ER Models

Evolution of Entity-Relationship Models

Last update: 4 April 2005
Number of models: 100

This site contains a sample of entity-relationship (ER) models. The short citation refers to the reference used for the purpose of evaluation; there may be alternative publications, which have the same content. In the year stated, the ER model was mentioned for the first time, according to the year of the conference, additional materials or information by the author. Consequently, it may differ from the year of the reference. Any special names of the new ER models are listed. If the ancestor of a new ER model is explicitly stated in the reference, it can be found in the column descendant of. In the columns, the following abbreviations have been used:

Constructs:
l: construct formed by labelling, f: construct formed by assigning a function
Entity Type, Relationship Type, Attributes:
d: deleted, r: renamed, rf: refer to, sg: general specialization, sd: domain-specific specialization, X: semantics and name of the original construct not changed
Relationship Type:
mapping: 1:N- and M:N-types of relationships; min/max: relationship types, characterized by min/max-cardinalities
Attributes:
d: deleted, r: renamed, rf: refer to, sg: general specialization, sd: domain-specific specialization, X: semantics and name of the original construct not changed
Syntactic rules:
X: original rules not changed
Focus:
g: graphical notation, t: textual notation
Data Definition/Manipulation:
e: explanations and examples, l: dedicated language
Binary vectors:
1: expressed according to focus, f: focus shift, a: assumption, i: implied by another proposition, m: missing statement (implied by descent)
Partially redundant constructs:
S: specialization/general, J: specialization (disjoint/overlapping), T: specialization (total/partial), W: weak entity types, Z: time attributes (multivalued)

Some papers on theoretical and practical problems of ER modeling (ERM) as well as ER analogies, ER homologies and ERM surveys are contained in the list of references.The evolution of Petri nets has been investigated as well.

No.

Constructs

Syntactic Rules

Semantics

Focus

Data Definition

Data Manipulation

Remarks

Cause

Alternative Publications

Binary Vector

Expressiveness Values

Partially Redundant Constructs

Inherent

New

Citation

Content

Structure

Integrity

Behavior

Application

Domain-Specifics

Value Sets / Attributes

Invented

Borrowed

Expressiveness

Applicability

Correction

Integration

Implementation

Other

Adapted

Included

Adapted

Included

(Chen 1976)

1975

sg: weak

sg: existence dependency, sg/l: mapping (1:N,M:N, 1:1)

sg: key, sg/l: role

(1) Relationship types connect entity types. (2) Entity types are assigned attributes. (3) Relationship types can be assigned attributes.

formalized

ERM incorporates ... semantic information about the real world (p. 9),... framework from which the three existing data models may be derived (p. 10).

structure

(Chen 1975)

0.15

0.28

0.00

(dos Santos et al. 1980)

1979

(Chen 1976)

r: type

r: type, sg/l: role

type constructors: sum, product, correspondence (theory of data types)

Different types are connected by a type constructor to form a new type.

formalized

reference to (Smith/Smith 1977)

... expressing the semantic contents and the semantic structure of the data base, ... natural expression of the real world (p. 103).

structure

0.27

0.50

0.00

(Schiffner/Scheuermann 1979)

1979

(Chen 1976)

sg: selector

sg: generalization, aggregation, sg/l: mapping

informal

reference to (Smith/Smith 1977)

...support more abstractions (views) (p. 139).

structure

0.24

0.44

0.00

(Scheuermann et al. 1980)

1979

(Scheuermann/Schiffner 1979)

sg: generalization, aggregation, total, weak, sg/l: mapping

formalized

... lack of capabilities to express abstractions (p. 122), the original ERM ... [did] not ...[allow to express] any additional constraints (p. 124).

structure, integrity

0.33

0.61

0.00

S, W

(Tjoa/Wagner 1980)

1979

(Chen 1976)

sg: weak

sg/t: attribute values 'undefined', 'unknown'

formalized

0.15

0.28

0.00

(Atzeni et al. 1983)

1981

(Chen 1976)

sg: subset, generalization

sg: key

primarily informal

tool, high level language to express transactions

(X)

0.27

0.50

0.00

(Caldiera/Quitadamo 1983)

1981

(Chen 1976)

sg: weak

sg: generalization, existence dependency, optional

formal

0.21

0.39

0.00

S, W

(De et al. 1983)

1981

CR: ConCept Relationship Model

(Chen 1976)

r: concept

r: modeling functions, sg/l: horizontal (logical, causal), vertical (logical, mathematical, part-whole, classificatory)

formal

... capture information requirements for various levels of the organizational hierarchy (p. 455).

structure, integrity

0.21

0.33

0.11

0.00

(Dogac/Chen 1983)

1981

sg: weak

sg: total, weak, aggregation, generalization, sg/l: mapping

informal

... include the recent extensions (p. 357).

