2384 Patterns for Designing Agent-Based E-Business Systems events regarding the current auction (e.g., if they are winning or losing the auction). While buyer agents only notify a user about the current auc- WLRQQRWL¿HUDJHQWV² NOTIFICATION AGENT—can be used to monitor other auctions on a user’s behalf. 7KHXVHUFUHDWHVDQRWL¿HUDJHQWE\LQLWLDOL]LQJLW ZLWKDVSHFL¿FDWLRQRIWKHSURGXFWRUVHUYLFHVKHLV looking for. On receiving a NOTIFICATION about an auction, a user can decide to create a new buyer agent to join it. Locating Sellers by Product or Service This example switches the positions of buyer and seller agents in the online auction example. Bob wants to buy a certain product or service and, via his USER AGENT, creates a BUYER AGENT and LQLWLDOL]HVLWZLWKDVSHFL¿FDWLRQRIWKHSURGXFW or service. This buyer agent queries a MEDIATOR AGENT —in this case, a search agent or a directory agent—for the locations of sellers for the product or service in question. At this architectural level, we could use static or mobile agents to represent buyer agents. Greengrass et al. (1999) describe a mobile agent scenario in which a buyer agent moves to the known site of a directory agent, que- ries this directory agent, and obtains an itinerary of the locations of SELLER AGENTS to visit. Fonseca et al. (2001) document a mobile shopping system, where an agent representing a shopper in a mall contacts a directory facilitator agent on entry to the mall, and obtains directions to stores which carry products of interest to the user, provides those directions to the user, and allows him to check the prices of other stores. For further details on this case study, and an illustration of how the patterns can be used as part of a pattern-driven design process, see Weiss (2003). Customizing Navigation Another example is using agents to assist users in navigating through an (online) product cata- log. From a front-end perspective, a USER AGENT FDQEHXVHGWRFROOHFWSUR¿OHLQIRUPDWLRQDERXW a user and control how this information is made DYDLODEOHWRVHOOHUDJHQWV7KHSUR¿OHLQIRUPDWLRQ IRUGLIIHUHQWVHOOHUVLVVWRUHGLQDSHUVRQDOSUR¿OH r ep os ito r y i n t he u s er age nt — USER PROFILING. Sec- WLRQVRIWKHSUR¿OHFDQEHUHVWULFWHGWRDVXEVHWRI WKHVHOOHUVRQO\8VLQJWKHSUR¿OHLQIRUPDWLRQWKH VHOOHUDJHQWFDQFODVVLI\WKHXVHULQWRSUHGH¿QHG VWHUHRW\SHVDQGWKXVLQIHUDSUR¿OHRISURGXFWSUHI- erences— USER PROFILING. With these preferences, a product extractor agent— CATALOGING—can tailor the contents of the product catalog, such as in the personalized Web store in Ardissono HWDO,QDGGLWLRQWR¿OWHULQJWKHFDWDORJ FRQWHQWVDJDLQVWWKHEX\HU¶VSUR¿OHWKHSURGXFW Figure 10. Collaboration diagram for the agent-based online auction example 2385 Patterns for Designing Agent-Based E-Business Systems extractor can propose discounts to retain existing or attract new customers. Although we have focused on using user pro- ¿OHVLQWKHIURQWHQGWKLVLVQRWWRVD\WKDWWKH\ cannot equally be used in the back-end. However, the basic trade-off in this case is that users will have OHVVFRQWURORYHUSUR¿OHVFROOHFWHGDQGVWRUHGE\ sellers. Customers may decide to do business with VHOOHUVZKRGRQRWFROOHFWSUR¿OHVLQWKLVZD\RU PDNHWKHSURFHVVRIFROOHFWLQJSUR¿OHVWUDQVSDUHQW DQGWKHSUR¿OHVWKHPVHOYHVDFFHVVLEOH CONCLUSION In this chapter, we described a group of ar- chitectural patterns for designing agent-based e-business systems and gives several examples illustrating their use. These patterns relate to front-end e-business activities that involve in- teraction with the user and delegation of tasks from users to agents. Future work will describe pa t t er n s for b ack-e nd e- bu si ne ss ac t ivi tie s t h at d o not involve direct interaction with the user, but rather depict mechanisms for mediating between agents representing users. These patterns will expand on the description of mediator agents in this chapter. Together, these patterns are just the beginnings of a pattern language for agent-based e-business system design, based on our current understanding of the technology. As the use of agent technology in e-business matures, this lan- guage will evolve as well. It is our hope that the proposed set of patterns and the pattern format may provide a starting point for future effort in this direction. FUTURE RESEARCH DIRECTIONS To date, only a