138 77
English Pages 179 [189] Year 2000
Lecture Notes in Computer Science Edited by G. Goos, J. Hartmanis and J. van Leeuwen
1921
Berlin Heidelberg New York Barcelona Hong Kong London Milan Paris Singapore Tokyo
Stephen W. Liddle Heinrich C. Mayr Bernhard Thalheim (Eds.)
ER 2000 Workshops on Conceptual Modeling Approaches for E-Business and The World Wide Web and Conceptual Modeling Salt Lake City, Utah, USA, October 9-12, 2000 Proceedings
Series Editors Gerhard Goos, Karlsruhe University, Germany Juris Hartmanis, Cornell University, NY, USA Jan van Leeuwen, Utrecht University, The Netherlands Volume Editors Stephen W. Liddle Brigham Young University Marriott School, School of Accountancy and Information Systems 585 TNRB, P.O. Box 23087, 84602-3087 Provo, Utah, USA E-mail: [email protected] Heinrich C. Mayr University of Klagenfurt Insitute for Business Informatics and Application Systems Universit¨atsstr. 65-67. 9020 Klagenfurt, Austria E-mail: heinrich@it.uni-klu.ac.at Bernhard Thalheim Brandenburg University of Technology at Cottbus Computer Science Institute Postfach 101344, 03013 Cottbus, Germany E-mail: [email protected] Cataloging-in-Publication Data applied for Die Deutsche Bibliothek - CIP-Einheitsaufnahme Conceptual modeling for E-business and the web : proceedings / ER 2000 ; Workshops on Conceptual Modeling Approaches for E-Business and the World Wide Web and Conceptual Modeling, Salt Lake City, Utah, USA, October 9 - 12, 2000. Stephen W. Liddle . . . (ed.). - Berlin ; Heidelberg ; New York ; Barcelona ; Hong Kong ; London ; Milan ; Paris ; Singapore ; Tokyo : Springer, 2000 (Lecture notes in computer science ; Vol. 1921) ISBN 3-540-41073-2
CR Subject Classication (1998): H.2, H.3-5, C.2.4-5, J.1 ISSN 0302-9743 ISBN 3-540-41073-2 Springer-Verlag Berlin Heidelberg New York This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microlms or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable for prosecution under the German Copyright Law. Springer-Verlag Berlin Heidelberg New York a member of BertelsmannSpringer Science+Business Media GmbH © Springer-Verlag Berlin Heidelberg 2000 Printed in Germany Typesetting: Camera-ready by author, data conversion by PTP-Berlin, Stefan Sossna Printed on acid-free paper SPIN: 10722824 06/3142 543210
an e-business model is similar to a business process model, and so can be specied
commerce business case to be developed. It can be seen as a rst step in requirements notion of activity. Such models tend to be workow-oriented: they show the sequence branches in a workow sequence, parallel threads and synchronizations. Thus, a process these offerings are selected, negotiated, contracted and fullled
we nd them basically lacking. Therefore, we consider in Sec. 3 what the conceptual
is also pointed out in [9] in terms of benets and revenues.
tal improvement of processes (e.g. efciency); (3) support of processes by workow 5. 6.
what is the sequence of activities to be carried out for a specic case; which activities can be carried out in parallel for a specic case.
is possible to dene a set of generic concepts and relationships that together make up an an ontology for e-business models. Here, we briey survey the core concepts, only to (see below) actors makes prot. In a sound, actor is capable of making prot. protable add value and must be protable. The rationale for this is that during they are protable for them.
value exchange occurs between ports of opposite direction. A value object ows from an signicant ways from the workow-oriented ones in process models.
protable, and whether actors are willing to exchange objects of value with each protable activities, for instance to discuss new alternative assignments of such ac The rst two differences have been discussed in the previous sections. The latter two
conrmation or negative) conrmation after a submission of an ad. This conrmation object owing from the FAP to the contact searcher. A desired contact states a consumer Different subsets of object properties are identied for business models and process models. A business model identies those object properties, which can prole). Such a property partly determines the value a reader assigns to an ad read. of value objects. In a process model data ows and control ows are used The process model in Fig. 3 contains a data ow from the
experience of a desired contact to the FAP. This ow relates to the ow is necessary as itself. Note that in the process model, the conrmation ows from contact searcher to FAP, while the desired contact ows from the FAP to the contact No direct physical or information ow. A value exchange may coincide with a ow direct physical or information ow. The previous example illustrates this case also. Moreover, some value exchanges do not imply a physical or information ow directly, Value adding and protable. protable determine the amount of prot for each actor, and we want to address and discuss various activities if these are protable.