0.30

0.56

0.00

W, S

(Klopprogge 1983)

1981

TERM (Time-extended ERM)

(Chen 1976)

sg: key, role, history

formal

... allows a general and rigorous treatment of time,... express very general patterns of change (p. 474).

time

0.24

0.44

0.00

(Tabourier/Nanci 1983)

1981

(Chen 1976)

sg: functional dependency, sg/l: min/max

sg/l: role

restriction: binary relationship types

formal

alternative graphic representation: Occurrences structure diagram

Some kinds of integrity constraints cannot be represented (p. 73). ... produce an equivalent model with only "1 to n" binary relationships, without lack of integrity information (p. 97).

integrity

0.21

0.39

0.00

(Webre 1983)

1981

(Scheuermann et al. 1980)

sg: existence dependency, mapping class, restriction class, completeness class

formalized

... to be a need for mutual existence dependencies (p. 173). ...roles of relationship types ... not clearly defined (p. 174). ...semantic information not being modeled (p. 174).

structure, integrity

0.21

0.39

0.00

(Al-Fedaghi 1983)

1983

(Scheuermann et al. 1980)

sg: generalization

sg: calculated, uninherited, sg/l: key

Relationships connect attributes (equality).

informal

alternative representation of generalization hierarchy; analogy to multidimensional data analysis

... represent particular concepts of the analyzed [petroleum engineering] environment (p. 761)

domain (process industry), structure

0.21

0.39

0.00

(Bertino 1983)

1983

(Chen 1976)

sg: fragmentation

sg: key

formal

vertical fragmentation corresponds to creating new entity/relationship types

... represent fragmentation schema (p. 189).

structure

0.15

0.28

0.00

(Lenzerini/Santucci 1983)

1983

(Chen 1976)

sg: generalization, subset

sg: generalization, subset, sg/l: min/max

sg/l: identifier

Relationships connect relationships (generalization).

formalized

g/t

language to specify integrity constraints

... the problem of expressing cardinality constraints is always an open one (p. 529).

integrity, structure

0.36

0.67

0.00

(Nakano 1983)

1983

Logic-oriented ER

(Chen 1976)

sg: classification, aggregation, generalization,

sg: role

formal

language to specify integrity constraints

... describe a conceptual structure, a variety of integrity constraints and high-level queries (p. 551).

structure, integrity

0.27

0.50

0.00

(Sakai 1983)

1983

(Chen 1976)

sg: mulitvalued, compound, identifier

behavior description

formal

g/t

Petri Nets represent behavior of entities and relationships, drawn together with ER diagram

A conceptual schema should contain descriptions of both aspects [structure/behavior], (p. 111).

structure, behavior

(Sakai 1984)

0.27

0.50

0.00

(Tabourier 1983)

1983

EROS (Entity, Relationship, Occurences Structure)

(Tabourier/Nanci 1983)

r: structure

r: structure, sg:min/max

formal

language to specify integrity constraints

[ER diagrams] may present a lack of precision (p. 565).

(X)

0.27

0.50

0.00

(Elmasri et al. 1985)

1985

ECR (Entity-Category-Relationship)

(Chen 1976)

r/sg: category, generalization/subset category

sg/l: min/max-participation, sg: total/partial/functional participation (derived)

sg/l: compound, multivalued, role, d: identifier

formalized

... [ER model] not sufficient to represent some important data semantics (p. 77).

structure

0.48

0.89

0.00

S, J, T

(Ferg 1985)

1985

(Chen 1976)

sg/l: key

informal

ID-dependence on time periods

... representing the time dimension (p. 280).

time

0.21

0.39

0.00

(Gilberg 1985)

1985

Information Engineering

sg: subentity

sg: recursive, intersection data groups

informal

separate subschema for each primary entity

[large ER diagrams] proved unreadable (p. 320)

0.18

0.33

0.00

(Hsu 1985)

1985

(Chen 1976)

sg: semantic, d: weak

sg: semantic, operational (plural, functional, mandatory)

sg: key, role

informal

... integrating the ... top-down analysis and bottom-up synthesis methodologies (p. 56).

0.15

0.28

0.00

(Lenzerini 1985)

1985

SERM (Semantic ERM)

r: category

r: category, sg: subset, part-of, cardinality-ratio

r: category, sg: role

Subset-relationship connects relationships.

formal

... integration of the original Entity-Relationship Model and many proposed extensions (p. 271). . ... give a sound foundation to possible implementations of a high-level programming language (p. 272).

0.27

0.50

0.00

(Qian/Wiederhold 1985)

1985

sg: primary, secondary

sg: subtype, association

sg: key, single-/multi-valued, constrained, derived

informal

... provide syntactic embodiment of semantic data model and a uniform linguistic tool for data definition, query, manipulation (p. 46).

0.27

0.50

0.00

(Velez 1985)

1985

TIGRE

sg: min/max

sg: role, type

constructor (programming languages)

The application of a constructor results in a typ.

informal

The aim ... is to handle multi-media data in an office automation context (p. 82).

domain (multi-media)

0.18

0.33

0.00

(Bruno/Elia 1987)

1986

(Chen 1976)

sg: static/dynamic specialization, sg/l: cardinality

state, transition

informal

g/t

...[ERM] does not give any information on the behavior of entities (p. 169). .... integrating requirements specification and simulation (p. 169).

behavior

0.27

0.50

0.00

S, Z

(Dittrich et al. 1987)

1986

CERM (Complex-Entity/Relationship Model)

(Chen 1976)

r: objects, sg: complex objects, c: versions

sg/l: min/max

sg/l: predefined domains

primarily informal

... [the ERM] lacks a notion of... [composite] objects and a set of operators capable of dealing with these aspects (p. 422).