subset of the patterns of agent-based e-business systems have been documented. This provides many opportunities for future research. These opportunities can be grouped into two areas. On one hand, a better understanding of using agent patterns as part of a pattern-driven design process (Weiss, 2003) is required; we need to be able to reason about design trade-offs between patterns; see Mussbacher, Amyot, and Weiss (2007) for a typical approach and survey of related efforts, and capture pattern knowledge in pattern repositories (Knublauch & Rose, 2002) that can be consulted E\GHYHORSHUVWRDGGUHVVVSHFL¿FGHVLJQSUREOHPV during the development of an agent-based system. )LQDOO\ ZH ZDQW WR JHQHUDWH FRGH IRU VSHFL¿F agent frameworks from those patterns. On the other hand, many more agent patterns in the e-business domain remain to be mined and documented. Areas of particular interest are pat- terns where agents are used as mediators, ranging from directory agents to sophisticated broker and market maker agents (Deugo et al., 2001), and pat- terns for agents that provide mobile, context-aware services. One example of current research in this area is a set of auction patterns documented by Jureta, Faulkner, and Kolp (2005). However, this would also include patterns for search, reputation management, and integration. A taxonomy of e- business agents, which could provide a suitable starting point for a systematic mining effort is given in Papazoglou (2001). Conceptual frameworks for these activities are also required. Weiss (2003) proposes a pat- tern-driven design framework for harvesting recurring design solutions and documenting them as patterns, and guiding the designer through the VHOHFWLRQRISDWWHUQVDSSURSULDWHWRWKHLUVSHFL¿F GHVLJQFRQWH[W7KHDSSURDFKVXJJHVWVD¿YHVWHS SURFHVVWKH¿UVWWKUHHUHODWHGWRPLQLQJSDWWHUQV identify the forces in a domain, document the roles of pattern participants, and document pat- terns and their dependencies. The last two apply to patterns: identify the overall design goals and select patterns. Finally, we need to document patterns for nonfunctional design issues such as deployment, scalability, and security of agent-based e-business systems, and we must gain a better understanding 2386 Patterns for Designing Agent-Based E-Business Systems of how to integrate such patterns with the current agent development processes. Some progress along these lines has been made in the area of agent security patterns. A good starting point is Mouratidis et al. (2006). ACKNOWLEDGMENT An earlier version of this chapter was accepted for PLoP 2001 (Weiss, 2001). Thanks go to my shepherd Dirk Riehle for his valuable comments. 7KHPDWHULDOLQWKLVFKDSWHUKDVEHQH¿WHGIURP feedback in several tutorials and classes held since RQWKLVPDWHULDODQGZDVH[SDQGHGWRUHÀHFWWKH recent progress in the area of agent patterns. REFERENCES Alexander, C. (1979). The timeless way of build- ing. Oxford University Press. Aranguren, M. (2005). Ontology design patterns for the formalisation of biological ontologies. Master’s Thesis, Manchester University. Ardissono, L., Barbero, C., et al. (1999). An agent architecture for personalized Web stores. In Proceedings of the Conference on Autonomous Agents (pp. 182-189). ACM. Aridor, Y., & Lange, D. (1998). Agent design patterns: Elements of agent application design. In Proceedings of the Conference on Autonomous Agents (pp. 108-115). ACM. AuctionWatcher (2002). Retrieved August 14, 2007, from http://www.auctionswatch.info Avgeriou, P., & Zdun, U. (2005). Architectural patterns revisited: A pattern language. In Pro- ceedings of the European Conference on Pattern Languages of Programs. Brenner, W., Zarnekow, R., & Wittig, H. (1998). Intelligent software agents: Foundations and applications. Springer. Carlson, D. (2001). Modeling XML applications with UML: Practical e-business applications. Addison-Wesley. Collis, J., & Lee, L. (1999). Building electronic marketplaces with the ZEUS agent tool-kit. Lec- WXUH1RWHVLQ$UWL¿FLDO,QWHOOLJHQFH 1-24. Deugo, D., Weiss, M., & Kendall, E. (2001). Reusable patterns for agent coordination. In A. Omicini et al. (Eds.), Coordination of Internet Agents (pp. 347-368). Springer. Ferber, J. (1999). Multi-agent systems: An intro- GXFWLRQWRGLVWULEXWHGVUWL¿FLDOLQWHOOLJHQFH (pp. 