid Reasoning in BACK, in Methodologies for Intelligent Systems 3, Ras, Z.W., and Saitta, L., eds. Amsterdam: North-Holland NEBEL, B. (1990) Reasoning and Revision in Hybrid Representation Systems (Lectures Notes in Artificial Intelligence, vol. 422). Berlin: Springer-Verlag. PÉREZAND, A.G. AND BENJAMINS, V.R. (1999). Overview of Knowledge Sharing and Reuse Components: Ontologies and Problem-Solving Methods. In Proc. of the IJCAI’99 Workshop on Ontologies and Problem-Solving Methods, 1999. Barbara Catania Pag. 76 IIWAS 2001 - Linz, Austria Coupling DB and AI efforts AL-ZOBAIDIE, A., AND GRIMSON, J.B. (1987) Expert Systems and Database Systems: How Can They Serve Each Other ?, Expert Systems, 4(1), 30-37. BERTINO, E., CATANIA, B., AND ZARRI. G.P. (2001) Intelligent Database Systems, Addison Wesley, 2001 BORGIDA, A. (1995) Description Logics in Data Management, IEEE Transactions on, Knowledge and Data Engineering, 7, 671-682. BRESCIANI, P. (1994) Uniformly Querying Knowledge Bases and Data Bases, in Working Notes of the KRDB’94 Workshop, Reasoning about Structured Objects : Knowledge Representation Meets Databases (Document D-94-11), Saarbrücken: Deutsches Forschungszentrum für Künstliche Intelligenz GmbH. BRODIE, M.L. (1988) Future Intelligent Information Systems: AI and Database Technologies Working Together, in Readings in Artificial Intelligence and Databases, Brodie, M.L., ed. San Francisco Morgan Kaufmann. Barbara Catania Pag. 77 IIWAS 2001 - Linz, Austria Coupling DB and AI efforts BRODIE, M.L. AND MYLOPOULOS, J., eds. (1986) On Knowledge Base Management Systems. New York: Springer-Verlag. COHEN, B. (1989) Merging Expert Systems and Databases, AI Expert, 4(2), 22-31. JARKE, M. AND VASSILIOU, Y. (1984) Coupling Expert Systems with Database Management Systems, in Artificial Intelligence Applications For Business, Reitman, W., ed. Norwood (NJ): Ablex. KRDB WORKSHOPS. Proc. of the Workshop on Knowledge Representation Meets Databases. EXPERT DATABASE SYSTEMS CONFERENCE. (1986,1987,1988,1989) Menlo Park: Benjamin/Cummings. NORRIE, M.C., REIMER, U., LIPPUNER, P., RYS, M., AND SCHEK, H.-J. (1994) Frames, Objects and Relations: Three Semantic Levels for Knowledge Base Systems, in Working Notes of the KRDB’94 Workshop, Reasoning about Structured Objects : Knowledge Representation Meets Databases. Barbara Catania Pag. 78 IIWAS 2001 - Linz, Austria Coupling DB and AI efforts RUPAREL, B. (1991) Designing and Implementing an Intelligent Database Application : A Case Study, Expert Systems with Applications, 3, 411-430 SELLIS, T., LIN, C.-C., AND RASCHID, L. (1993) Coupling Production Systems and Database Systems : A Homogeneous Approach, IEEE Transactions on Knowledge and Data Engineering 5, 240-256. ZARRI, G.P. (1990) Expert Databases : The State of the Art, in Proceedings of the Second International Conference on Expert Systems Applications EXPERSYS'90, Gournay-sur-Marne: IITT-International. Barbara Catania Pag. 79 IIWAS 2001 - Linz, Austria IDBs and Web ARENS, Y., CHEE, C.Y., HSU, C., AND KNOBLOCK, C.A. (1993) Retrieving and Integrating Data from Multiple Information Sources. International Journal of Intelligent and Cooperative Information Systems. Vol. 2, No. 2. Pp. 127-158, 1993. CERI, S., FRATERNALI, P, AND BONGIO, A. (2000) Web Modeling Language (WebML): a Modeling Language for Designing Web Sites. In Computer Networks, 33, 137-157. FERNANDEZ, M.F., FLORESCU, D., LEVY, A., AND SUCIU, D. (1998) Web-Site Management: The Strudel Approach. Data Engineering Bulletin, 21(2): 14-20. FRATERNALI, P. (1999) Tools and Approaches for Developing DataIntensive Web Applications: a Survey. In ACM Computing Surveys, 31(3), 227-263. Barbara Catania Pag. 80 IIWAS 2001 - Linz, Austria IDBs and Web GRAVANO, L. AND PAPAKONSTANTINOU, Y. (1998). Mediating and Metasearching on the Internet. In Bulletin of the IEEE Computer Society Technical Committee on Data Engineering, 1998. GUHA, R.V., AND LENAT, D.B. (1994) Enabling Agents to Work Together, Communications of the ACM, 37(7), 127-142. MARTIN, P., POWLEY, W., AND ZION, P. (1998) A Metadata Repository API, in Proceedings of the KRDB’98 Workshop, Innovative Application Programming and Query Interfaces (Swiss Life Technical Report). MECCA, G., MERIALDO, P., ATZENI, P., and CRESCENZI, V. (1999) The (Short) Araneus Guide to Web-Site Development - Second Intern. Workshop on the Web and Databases (WebDB'99) . WIEDERHOLD, G. (1994). Interoperation, Mediation, and Ontologies. In Proc. of the Int. Symp. on Fifth Generation Computer Systems (FGCS’94), Workshop on Heterogeneous Cooperative Knowledge-Bases, Vol. W3, pages 33-48, ICOT, Japan, 1994. Barbara Catania Pag. 81 IIWAS 2001 - Linz, Austria View publication stats NKRL for semistructured data representation BERTINO, E., CATANIA, B., LARADI, D., MARIN, B., AND ZARRI, G.P. (2000) Repository Management in an Intelligent Indexing Approach for Multimedia Digital Libraries, ISMIS’2000. Berlin: Springer-Verlag. BERTINO, E., CATANIA, B., AND ZARRI, G.P. (2000) A Conceptual Annotation Approach to Indexing in a Web-Based Information System Economic Research and Electronic Networking Journal (NETNOMICS), 2(3):247-264, 2000. ZARRI, G.P. (1997) NKRL, a Knowledge Representation Tool for Encoding the ‘Meaning’ of Complex Narrative Texts, Natural Language Engineering — Special Issue on Knowledge Representation for Natural Language Processing in Implemented Systems, 3, 231-253. ZARRI, G.P: ET AL. (1999) CONCERTO, An Environment for the "Intelligent" Indexing, Querying and Retrieval of Digital Documents. In LNAI 1609: Foundations of Intelligent Systems, Proc. of the 11th Int. Symp. ISMIS'99, pages 226--234, Warsaw, Poland, June 1999. Barbara Catania Pag. 82 IIWAS 2001 - Linz, Austria