structure, time

0.27

0.50

0.00

(Eder et al. 1987)

1986

BIER (Behavior Integrated E-R)

(Chen 1976)

sg: weak, subentity

sg: generalization

sg/l: surrogate

states, activities (Petri nets)

(1) States are connected to entities.,(2) Activities connect states.

informal

Petri Nets represent behavior of entities

.. introduce a general framework ... which includes both static and dynamic aspects ... (p. 147).

behavior, time

0.24

0.44

0.00

(Junet 1987)

1986

(Chen 1976)

sg: weak

sg: sub-relation, generic-relation, sg/l: cardinality

sg/l: key, sg: multivalued-role, mandatory/not-mandatory roles

formal

g/t

...., the meaning of a role in the original E-R model is not powerful enough to capture some enhanced semantic of the real world (p. 305).

structure

(X)

0.30

0.56

0.00

W, S

(Lipeck/Neumann 1987)

1986

(Chen 1976)

r: objects; sg: complex, generalization, partition; sd: geoobjects, map objects

sg/l: functional

sg/l: identifier

formal

g/t

...[extended ERM] is adapted to geoscientific needs ... (p. 68).

domain (geographic IS)

0.30

0.50

0.00

0.17

J, T

(Teorey et al. 1986)

1986

sg: weak, subset, generalization

sg: connectivity, membership class

sg/l: identifier, descriptor

informal

... inadequacy of the initial [ER] modeling constructs (p. 201).

0.21

0.39

0.00

W, S

(Downs 1992)

1987

SSADM

sg: optional, exclusive

sg/l: key

restriction: binary relationship types

informal

ERM as a step of SSADM (Structured Systems Analysis and Design Method)

0.18

0.33

0.00

(Flory/Giard 1988)

1987

(Chen 1976)

sd: spacio-temporal

sgL: cardinality

sg/l: composite, identifier, role

informal

spacio-temporal entity types are generally hidden in ER diagrams

... [application in manufacturing]... leads to new requirements which are not always matched by conventional systems (p. 249).

domain (manufacturing)

0.15

0.28

0.00

(Hohenstein et al. 1987)

1987

(Chen 1976)

r: objects, sg/l: complex objects, versions

sg/l: min/max, sg: derived

sg/l: key, role, multivalued, sg/l: optional, derived

integrity boxes, type construction

(1) Integrity boxes are connected to entity types (relationship types, attributes). (2) Different input entity types are connected by type constructors to form separate output entity types [requirements].

formal

... [extended ERM] to be suited for the design of standard and non-standard applications ... (p. 59). ... integrate all important semantic concepts of recent semantic data models (p. 64). ... shall form the basis for several database and information systems design tools (p. 59).

structure

0.45

0.83

0.00

J, T

(ISO 1987)

1987

ISO

sg/l: min/max, functional dependent

sg/l: identifier

informal

... providing a framework for the design of conceptual schema languages (p. 2).

0.15

0.28

0.00

(Lazimy 1988)

1987

ERA (Entity-Relationship-Attribute)

(Chen 1976)

sg: action, sg/l: probabilistic

constraint set, transformation set

Constraint sets and transformation sets connect attributes.

formalized

[new ERM] .. sufficiently rich for representing both transactional and conceptual information (p. 133).

structure, uncertainty

0.18

0.33

0.00

(Kappel/Schrefl 1989)

1988

BIER

rf: BIER

complex activities

Complex activities connect states.

informal

... [new ERM] considers the structural as well as the behavioral aspects (p. 311).

structure, behavior

0.21

0.39

0.00

(Navathe/Pillalamarri 1989)

1988

OOER (Object-Oriented ERM)

(Chen 1976)

sg: weak, generalization, specialization, classification

sg: association

sg: key

operation (object-orientation)

Operations are connected to entity types.

informal

... enhance ER model to improve its abstraction capability (p. 186).

structure

0.33

0.61

0.00

W, S

(Put 1989)

1988

(Scheuermann et al. 1980)

sg: total, weak, aggregation, generalization, sg/l: mapping

action

formalized

g/t

actions are the leaf nodes of sequence diagrams, which are used in addition to ER diagrams, verification of conceptual design

A conceptual schema should include all relevant aspects of the real world, both static and dynamic (p. 423).

behavior

0.27

0.50

0.00

(Sinz 1988)

1988

SER (Structured ER)

(Chen 1976)

sg: ER-Typ

sg: complexity, subtype-hierarchy, is-a-hierarchy

sg/l: key

informal

... Verbesserung der semantischen Aussagefähigkeit von ER-Diagrammen (S. 191). ... fünf verbesserungsfähige Schwachstellen des ERM (S. 193).

structure

0.24

0.44

0.00

S, J, T

(Carlson et al. 1990)

1989

NER (Nested ER)

(Chen 1976)

sg: complex

sg: complex, key

formalized

[improve] comprehensibility, retrievability, updatability (p. 44).