13-16). Addison-Wesley. Fonseca, S., Griss, M., & Letsinger, R. (2001). An agent-mediated e-commerce environment for the mobile shopper (HP Labs Tech. Rep. No. HPL-2001-157). Greengrass, E., Sud, J., & Moore, D. (1999, June). Agents in the virtual marketplace: Component dtrategies. In Proceedings of SIGS (pp. 42-52). Griss, M. (1999). My agent will call your agent but will it respond? (HP Labs Tech. Rep. No. HPL-1999-159). Guttman, R., & Maes, P. (1999). Agent-mediated integrative negotiation for retail electronic com- merce. /HFWXUH1RWHV LQ$UWL¿FLDO ,QWHOOLJHQFH 1571, 70-90. Jureta, I., Faulkner, S., & Kolp, M. (2005). Best practices agent patterns for on-line auctions. In Proceedings of the International Conference on Enterprise Information Systems (pp. 814-822). Kendall, E. (1999). Role models: Patterns of agent system analysis and design. In Proceedings of the Agent Systems and Applications/Mobile Agents. ACM. Kendall, E., Murali Krishna, P., Pathak, C., et al (1998). Patterns of intelligent and mobile agents. In Proceedings of the Conference on Autonomous Agents. ACM. 2387 Patterns for Designing Agent-Based E-Business Systems Kolp, M., Giorgini, P., & Mylopoulos, J. (2001). A goal-based organizational perspective on multi-agent architectures. In Proceedings of the Workshop on Agent Theories, Architectures, and Languages (LNCS 2333, pp. 128-140). Springer. Lieberman, H. (1998). Integrating user interface agents with conventional applications. Paper presented at the Conference on Intelligent User Interfaces. ACM. Lind, J. (2002). Patterns in agent-oriented software engineering. In Proceedings of the Workshop on Agent-Oriented Software Engineering (LNCS 2585, pp. 45-58). Springer. Maes, P. (1994, July). Agents that reduce work and information overload. Communications of the ACM, pp. 31-41. Mouratidis, H., Weiss, M., & Giorgini, P. (2006). Modelling secure systems using an agent-oriented approach and security patterns. International Journal on Software Engineering and Knowledge Engineering, 16(3), 471-498. Mussbacher, G., Amyot, D., & Weiss, M. (2007). Formalizing patterns with the user requirements notation. In T. Taibi (Ed.), Design Pattern Formal- ization Techniques. Hershey, PA: IGI Global. Noriega, P., & Sierra, C. (1999). Agent-mediated electronic commerce./HFWXUH1RWHVLQ$UWL¿FLDO Intelligence, 1571. Springer. Papazoglou, M. (2001, April). Agent-oriented technology in support of e-business. Communica- tions of the ACM, pp. 71-77. Reich, J. (2000). Ontological design patterns: Modelling the metadata of molecular biological ontologies, information and knowledge. Paper presented at the Conference on Database and Expert System Applications. Riehle, D., & Gross, T. (1998). Role model based framework design and integration. In Proceedings of the Conference on Object-Oriented Programs, Systems, Languages, and Applications (pp. 117- 133). ACM. Schelfthout, K., Coninx, T., et al (2002). Agent implementation patterns. Paper presented at the OOPSLA Workshop on Agent-Oriented Meth- odologies. Shu, S., & Norrie, D. (1999). Patterns for adaptive multi-agent systems in intelligent manufacturing. Paper presented at the International Workshop on Intelligent Manufacturing Systems. Staab, S., & Stuckenschmidt, H. (Eds.). (2006). Semantic Web and peer-to-peer. Springer. Steinmetz, E., Collins, J., Jamieson, S., et al. (1999). Bid evaluation and selection in the MAG- NET automated contracting system. Lecture Notes LQ$UWL¿FLDO,QWHOOLJHQFH, 105-125. Tahara, Y., Oshuga, A., & Hiniden, S. (1999). Agent system development method based on agent patterns. Paper presented at the International Conference on Software Engineering. ACM. Ulmer, D. (2004). Architectural solutions to agent-enabling e-commerce portals with pull/ push abilities. Unpublished doctoral thesis, Pace University. Voss, A., & Kreifelts, T. (1997). SOaP: Social agents providing people with useful information. Paper presented at the Conference on Supporting Groupwork. ACM. W3C. (1997). 2SHQSUR¿OLQJVWDQGDUG(version 1.0, note 1997/6). Weiss, M. (2001). Patterns for e-commerce agent architectures: Agents as delegates. Paper pre- sented at the Conference on Pattern Languages of Programs. Weiss, M. (2003). Pattern-driven design of agent systems: Approach and case study. In Proceed- ings of the Conference on Advanced Information System Engineering (pp. 711-723). Springer. 