are protable. In [7] and [9], this is referred to as value chain deconstruction, as a way
receives a conrmation. Also, the ad is offered to the Ad Association, which pays for protable activities. The decomposition operation is dened as follows: (1) a value activity can is protable; (2) consider for each pair of sub-value activities new value interfaces and
nd contact . Both activities are likely to be protable for contact searchers; they both enhance the chance to nd a desired contact. The activity is considered to be protable; this is especially the case when a FAP process ow in detail, while in a business model it is led by a search for commercially business models, it should be possible to identify the protability of a business model Value adding and protable.
protability publisher, a redistributor, and an administration ofcer.
both are necessary for good e-business design, but they differ in several signicant operational fullment. 2. In an e-business model, an actor adds value and is protable, while in a process the steer the process ow. a process model a ow of information or goods implies a change of state; adding value and are protable. Decomposition of such activities is done to discover smaller chunks of activities that still add value and are protable. Discovering del, decomposition serves the goal of clarication of the workow or to show the
of concerns of stakeholders and claries the discussions.
The Unied Modelling Language Refe
A method for supporting WWW retrieval by constructing a exible viewpoint is a nite set of consistent category names. A set of documents retrie
which can extract only documents relevant to the user’s purpose of retrieval is difcult, provides a user with a human-classied collection information on ofcial research facilities in Nara which are responsible for environmental
In the above case, the user hesitates over which category to select rst, and even if the purpose of retrieval and one xed structure cannot cope with this versatility of criteria. exible that classies a subset of documents which are related to the user’s interest. Providing exibility the user’s purpose of retrieval. For the above example, where ofcial research facilities
gorization viewpoint is a nite set of consistent category names. A set of documents
and a query issued by the user is automatically rened to achieve higher precision and However, it seems difcult to decompose a document set in a consistent way and/or cluster appropriately, sufcient information on each cluster should be provided for the user. However, it is difcult to automatically associate a title or a summary with a cluster.
which roughly reect the user’s purpose of retrieval. (ii) Construct a category structure is a nite set of categories. For example, a categorization viewpoint and a nite set of category names [13] is used for dening the similarity coefcient between a
is dened as the frequency of similarity coefcient is dened as the
which has the maximum similarity coefcient is dened as follows:
dened as tion will be dened in subsection 2.3. In the decomposition based on the initial value smooth the categorization by redening the category vector
the nal result of the session or to further decompose of documents is dened as expression (2) when the clearness is used, and is dened as expression (3) when the entropy is used. The effectiveness of the two criteria belongs; that is, the similarity coefcient between
similarity coefcient between a categorization viewpoint resulting in high clearness does not always reect the user’s of documents is dened
which is the average of the similarity coefcient between each document and the category
, let us dene
A categorization viewpoint resulting in high entropy offers an efcient means of nar
retrieve documents containing specied keywords. The system also uses a Japanese mor To evaluate the method, we dened the following four experimental systems and we
are dened as follows:
is dened as follows [16]:
in Figure 5. In the gure, the curves labeled with Clearness, Entropy, Clustering and and can provide the decomposition which reects the user’s purpose of retrieval. When very low is not important since the recall is dened relative to not the entire document
recall and the precision becomes higher. By denition, the entropy of a categorization
Calculate the correlation coefcient of results in a signicant difference between clearness and entropy in the case of Averages of Correlation Coefcients between the Score and the Precision *1: The average of the correlation coefcient between the score and the precision for each topic.
In this paper, we proposed a method for supporting WWW retrieval using a exible
the proposed method. For example, it should be claried whether a user looses the advantage of being familiar with a xed category when documents are dynamically
11. Pollitt, A. S.: The key role of classication and indexing in view-based searching, in