0.24

0.44

0.00

(Davis/Bonnell 1990)

1989

EARL

(Chen 1976)

sg: IS-A, PART-OF, AGGR-OF, INST-OF, MEMBER-OF, ROLE-OF,sg/l: min/max

sg: identifier

constraint-links

Constraint links connect relationship types.

informal

... capture abstraction relations (p. 95). ... express certain types of cardinality constraints (p. 96). coupling concept abstraction with constraint specification (p. 95).

structure, integrity

(X)

0.30

0.56

0.00

(Elmasri/Navathe 2000)

1989

EER (Enhanced ER)

(Chen 1976)

sg: weak, subclass, superclass, category

sg/l: cardinality, sg: total participation, identifying, disjoint/overlapping total/partial specialization, aggregation

sg: composite, multivalued, derived, key, sg/l: role

formalized

... To represent these requirements [of newer applications] as accuratly and clearly as possible.... ... the ER model can be enhanced to include these concepts [of semantic data models] .. (p. 73)

structure

0.33

0.61

0.00

W, S, J,T

(Embley/Ling 1990)

1989

E2R (E-squared-R)

(Chen 1976)

sg: lexical, non-lexical

sg: functionally dependent, participation, union/partition generalization/specialization, sg/l: min/max

sg: composite, sg/l: role

informal

normalized EER models

... capture a wider range of design-dependent real-world semantics (p. 111). ... improved, synergistic database design methodology (p. 111). ... design work cannot be completed in the ER model alone (p. 112).

0.30

0.56

0.00

S, J, T

(Lazimy 1990)

1989

E2R (E-squared-R)

(Chen 1976)

sg: association, aggregation, class-subclass, membership

sg: derived, objective/goal

functions

(1) Relationships connect attributes. (2) Functions connect attributes.

formalized

knowledge-based design

.. a need ... [to] represent semantically-rich knowledge as well as a variety of activities (p. 130).

knowledge

0.24

0.44

0.00

(Kerschberg et al. 1990)

1989

KER (Knowledge-based ER)

(Teorey et al. 1986)

rf (Teorey et al. 1986)

sg: cardinality, min/max

rf (Teorey et al. 1986), sg: cardinality

formulas

inference rules (knowledge-based systems)

Formulas and inference rules connect attributes.

informal

There was a ... need to specify rule-based knowledge as well as object-oriented data specifications (p. 256).

knowledge

0.27

0.50

0.00

(Parent/Spaccapietra 1992)

1989

ERC+

sg: complex objects

sg: optional, min/max, generalization

sg: surrogate key, complex, sg/l: role

formal

g/t

A richer semantic model [for integration of heterogeneous schemas] was needed (p. 69). ... besides .. expressive power, friendliness was assigned as additional requirement... (p. 69).

structure

0.24

0.44

0.00

O, C

(Teorey et al. 1989)

1989

(Teorey et al. 1986)

rf (Teorey et al. 1986)

rf (Teorey et al. 1986), sg: grouping, exclusive

rf (Teorey et al. 1986)

informal

In an ER diagram with 1000 entities [...] the overall structure will probably not be very clear

0.27

0.50

0.00

(Barker 1990)

1990

(Chen 1976)

sg: subtype

sg: cardinality, optional, exclusive, non-transferable

sg: identifier

restriction: binary relationship types

informal

0.30

0.56

0.00

H, X

(Brunet et al. 1991)

1990

ERC+

rf ERc+

action, event (object orientation)

(1) Actions are connected to entity types. (2) Events connect actions.

informal

... for the description of behavioral aspects (p. 211).

behavior

0.27

0.50

0.00

(Elmasri et al. 1991)

1990

TEER

(Elmasri/Navathe 2000)

rf (Elmasri/Navathe 2000)

role type

(1) Each role type is associated with a single entity type. (2) Role types can be connected by relationship types.

formal

.. conventional DBMS lack the power to describe and query information related to time aspects (p.239).

time

0.39

0.72

0.00

(Hsu/Rattner 1990)

1990

TSER (Two-Stage ER)

(Chen 1976)

sg: mandatory, functional

sg: key

context, subject

informal

focus on system implementation, separate SER and OER diagrams

... consolidating data structures across the manufacturing facility (p. 758).

0.12

0.22

0.00

(Martin 1990)

1990

Information Engineering

sg: subtypes

sg: min/max, exclusive

sg/l: role

restriction: binary relationship types

informal

exclusive relationships

0.21

0.39

0.00

(Schrefl 1991)

1990

BIER

rf BIER

abstract activities, abstract states

rf BIER

informal

[BIER] ... lacks refinement primitives for states and activities (p. 119).

behavior

0.24

0.44

0.00

(Rochfeld et al. 1991)

1990

(Chen 1976)

sg/l: min/max

sg: key

(inclusive/exclusive) functional integrity constraint

Relationships connect relationship types.

primarily informal

[remove] ... a modelling constraint (p. 149).