2388 Patterns for Designing Agent-Based E-Business Systems Weiss, M. (2004). A pattern language for mo- tivating the use of agents. In Agent-Oriented Information Systems: Revised Selected Papers (LNAI 3030, pp. 142-157). Springer Weiss, M., Gray, T., & Diaz, A. (1997). Experi- ences with a service environment for distributed multimedia applications. feature interactions in telecommunications and distributed systems, IOS, 242-253. Wooldridge, M., & Jennings, N. (1995). Intelli- gent agents: Theory and practice. The Knowledge Engineering Review, 10(2), 115-152. Zacharia, G., Moukas, A., & Maes, P. (1999). Col- laborative reputation mechanisms in electronic marketplaces. Paper presented at the Hawaii International Conference On System Science. IEEE. Zambonelli, F., Jennings, N., et al. (2001). Agent- oriented software engineering for Internet applica- tions. In A. Omicini, et al. (Eds.), Coordination of Internet Agents (pp. 326-346). Springer. ADDITIONAL READING Deugo, D., Kendall, E., & Weiss, M. (1999). Agent pPatterns. Tutorial at the International Symposium on Agent Systems and Applications/Mobile Agents. Retrieved August 15, 2007, from http://www.scs. carleton.ca/~deugo/Patterns/Agent/Presentations/ AgentPatterns Jureta, I., Faulkner, S., & Kolp, M. (2005). Best practices agent patterns for on-line auctions. In- ternational Conference on Enterprise Information Systems (pp. 814-822). Knublauch, H., & Rose, T. (2002). Tool-supported process analysis and design for the development of multiagent systems. Workshop on Agent-Ori- ented Software Engineering (LNCS 2585, pp. 186-197). Springer. Mussbacher, G., Amyot, D., & Weiss, M. (2007). Formalizing patterns with the user requirements notation. In T. Taibi (Ed.), Design Pattern Formal- ization Techniques. Hershey, PA: IGI Global. ENDNOTE 1 It is customary to indicate references to patterns through a SMALL CAPS font. This work was previously published in Agent Systems in Electronic Business, edited by J. Wang, pp. 1-24, copyright 2008 by Information Science Reference (an imprint of IGI Global). 2389 Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited. Chapter 8.4 Dialogue Act Modeling: An Approach to Capturing and Specifying Communicational Requirements for Web-Based Information Systems Ying Liang University of Paisley, UK ABSTRACT Web-based information systems (WBIS) aim to support e-business using IT, the World Wide Web, and the Internet. This chapter focuses on the Web site part of WBIS and argues why an easy-to-use and interactive Web site is critical to the success of WBIS. A dialogue act modeling approach is presented for capturing and specifying user needs for easy-to-use Web site of WBIS by WBIS analysis; for example, what users want to see on the computer screen and in which way they want to work with the system interactively. It calls such needs communicational requirements, in addition to functional and nonfunctional requirements, and builds a dialogue act model to specify them. The author hopes that development of the Web site of WBIS will be considered not only an issue in WBIS design but also an issue in WBIS analysis in WBIS development. INTRODUCTION Web-based information system (WBIS) is a new type of information system that uses information technology, the World Wide Web (WWW), and the Internet to support e-business and information source management worldwide. They provide a new way of managing, manipulating, exchang- ing, sharing, and supplying global information and services online. They enable customers and companies worldwide to communicate with each other through the Internet and to demand and supply business information resources and services around the world without meeting each other. Appearance and use of WBIS in business have changed people’s lives because it brings a new culture into business. People living with this new culture do not have to buy things in local shops; instead, they can buy things in global e- shops using the Internet. Suzuki (1997) in general 2390 Dialogue Act Modeling GH¿QHVFXOWXUHDVWKHUHVSRQVHSDWWHUQVKDUHGE\D VSHFL¿FJURXSRISHRSOHZKLFKLVVKDSHGWKURXJK interaction with the environment. Furthermore, 