Zusammenfassung
Unternehmensportale sollen Geschäftsprozesse unterstützen und die Produktivität der Mitarbeiter steigern. Die erwartete Produktivitätssteigerung wird allerdings nur dann erfüllt werden, wenn die Nutzer hinreichend über die Möglichkeiten des Unternehmensportals informiert sind. Diese Problematik betrifft vor allem große Unternehmensportale deren Dienstangebot sich ständig weiterentwickelt und zu denen oft neue Nutzer hinzugefügt werden. In dem Artikel wird ein Empfehlungssystem für Unternehmensportale vorgeschlagen, um die Wahrnehmung und Nutzung für Dienste zu steigern. Dem gestaltungsorientierten Ansatz folgend wird ein passendes Empfehlungskonzept entwickelt und mehrere Implementierungsmöglichkeiten in einem Feldexperiment bei einem der größten deutschen Unternehmen evaluiert. Es wird dargelegt, dass das Empfehlungssystem die Anzahl der neu aufgerufenen Dienste und ebenso die Anzahl der neu genutzten Dienste im betrachteten Unternehmensportal um etwa 20 % steigern konnte.
Abstract
Corporate portals are supposed to support a company’s business model and to increase productivity of the employees. However, the productivity gain that can be achieved by corporate portals is often undermined because the users of the portal are not sufficiently informed about the portal’s capabilities. This is of particular concern for large corporate portals whose service portfolio is constantly evolving and to which new users are added frequently. In the article, we propose a recommender system for corporate portals in order to increase service awareness and usage. Following the design science methodology, a suitable recommender concept is developed and several implementation options are evaluated in a field experiment at one of Germany’s largest companies. It is found that the recommender system increases the number of newly visited services as well as the number of newly used services in the corporate portal by about 20 %.





Notes
Eine ausführliche deskriptive Statistik dieses Unternehmensportals findet sich in Abschn. 4.
So haben beispielsweise verschiedene Abteilungen manchmal unterschiedliche Formulierungen für die gleichen Sachverhalte. Dies müsste in der Dienstbeschreibung und den Präferenzen der Nutzer berücksichtigt werden. Darüber hinaus können komplexe und spezialisierte Dienste oftmals nicht einfach textuell beschrieben werden.
Jede Vorhersage p j jedes Objekts j wird mit der tatsächlichen Bewertung r j mittels \(e_{\mathrm{MAE}} = \frac{1}{M} \sum_{j=1}^{M} |p_{j} - r_{j}|\), beziehungsweise \(e_{\mathrm{MSE}} = \sqrt{\frac{1}{M} \sum_{j=1}^{M} (p_{j} - r_{j})^{2}}\) verglichen, wobei M ist die Anzahl an Vorhersagen ist.
Diese Statistik beruht auf insgesamt 466 neuen Nutzern in einem Zeitintervall von Dezember 2009 bis Juli 2010. Vor dieser Periode konnten neue Nutzer nicht zweifelsfrei von wiederholten Nutzern unterschieden werden. Neue Nutzer, die später zur Beobachtungsgruppe hinzugefügt wurden, wurden nicht berücksichtigt.
Die ANOVA ist äquivalent zu dem bekannten t-Test, sofern nur eine Gruppenvariable mit zwei Ausprägungsmerkmalen existiert. Da hier ein vollständiger Versuchsplan mit drei Gruppenvariablen (Bewertungsregel, Ähnlicheitsmaß, Vorhersagealgorithmus) vorliegt, von denen jede zwei oder mehr Merkmalsausprägungen annehmen kann, ist die ANOVA der angemessenere Test. Ähnlich wie bei einem t-Test ist die Nullhypothese, dass die mittlere Vorhersagequalität unter den unterschiedlichen Ausprägungen der Gruppenvariable identisch ist. Unter dieser Nullhypothese besitzt der ANOVA-Test eine F-Verteilung. Die ANOVA leitet neben einem F-Wert für jede Gruppenvariable, die den Haupteffekt charakterisiert, auch einen F-Wert für alle möglichen Kombinationen der Gruppenvariablen ab, was Interaktionseffekt genannt wird. Sind die Freiheitsgrade der Gruppenvariable (also die Anzahl Merkmalsausprägungen minus Eins) und die verbleibenden Freiheitsgrade des Modells (also die Anzahl an Beobachtungen minus der Summe der Freiheitsgrade der Gruppenvariablen minus Eins: hier 89) bekannt, so kann der p-Wert (Signifikanzniveau) berechnet werden.
Zur Berechnung des Bonferroni-korrigierten p-Werts, wird der unkorrigierte p-Wert mit der Gesamtanzahl an paarweisen Vergleichen multipliziert. Falls das Ergebnis 1.0 übersteigt, wird der korrigierte p-Wert mit 1.0 angegeben.
In diesem Fall stimmt die ANOVA mit dem t-Test überein. Um eine einheitliche Darstellung zu gewährleisten, beziehen wir uns jedoch auf eine ANOVA.