0.24

0.44

0.00

(Zhu et al. 1991)

1990

HAMMER (Hierarchical Abstraction Mechanism on Models in ER)

sg: hyperentity

sg: min/max-participation, association, aggregation, hierarchy

sg: roles

model, hypermodel

(1) Entity types and relationship types are connected to models. (2) Hypermodels are connected to models.

formalized

[HAMMER] grew out of the concerns with the complexities associated with large conceptual models (p. 75)

0.27

0.50

0.00

(Ferg 1991)

1991

sg: participation cardinality, look across cardinality, generalized cardinality

formalized

[certain ER notations] are found not to be capable.. of expressing [...] cardinality constraints (p. 1)

integrity

0.15

0.28

0.00

(Gogolla/Hohenstein 1991)

1991

EER

different EER

sg: min/max,derived, functional

sg: multivalued, optional, derived, components, role, key

type construction

Different input entity types are connected by type constructors to form separate output entity types [requirements].

formal

... concentrating nearly all concepts of ... semantic data-models...., ... provide our extended ER model with a formal mathematical semantics (p. 369)

(Engels et al. 1992/93)

0.48

0.89

0.00

(Heuser/Peres 1991)

1991

ER-T

(Chen 1976), Petri nets

sg: control places

external agents, transaction sets (Petri nets)

Transaction sets are connected to external agents, control places, entity/relationship types.

informal

... integrated modeling of static and dynamic properties... (p. 247).

behavior

0.27

0.50

0.00

(Kouramajian/Elmasri 1991)

1991

TEER (Temporal Extended ER)

sg: weak

sg: superclass/subclass

sg: temporal, multivalued, composite, surrogate key

temporal element

each entity/relationship type is associated with a temporal element

formal

[current DBMS] ... lack the power to describe and query information related to time aspects... ; handle both valid and transaction time (p. 671).

time

0.33

0.61

0.00

W, S

(Loucopoulos et al. 1991)

1991

ERT (ER-Time)

r: class, sg: time period class, complex object, derived

sg: generalization/specialization, derived, sg/l: cardinality

sg/l: complex, role

formalized

g/t

... capturing and representation of business rules (p. 323). ... the need for expressiveness power of conceptual modelling formalisms... (p. 324).

integrity

(Theodoulidis et al. 1991), (Theodoulidis et al. 1992)

0.36

0.67

0.00

S, J, T

(Moyne et al. 1991)

1991

(Teorey et al. 1989)

rf (Teorey et al. 1989)

sequential ordered function

Sequential ordered functions connect relationship types.

formalized

... represent ordering and time sequencing of process recipe information (p. 421).

domain (process industry)

0.27

0.50

0.00

H, OR

(Tanaka et al. 1991)

1991

ER-R

sg: total participation, sg/l: cardinality

sg/l: key

rule objects

Rule objects are connected to entity types, relationship types or attributes.

informal

g/t

... introduction of active aspects of information systems in the conceptual schema (p. 60).

behavior

0.30

0.56

0.00

(Tauzovich 1991)

1991

TER

sg/l: min/max, optional, snapshot cardinality, lifetime cardinality

informal

[current ER approaches] ... ignore time (p. 163)

time

0.18

0.28

0.11

0.00

(Thalheim 2000)

1991

HERM+

sg: cluster, weak

sg: min/max

sg: nested, key

integrity constraints, operations, type construction

(1) Relationships connect relationships. (2) Integrity constraints/operations are connected to entity or relationship types.

formal

[HERM] ... has a strong theoretical basis (p. 9). The modeling is more natural and can be applied in a simple manner (p. 9). The theory is applicable to practical needs (p. 10). The results of the design are much simpler than with other approaches (p. 10). The model is easy to understand, simple and comprehensible (p. 11). The model is capable to support sophisticated database design techniques (p. 11).

(Thalheim 1999)

0.58

1.00

0.11

0.00

(Wei/Teorey 1991)

1991

ORAC

(Teorey et al. 1989)

r: abstract objects

sg: generalization, categorization, aggregation, min/max

r: literal objects, sg: complex, multivalued, surrogate keys, sg/l: role

constraint primitives

Constraint primitives connected to relationship types.

formal

... answer these criticisms [concerning the ER model]... (p. 32). ... humanly intuitive, semantically expressive, and can be mapped into simple mathematical concepts (p. 32).

(X)

0.33

0.61

0.00

(Batini et al. 1992)

1992

(Chen 1976)

sg: generalization (sg/l: total/partial, disjoint/overlapping), subset

sg: rings, sg/l: min/max

sg: identifier (external/mixed), composite, sg/l: multi-valued, optional, mandatory

formalized

... present basic elements and advanced features of the ER model ... (p. 31).

0.30

0.56

0.00

(Garcia/Sheng 1992)

1992

SEER (Synthesized Extended ER)

(Teorey et al. 1986)

rf (Teorey et al. 1986)

focal objects, transaction access, statistics, procedures (transaction-modeling, frames)

informal

separate diagram

... formalize the descriptions of distributed transaction information (p. 181). ...., facilitating relational distribution design (p. 181).

behavior

0.27

0.50

0.00

(Hainaut 1992)

1992

(Chen 1976)

sd: space

sg/l: min/max, specialization

sg/l: compound, role, optional/mandatory, multivalued, identifier

group

Group connects attributes or roles.

formalized

.. extension of the [ER] model to describe main statistical aspects of data (p. 79).

domain (IS planning)

0.27

0.50

0.00

(Han/Li 1992)

1992

(Chen 1976)

sg: subentity, virtual

sg: generalization, virtual, cardinality

sg: complex, virtual, multivalued, key

informal

... design of deductive databases (p. 192).