5DWQHUVSHFL¿FDOO\GH¿QHVJOREDOFXOWXUH as what is common to all human beings and the response pattern as how people interact with the Web site in the context of the Web site. Based on WKH6X]XNLDQG5DWQHUGH¿QLWLRQV5DWQHU S:%,6FXOWXUHLVGH¿QHGLQWKLVFKDSWHU as follows: WBIS culture is what is common to all users of WBIS and that the “response pattern” is how the users will interact with the Web site of WBIS in the context of WBIS. Such response patterns can be perceived as an interactive communication procedure of getting the goal of a user’s task by using WBIS within a business context. A well-accepted WBIS should allow the user to control the interactive commu- nication procedure and to decide how to interact with the Web site of WBIS. It is agreed that in order to have an easy-to-use Web site, WBIS basically needs to be user-centered and inter- active as a computerized business tool linking customers to companies or linking companies to companies in e-business and information source management. They must be attractive to online users and compatible with other similar systems. Usability of WBIS relies greatly on the developer’s understanding of the user’s need for the Web site of WBIS (e.g., what users want to see on the computer screen and in what way they want to work with the system interactively). Un- fortunately, understanding such a need currently is ignored in WBIS analysis because usability is often thought of as an issue in design rather than in analysis. For example, Cato (2001) emphasizes that the developer needs to pay greater atten- tion to user interface and interaction design in WBIS design if he or she wants a system to be effective. However, in principle, understanding a user’s need is the task of system analysis but not the task of system design. This means that the user requirements for effective interactive communication with the Web site must be cap- tured in WBIS analysis. This is not supported by current modeling approaches, and new modeling approaches are needed for this purpose. Actually, a new modeling approach called dialogue act modeling (Liang, 2004, 2005a, 2005b) has been created specially by the author for capturing and specifying such user need and will be introduced in this chapter. The objective of the chapter is to help the WBIS developer (in particular, WBIS analysts) to understand the following: • Different cultures of WBIS and traditional IS. • Role of the Web site in the WBIS culture and its impact on user-centered and interactive WBIS. • Need of considering usability of the WBIS Web site in WBIS analysis. • Need of understanding the user need for easy-to-use Web site in WBIS analysis. • The dialogue act modeling approach and how to use it to capture and specify such user need in WBIS analysis. WBIS CULTURE AND ITS IMPACT ON WBIS DEVELOPMENT Observation has shown many differences between WBIS and traditional IS (Deshpande et al., 2002) and, unlike traditional IS, WBIS provides a Web site as the communication medium to interact with the user over the Internet. This drives the two systems to run in different environments with d i f f e r e n t c u l t u r e s (i .e ., w h a t i s c o m m on t o t h e u s e r and that the response pattern is how the user will interact with the system in its context are differ- ent). Figure 1(a) and 1(b) show that the response pattern of traditional IS culture is how the user will interact with IS directly in the context of IS, whereas the response pattern of WBIS is how the 2391 Dialogue Act Modeling user will interact with IS through the Web site of WBIS in the context of WBIS. The WBIS culture requires WBIS analysis to involve users much more than traditional IS analy- sis and requires WBIS development to make WBIS ZLWKWKHIROORZLQJVSHFL¿FFKDUDFWHULVWLFV • WBIS should have a Web site and Web pages: T he u se r of W BIS u se s t he We b sit e to get information and services on the Internet, and WBIS uses Web pages to display the result on the computer screen. The Web site must enable the user to complete his or her tasks effectively within the minimum time. Thus, developing a Web site and Web pages in consultation with users is vital to Web site success (Lawrance, Gorbitt, Tidwell, Fisher, & Lawrance, 1998). • WBIS have an unlimited number of varied users: