Data Modeling Comparison of data modeling techniques By

Data Modeling [Comparison of data modeling techniques ] By Renjini Sindhuri

Contents Introduction E-R modeling l Peter Chen l Information Engineering l Barkers Notation l IDEFIX l UML modeling l X- Entity modeling l XUML l Conclusion l l

Introduction Data modeling is the act of exploring data oriented structures. l Examines and compares different data modeling techniques l In the data modeling techniques we have traditional modeling and object oriented modeling of data l

E-R modeling It is a conceptual data model that views the real world as consisting of entities and relationships l It is used to transform relational tables that are easy to understand that enables easy communication with the end user l Peter –Chen developed E-R model l

Peter –Chen notation Entities are represented in the squared cornered and circles as attributes l Many –Many relationships can be represented without associative entity l Relationship itself has attributes and are considered as objects l It failed to represent unique identifier l

Peter Chen’s Model

Information Engineering model Developed by Clive Finkelstein l Entities are represented in the squared cornered and attributes are not shown at all they are shown in a separate list called entity list l Relationships like mandatory 1 and many can be represented l Unique identifiers are not represented l

Information Engineering model diagram

Barkers Notation l l l Adopted by Oracle corporation for its CASE method Entities can be represented by round cornered rectangle Same entity can be represented for role an interaction or another kind of association Relationship names are prepositions and not verbs Unique identifiers can be represented by hash marks next to the attribute

Barkers Notation diagram

IDEFIX Notation It is a modeling technique that is used by many branches of the United States Federal government l A relationship name is a verb l IDEFIX shows subtypes as separate entity boxes l IDEFIX permits multiple inheritance and multiple type hierarchies l

IDEFIX diagram

UML l l l UML is an object modeling technique It models object classes instead of entities In the object oriented world the relationships are called as associations Cardinality and optionality in UML is conveyed by characters or numbers l Express in the form of more complex upper and lower limits UML introduces a small flag that includes text describing any business rules

UML diagram

XML Notation Describing data and interchanging structured and unstructured data on the Internet l It is a universal language of data on web l XML tags are used to create data structures l XML documents have been widely used for interchanging data between heterogeneous systems. l

XML notation l An example of XML notation http: //www. essentialstrategies. com/publica tions/modeling/xml. htm

X-Entity model Conceptual model of XML uses X entity model in order to represent additional features l The entity can be denoted by ‘E’ l ({A 1, …. An}, {R 1, …Rm}, {D 1, …. Dk}) l Each attribute A is associated with a domain Dom(Ai) Which specifies its value set Cardinality is denoted by Card(Ai)=(min, max) l

X entity model diagram

XUML comprises the characteristics of XML and UML 2. l It is used to express the containment semantics more explicitly l Supporting the concept of Business Components l Specifying the data dependencies in multiple context l

XUML diagram UML and XUML model of a book store

Comparison of data modeling techniques S. No Modeling Technique Peter Chen Information Engineering IDEFIX Richard Barker’s notation UML 1. Entities squared cornered and circles as attributes Squared cornered, attributes are not shown at all. Round or square cornered rectangle Round cornered rectangle Models object classes 2. Relationship Nouns. So the relationships can be represent as objects and has attributes Verb or verb phrase Preposition not verb Associations 3. Constraints between relationships Failed to represent the constraints directly exclusive or) Can represent Constraints exclusive or , inclusive Cannot represent Can represent Constraints exclusive or) Can represent Constraints exclusive or )

Comparison of Data modeling techniques S. No Modeling Technique Peter Chen Information Engineering IDEFIX Richard Barker’s notation UML 4. Cardinality Many to Many relationships can be represented between the entities without the associative entity Can represent in different ways Can represent zero or more , atleast at least one express more complex upper limits, zero, 3, 6 -7, or 9 Cannot represent the sub types and sub-types can be represented inside their super-type 5. Sub types/ Super Types super type 6. Unique Identifier Cannot represent , up to many up to one relationships sub-types can be represented inside their super-type box Sub types can be represented as separate entity boxes separate from its super type. Cannot represent Represented in the form of hash next to Can represent

Comparison of Data Modeling techniques S. No Modeling Technique Peter Chen Information Engineering IDEFIX Richard Barker’s notation UML 7. Aggregation Cannot represent Can represent only binary aggregations 8. Business Rules / Cannot Represent Can Represent Components

Comparison of Data Modeling techniques S. No Modeling Technique 9. Peter Chen Information Engineering IDEFIX Richard Barker’s notation UML High Medium Low High Aesthetic Simplicity Score 10. Completenes s Score Low medium 11. Language Notation Score medium low medium High

Advantages of XUML can express the containment semantics more accurately. l Support the concept of Business Component. l Can specify the data dependencies in l multiple context.

Contd. . XUML is more expressive, precise and understandable. l More rigorous and accurate. l

Conclusion By comparing the aesthetic simplicity, completeness, language notation (relationship) Mr. Barker's notation is favorable for requirement analysis model l XML is used in recent trends it follows a standard format for representing structured and semi structured data on web l X-Entity model has the advantages of both XML schemas and extends the ER model so that it can explicitly represent important features of XML schemas l The distinctive features of XUML made this technique of data modeling the latest trend for conceptual modeling of data. l

References l 1. Conceptual Modeling of XML schemas, Bernadette Farias Losio, Ana Carolina Salgado , Year: 2003, Publisher: ACM l 2. XML conceptual modeling with XUML, Hong. Xing Liu Hua. Zhong University of Science and Technology, P. R. China, Yan. Sheng Lu Hua. Zhong University of Science and Technology, P. R. China, Qing Yang Wuhan Uni Pages: 973 – 976, Year of Publication: 2006, Publisher: ACM Press l 3. PETER PIN-SHAN CHEN, “The Entity Relationship Model. Toward a Unified View of Data” , Massachusetts Institute of Technology, ACM Transactions on Data base System Volume 1, Issue 1, Publisher-ACM l 4. Data modeling in the understanding database course: adding UML and XML modeling to the traditional content. Journal of Computing Sciences in Colleges, Volume 17, Issue 5 (April 2002)

References l 5. Data Modeling 101. http: //www. agiledata. org/essays/data. Modeling 101. html 6. A comparison of Data Modeling , David C Hay, Essential Strategies Inc, October 1999.