knowledge

0.30

0.56

0.00

(Behm/Teorey 1994)

1993

(Chen 1976)

sg: aggregate

sg: subset, sg/l: mapping

s/l: structured, derived

relationship operators, constraints

(1) Relationship operators connect relationship types. (2) Domain-based constraints connect attributes.

informal

g/t

... extend the ER model to support these additional business rules and constraints (p. 46)

integrity

0.33

0.61

0.00

(Dijkstra 1994)

1993

O2XER (Object Oriented eXtensions to the ERM)

ERC+

r: objects, sg: complex objects, c: versionable

sg: generalization

sg: complex, multivalued, optional, role

informal

g/t

... [EER models] lack the support for complex objects, versioning, and slightly explore the aggregation concept (p. 13).

structure, time

0.30

0.56

0.00

(Garzotto et al. 1994)

1993

HDM (Hypertext Design Model)

different EER

sg: derived

r: application web type, sg: derived application web, cardinality, optional

r: unit, sg: component

index, guided tour

Application web types/indexes/guided tours connect entity types, components or application web types.

formalized

Goal of HDM is ... the specification of structural and navigational aspects of hypermedia (p. 179).

domain (hypermedia)

0.21

0.39

0.00

(Heuser et al. 1993)

1993

(Batini et al. 1992)

r: entity place

r: relationship place, dead links

r: attribute place

live links (Petri nets)

Live links connect places, transitions, entity types, relationship types.

formal

ER diagrams are translated into Petri nets.

... the need also to include dynamic properties (p. 275). ... the semantic of an E-R diagram ... can be expressed in terms of a high-level Petri net (p. 276).

behavior

0.27

0.50

0.00

(NIST 1993)

1993

IDEF1X

sg: identifier-dependent

sg: identifying, non-identifying, optional, mandatory, categorization, cardinality

sg: typed, key, sg/l: role

informal

.. support integration (p. i). ...meet the following requirements: .... consistent, extensible, transformable; .. easy for users to grasp, yet powerful and robust; ... teachability; ...be automatable (p. ii).

structure

0.21

0.39

0.00

(Norrie 1994)

1993

sg/l: min/max, sg: is_a

sg: complex

formal

g/t

How can [ER and object-oriented modeling] be combined to support the development of database application systems (p. 391)?

0.24

0.44

0.00

(Pernul et al. 1994)

1993

extended ER

sd: security object

r: security object, sg/l: cardinality

sg: identifier

security constraints

Security constraints connect attributes.

formal

g/t

... capture security semantics (p. 166).

domain (security)

0.27

0.50

0.00

(Winter 1994)

1993

sd: derived

sg: existential dependency, aggregation, generalization, cardinality

sg: key

informal

... allow an utilization of database trigger mechanisms (p. 60).

0.24

0.44

0.00

H, D

(Gandhi et al. 1994)

1994

LER (Leveled ER)

(Chen 1976)

sg: complex, c: aspect

sg: is_part_of

sg/l: identifier, sg: multi-valed aspect

formal

... the current proposal overcomes ... [limitations of existing ER models] (p. 421).

0.27

0.50

0.00

(Ling/Teo 1994)

1994

OOER (Object-Oriented ER)

(Chen 1976)

sg: weak, complex

sg: m-n, recursive, existence dependent, identifier dependent, ISA, UNION, INTERSEcTION

sg: multi-valued, composite, derived, methods

formalized

... inability to judge the quality of an OO schema design, the presence of inheritence conflicts ..., the lack of ... support for different relationship types ... and views (p. 241).

structure

0.30

0.56

0.00

(Liu et al. 1994)

1994

EVER (Evolutionary ER)

(Teorey et al. 1986)

sg: visible, defunct

sg: domain-changed, renamed, resumed, derived, dependent, independent, key

version derivation

Version derivation connects defunct and visible entity types.

formalized

... provide for the specification of the derivation of relationships between schema versions (p. 133).

time

0.27

0.44

0.11

0.00

(Oh/Navathe 1995)

1995

SEER (Security Enhanced ER)

(Chen 1976)

sd: user group

sd: security

Security relationship connects user, entity or relationship types.

informal

... address the conceptual modeling of security features and authorization histories... (p. 170).

domain (security)

0.12

0.22

0.00

(Green/Benyon 1996)

1996

ERMIA (ER for Information Artifacts)

sd: pile, chain, sorted list, unsearchable, hashed, conceptual

sg: optionality, sg/l: cardinality

sd: perceptually coded

informal

... analyse structures with respect to human abilities rather than computer ones (p. 803).

domain (cognitive science)

0.18

0.17

0.00

0.50

(Lee et al. 1996)

1996

TEER

sg/l: cardinality

sg: composite, multivalued, time series

informal

The concept of time series, ..., does not fit well within these [temporal data] models (p. 341).

time

0.42

0.78

0.00

(Steeg 1996)

1996

ERBM (ER and Behavior Model)

HERM

rf HERM

behavior properties, behavior specification

formal

... derive an efficient database application to avoid logical/physical tuning (p. 106). [by applying the proposed approach]... logical/physical tuning measures ... can be ... avoided (p. 105).