Users of WBIS are unlimited and live in different places of the world (Conallen, 2003). This makes it impossible to train all of them in training sessions. Untrained us- ers may have fear and anxiety when using WBIS. Thus, easy-to-use Web sites of WBIS is critical to such a user. • WBIS often have global users: WBIS in principle is used by anyone, anywhere in the world. Global users normally have differ- ent cultural backgrounds and use different languages (Ratner, 2003). It has been found that the possibility of information systems failing through culture shock increases be- cause of business globalization and creation of WBIS connecting the world. When it becomes apparent that some users do not behave as expected, the problem of what to do next becomes crucially important in Web site development (Szewczak & Snodgrass, 2002). • WBIS have a volatile development pro- cess: This makes requirements analysis DQGVSHFL¿FDWLRQGLI¿FXOW%DUU\</DQJ 2001). These characteristics of WBIS make WBIS GHYHORSPHQW PRUH GLI¿FXOW WKDQ WUDGLWLRQDO ,6 development, because the problem of the devel- opment is no longer constructing the technical infrastructure of the system but is creating a good Web site that is easy-to-use and meets the Figure 1(a). Traditional IS culture Figure 1(b). WBIS culture TIS culture interacts Limited, trained and local user IS WBIS culture interacts Unlimited, untrained and global user WBIS Website part IS part Web page 2392 Dialogue Act Modeling user’s needs (Lazar, 2001). Analysis and design of WBIS are inevitable to face this problem. A lot of effort already has been made to solve it in WBIS design (Burdman, 1999; Corry, Frick, & Harsen, 1997; Fuccella, 1997; Lazar, 2001; Lynch & Horton, 1999), but not much has been done for this in WBIS analysis. Thus, the WBIS analyst has no approach available to capture and understand the user’s needs for easy-to-use Web site in WBIS analysis, although it is the task of analysis in software engineering. In order to help to improve the situation, this chapter will address this issue and represent a new modeling approach called dialogue act modeling approach as a solu- tion to this problem. The author hopes it is useful to capture and specify the user’s need for the Web site of WBIS in WBIS analysis. THE DIALOGUE ACT MODELING APPROACH: A WAY OF COPING WITH THE WBIS CULTURE IN WBIS DEVELOPMENT The dialogue act modeling approach presented in this chapter aims to cope with the WBIS culture in WBIS development. It provides a new modeling technique for capturing and specifying interactive communications between the user and WBIS in WBIS analysis with the concern about usability of the Web site of WBIS, because easy-to-use and the usefulness of the technology have the most LQÀXHQFHRQWKHXVHURIFRPSXWHUL]HGV\VWHPVLQ general (Davis, 1989), and they also are applied to the user of a Web site (Lederer, Maupin, Sena, & Zhuang, 1998). IS needs to be developed around the user’s needs (Norman & Draper, 1986), which becomes even more critical in WBIS development because traditional IS without Web can control where a user can go, but WBIS cannot do so, as users of WBIS want to be active when they are on the Web, and they even can take paths that were never thought or intended by the developer (Nielsen, 2000). They intend to control the paths toward the goal of their tasks on the Web and to c o m p l e t e t h e i r t a s k s i n a m i n i m u m a m o u n t of t i m e with a minimal amount of frustration. Such a user need should be understood in order to avoid the failure of WBIS, because some of e-business Web sites have failed completely and have been shut down because of poor usability (Ratner, 2003). Users will not purchase things from a Web site that is complicated and frustrating to use (Lazar, 2001), and they are extremely goal-driven on the :HE1LHOVHQ'LI¿FXOWWRXVH:HEVLWHV of WBIS can cause computer anxiety when us- ers use the computer, which is thought of as one of the most prevalent emotions experienced by users (Downton, 1993; Negron, 1995). However, it can be remedied through positive computing experiences (Cambre & Cook, 1987) with an easy-to-use Web site. How to make easy-to-use Web sites for WBIS is a big challenge of WBIS development, because it is quite hard to produce a useful product that is easy to learn and use (Cato, 2001). It needs ef- fective development approaches and techniques such as modeling techniques that enable WBIS developers to consider the user’s attitude and motivation and to allow users to build interactions in the way they like and create their own power over interaction. Traditional IS development ap- proaches and techniques are not really proper or adequate to WBIS development because of the dynamic and evolving nature of WBIS (Lang, 2002; Zelnic, 1998). The dialogue act modeling approach was created, therefore, for overcoming the lack and by considering the following: +RZ WR FUHDWH D :%,6 VSHFL¿FDWLRQ WKDW will lead to a success of user-centered and interactive WBIS in development. • What kind of modeling techniques can help to promote usability of the Web site of WBIS. • How to make users involved in Web site analysis. 2393 Dialogue Act Modeling ([SHUWVLQWKLV¿HOGKDYHDWWHPSWHGWRFUHDWH a new discipline, Web Engineering,for research DQGHVWDEOLVKPHQWRIVRXQGVFLHQWL¿FHQJLQHHU- ing, and management principles and system- atic approaches to successful development and maintenance of high quality WBIS (Murugesan, Deshpande, Hansen, & Ginige, 2001). Engineer- ing a Web application is to diversify problems to application domain analysis, navigational structures, and user interface design (Conallen, 2003). The dialogue act modeling approach falls LQWRWKHDSSOLFDWLRQGRPDLQDQDO\VLVVSHFL¿FDOO\ in Web Engineering, as illustrated in Figure 2. It will analyze the Web site part of WBIS through dialogue act modeling and analyze the IS part of WBIS by object modeling and behavior model- ing (Booch, Rumbaugh, & Jacobson, 2005). The analysis process was established based on the RQHGH¿QHGE\6RPPHUYLOOH7KHGHWDLOV of the approach are described as different parts in the following. Part One: Pragmatic View and De- scriptive View Used by the Approach The pragmatic view and the descriptive view are the two typical modeling views used in traditional IS modeling in the past: • Pragmatic view: Used for observing the pragmatic aspects of IS (e.g., business prop- erties such as customers) as part of reality within the business context. The action ZRUNÀRZ DSSURDFK 'HQQLQJ 0HGLQD Mora, 1995) used this view. It has much focus on pragmatic concepts but little focus on semantics of the system (Agerfalk, 2002; Erickson & Kellogg, 2000; Eriksen, 2002; Holm & Ljungberg, 1996). • Descriptive view: Used for observing the semantic aspects of IS (e.g., business pro- cesses and entities) as an image of reality. Data modeling (Chen, 1976), process mod- eling (DeMacro, 1978), and object model- ing (Booch, 1994; Coad & Yourdon, 1991; Jacobson, 1992; Liang, 2003; Rumbaugh, Blaha, Premerlani, Eddy, & Lorensen, 1991) used this view. In general, it has much focus on business processes but little focus on business properties such as customers. Nearly every existing modeling approach used one of these two modeling views only but not both in IS modeling because they aimed to focus on either pragmatics or semantics of a system during analysis. However, the dialogue act modeling approach presented in this chapter dialogue act modelling, object modelling, behaviour modelling Pragmatic view Descriptive view (class diagram) (business context map, (sequence diagram, dialogue act diagrams, state diagram) user interface tree) Requirements specification documents Business contexts Object model (WBIS structure) Behaviour model (WBIS behaviour) Dialogue act model (WBIS communication) Business case study Requirements analysis System specification Specification validation )LJXUH'LDORJXHDFWPRGHOLQJDSSURDFKIRU:%,6UHTXLUHPHQWVDQDO\VLVDQGVSHFL¿FDWLRQ . information and services online. They enable customers and companies worldwide to communicate with each other through the Internet and to demand and supply business information resources and services. Foundations and applications. Springer. Carlson, D. (2001). Modeling XML applications with UML: Practical e-business applications. Addison-Wesley. Collis, J., & Lee, L. (1999). Building electronic. information and knowledge. Paper presented at the Conference on Database and Expert System Applications. Riehle, D., & Gross, T. (1998). Role model based framework design and integration.