0.48

0.89

0.00

(Chu/Zhang 1997)

1997

general EER

r: association, sg: ISA, sg/l: min/max

role classes (object-oriented)

Role classes are connected to entity types or role classes.

informal

... support object evolution and extension for long lived objects (p. 257).

structure

0.21

0.39

0.00

(Moody 1997)

1997

Information Engineering

sg: cardinality, optionality

subject areas

Each entity type is assigned to one subject area.

informal

On of the ... limitations of the Entity Relationship Model is its inability to cope with the size and complexity of data models encountered in real world situations (p. 184).

0.24

0.44

0.00

(Bergamaschi/Sartori 1998)

1998

CHRONO

IDEFI1X

sg/l: temporal

sg: aggregation, generalization, cardinality, optionality

temporal element

Each entity type is associated with a temporal element.

formalized

... necessity of representing time varying information (p. 35).

time

0.39

0.72

0.00

S, Z

(Lee/Elmasri 1998)

1998

ITDM (Integrated Temporal Data Model)

EER (enhanced)

sg: generalization, sg/l:min/max

sg: time-varying, time-series, surrogate,sg/l: role

formal

... we ... formalize a conceptual model that supports time-series objects as well as traditional version-based objects (p. 21).

time

0.36

0.67

0.00

S, Z

(Sapia et al. 1999)

1998

ME/R (Multidimensional ER)

(ISO 1987)

sg: dimension level

sg: fact, classification

formalized

In order to allow the natural representation of the multidimensional semantics ... (p. 109)

multidimensionality

0.21

0.22

0.33

0.00

(Balaban/Shoval 1999b)

1999

EER (Enhanced ER)

general EER

sg: cardinality, specialization

sg: composite, multivalued, key, role

primitive methods (abstract data types)

Primitive methods are associated with entity and relationship types.

formal

The integrity methods ... are automatically created for a given EER diagram (p. 14).

(Balaban/Shoval 2002)

0.42

0.78

0.00

S, J, T

(Badia 2000)

2000

(Chen 1976)

sg: cardinality, total, partial

sg: complex, multi-valued, role

constraint

generalized quantifier (logics)

formal

... limitations on what can be expressed in the [ER] model (p. 323).

integrity

0.42

0.78

0.00

S,T

(Castellani et al. 2000)

2000

MCX

(Tardieu et al. 1974)

sg/l: min/max-cardinality, target class

informal

.. one must be able to mention multiple cardinality couples (p. 34). ... the cardinalities of n-ary (n>=3) relationships defined with the chen model are ambiguous (p. 32).

integrity

0.15

0.28

0.00

(Karlapalem et al. 2001)

2001

EREC (ER for e-ContraCts)

ER-R

rf ER-R

rf ER-R, sg/l: min/max

rf ER-R

sd: exception, contract event

Rule objects are connected to entity types, relationship types or attributes.

informal

g/t

exceptions are represented as rules

... conceptually modelling e-contracts ... (p. 193).

domain (electronic commerce)

0.39

0.72

0.00

(Ma et al. 2001)

2001

FEER (Fuzzy Extended ER)

EER

sg: weak

sg: specialization, category, aggregation

sg: key, multivalued, composite, fuzzy

grade of membership

grades of membership are associated with entity types, relationship types, attributes

formalized

g/t

.. to cope with imperfect as well as complex objects ... (p. 697).

uncertainty

0.42

0.61

0.33

0.00

W, S, J, T

(Wagner 2001)

2001

AOR (Agent-Object-Relationship)

(Barker 1990)

sd: event, action, claim, commitment, agent, object

sg: specialization, composition, sd: sends, receives, does, perceives

rf: Barker (1990)

informal

ER modelling does not account for the dynamic aspects of information and knowledge processing systems (p. 114).

domain (institutional agents)

(Wagner 2002)

0.36

0.56

0.00

0.33

(Bowers et al. 2002)

2002

(Chen 1976)

r: schematic, sd: authoritative

r: schematic, sg/l: min/max

sg: complex, identifier, sd: mark values, sd/l: role, optional,

anchor

Entity types and relationship types may be associated with at most one anchor.

formal

g/t

... providing structured access to unstructured information sources (p. 90)...

domain (hypermedia)

0.24

0.44

0.00

(Vert et al. 2002)

2002

sg: subset, cardinality, optionality, sd/l: fuzzy spatial

formalized

g/t

uncertainty is extensively treated

... express ... different degrees of relevance to a GIS user's specific needs (p. 165).

domain (geographic IS)

0.30

0.39

0.33

0.00

(Psaila 2003)

2003

ERX (ER for XML)

(Batini et al. 1992)

sg: alternative, sd: containment, sg/l: min/max

sg: implied

informal

g/t

[ERX] allows to deal with concepts coming from XML documents at the conceptual level (p. 378).

domain (XML)

0.27

0.50

0.00

100

(Shimazu 2003)

2003

RER (Refined ER)

(Chen 1976)

sg: generalization

sg: identifier, derived, role

formal

derived attributes are indicated, but stored

The purpose of an RER model is to apply ILP [Inductive Logic Programming] (p. 390).

knowledge

0.18

0.33

0.00

perceptual(!a)

The attributes (e.g., shape, color) differ through perceptual aspects, in contrast to attributes differing in their symbolic labels (e. g., book titles).

agent_r(!r)

The relationship type represents involvements of agents, e.g., receive/send messages, do actions.

agent_e(!e)

The entity type represents an agent that can, e.g., perform actions, communicate and make commitments.

conceptual(!e)

Entities of a conceptual entity type (e.g., a plot of a book) represent only information.

search(!e)

Some entity type represents a class of items that share time and memory requirements when searched (e. g., pile, chain, stored list).

geoscientific(!e)

Some entity type represents a special class of geo-scientific objects (e.g., geo-objects, map objects).

time_dependent(!a)

During schema evolution, the name, the domain or the existence (addition/deletion) of some attribute may change.

time_participate(!e, !r)

The allowed or required participation of entities in relationships depends on the time horizon considered.

causal_relation(!e1, !e2, !r)

The relationship type indicates a cause and effect connection between the involved entity types.

a_fuzzy(!a, !rn)

Some attribute is not with certainty but with some grade !rn member of a schema.

r_fuzzy(!r, !rn)

Some relationship type is not with certainty but with some grade !rn member of a schema.

e_fuzzy(!e, !rn)

Some entity type is not with certainty but with some grade !rn member of a schema.

classification(!e1, !e2, !r)

The (binary) relationship type relates instances of entity type !e1 (e.g., ’city’) to instances of the more aggregated entity type !e2 (e.g., ’country’).

dimensional_level(!e)

In the field of multi-dimensional data analysis, the entity type represents a set of members (e.g., days/months; towns/countries) of a dimension (e.g., time, geography).

fact(!r)

The relationship type represents a subject of multidimensional data analysis.

r_transaction(!r)

Transactions are defined on the relationship type. A transaction is a logical unit of database manipulation (insert, update, delete) that transforms a consistent database state into another one. To preserve consistency, pre- and postconditions of the transaction are stated.

e_transaction(!e)

Transactions are defined on the entity type. A transaction is a logical unit of database manipulation (insert, update, delete) that transforms a consistent database state into another one. To preserve consistency, pre- and postconditions of the transaction are stated.

r_ex_constraint(!r)

General restrictions are defined on the relationship type that must hold for all of its instances and in any database state.

e_ex_constraint(!e)

General restrictions are defined on the entity type that must hold for all of its instances and in any database state.

ident_participate(!e1, !e2, !r)

The entities of distinct types that are associated by some relationship are identical.

excl_participate(!e, !r1,... !rj)

Entities of some type are involved in only one of several types of relationships.

option_participate(!e, !r)

Some entities of a given type are not involved in instances of some relationship type.

min_participate(!e1, ..., ei, !r, N)

All entities of some type(s) must participate in at least one instance of a relationship type.

max_participate(!e1,..., !ei, !r, N)

All entities of some type(s) participate in more than one instance of a relationship type.

value_description(!a)

The allowed attribute values are specified, e.g., by listing them (including NULL or UNDEFINED), by providing data types or by stating ranges. This also covers composite attributes.

multivalued(!a)

The attribute has more than one value for some entity or relationship.

key_description(!e)

The primary key of the entity type is described.

r_condition(!r)

An arbitrary condition is defined on a relationship type to divide its instances into subsets.

e_condition(!e)

An arbitrary condition is defined on an entity type to select some of its instances in a given database state.

derived_rel(!r)

A relationship is not stored but derived from existing relationships, respectively; derivation follows a derivation rule.

derived_att(!a)

An attribute and its values are not stored but derived from the values of other attributes; derivation follows a derivation rule.

role(!e, !r)

An entity plays a certain role in a relationship.

e_type_treatment(!r, !e)

Some relationship type is treated as an entity type. This is also called aggregation.

d: deleted, l: label, r: renamed, rf: refer to, sg: general specialization, sd: domain-specific specialization, X: semantics and name of the original construct not changed

d: deleted, l: label, r: renamed, rf: refer to, sg: general specialization, sd: domain-specific specialization, X: semantics and name of the original construct not changed, mapping: 1:N- and M:N-types of relationships, min/max: relationship types, characterized by min/max-cardinalitied

d: deleted, l: label, r: renamed, rf: refer to, sg: general specialization, sd: domain-specific specialization, X: semantics and name of the original construct not changed

C: composite attributes/objects, D: derived entity types, F: uncertain attributes, H: dependency/ total relationship, O: optional attributes, OR: order relation, R: anker, X: non-transferable relationship

S: specialization/general, J: specialization (disjoint/overlapping), T: specialization (total/partial), W: weak entity types, Z: time attributes (multivalued)

RD: Literals [E28] to [E33]

RD: Literals [E19] to [E27]

RD: Literals [E1] to [E18]

RD: Literals [E1] to [E33]

1: expressed according to focus, f: focus shift, a: assumption, i: implied by another proposition, m: missing statement (implied by descent)

publications that are equal in content

e: explanations and examples, l: dedicated language

g: graphical notation, t: textual notation

X: original rules not changed

if stated by the author

ER model mentioned for the first time (according to the year of the conference, references or information by the author); may differ from the year of the evaluated publication

Publication used for evaluation