Literatur
Adomavicius G, Tuzhilin A (2005) Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering 17(6):734–749
Arbin K, Essler U (2002) Emerging industrial eMarkets: the case of covisint in Europe. In: Proceedings of the 15th International electronic commerce conference, Bled, S 279–292
Ashenfelter O, Card D (1985) Using the longitudinal structure of earnings to estimate the effect of training programs. The Review of Economics and Statistics 67(4):648–660
Balabanović M, Shoham Y (1997) Fab: content-based, collaborative recommendation. Communications of the ACM 40(3):66–72
Baudisch P (1999) Joining collaborative and content-based filtering. In: Online proceedings of interacting with recommender systems, CHI ’99 workshop
Bell RM, Koren Y (2007) Improved neighborhood-based collaborative filtering. In: KDDCup’07, San Jose
Benbya H, Passiante G, Belbaly NA (2004) Corporate portal: a tool for knowledge management synchronization. International Journal of Information Management 24(3):201–220
Berkovsky S, Eytani Y, Manevitz LM (2007) Efficient collaborative filtering in content-addressable spaces. International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI) 21(2):265–289
Breese JS, Heckerman D, Kadie C (1998) Empirical analysis of predictive algorithms for collaborative filtering. In: Proceedings of the 14th annual conference on uncertainty in artificial intelligence (UAI-98). Morgan Kaufmann, San Francisco, S 43–52
Buono P, Costabile MF, Guida S, Piccinno A (2002) Integrating user data and collaborative filtering in a web recommendation system. In: Reich S (Hrsg) Hypermedia: openness, structural awareness, and adaptivity. Springer, Berlin
Burke R, Felfernig A, Gker M (2011) Recommender systems: an overview. AI Magazine 32(3):13–18
Calderón-Benavides ML, González-Caro CN, de Pérez-Alcázar JJ, Garciaz JC, Delgado J (2004) A comparison of several predictive algorithms for collaborative filtering on multi-valued ratings. In: Proceedings of the 2004 ACM symposium on applied computing (SAC ’04). ACM, New York, S 1033–1039
Davydov MM (2001) Corporate portals and e-business integration. Emerging business technology series. McGraw-Hill, New York
Delgado J, Nahoiro I (1999) Memory-based weighted-majority prediction for recommender systems. In: Workshop on recommender systems: algorithms and evaluation (SIGIR ’99), Berkeley
Dias C (2001) Corporate portals: a literature review of a new concept in information management. International Journal of Information Management 21(4):269–287
Eckerson W (1999) Plumtree blossoms: new version fulfils enterprise portal requirements. Whitepaper. Boston
Ekstrand M, Riedl J, Konstan J (2011) Collaborative filtering recommender systems. Now Publishers, Boston
Elsner H, Krämer J (2010) Einsatz von Recommender Systemen zum Management von Unternehmensportalen. In: Schumann M, Kolbe L, Breitner M, Friedrichs A (Hrsg) Multikonferenz Wirtschaftsinformatik 2010. Universitätsverlag, Göttingen, S 1819–1830
Firestone JM (2000) Defining the enterprise information portal. http://www.tgc.com/dsstar/00/0822/102054.html. Abruf am 2009-04-02
Girgensohn A, Shipman F, Chen F, Wilcox L (2010) DocuBrowse: faceted searching, browsing, and recommendations in an enterprise context. In: Proceedings of the 15th international conference on intelligent user interfaces. ACM, New York, S 189–198
Goldberg D, Nichols D, Oki BM, Terry D (1992) Using collaborative filtering to weave an information tapestry. Communications of the ACM 35(12):61–70
Hazra TK (2002) Building enterprise portals: principles to practice. In: Proceedings of the 24th international conference on software engineering (ICSE 2002), S 623–633
Herlocker JL, Konstan JA, Borchers A, Riedl J (1999) An algorithmic framework for performing collaborative filtering. In: Proceedings of the 22nd annual International ACM SIGIR conference on research and development in information retrieval (SIGIR ’99). ACM, New York, S 230–237
Herlocker JL, Konstan JA, Terveen LG, Riedl JT (2004) Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems 22(1):5–53
Hevner A, March S, Park J, Ram S (2004) Design science in information systems research. MIS Quarterly 28(1):75–105
Hill W, Stead L, Rosenstein M, Furnas G (1995) Recommending and evaluating choices in a virtual community of use. In: Proceedings of the SIGCHI conference on human factors in computing systems (CHI ’95). Addison-Wesley, New York, S 194–201
Jannach D, Zanker M, Felfernig A, Friedrich G (2010) Recommender systems: an introduction. Cambridge University Press, Cambridge
Knorr E (2004) The new enterprise portal: the browser-based portal is fast becoming the enterprise UI and the nexus for a new breed of integration and app dev. http://www.infoworld.com/d/developer-world/new-enterprise-portal-131. Abruf am 2009-12-06
Konstan JA, Miller BN, Maltz D, Herlocker JL, Gordon LR, Riedl JT (1997) GroupLens: applying collaborative filtering to Usenet news. Communications of the ACM 40(3):77–87
Linden G, Smith B, York J (2003) Amazon.com recommendations: item-to-item collaborative filtering. IEEE Internet Computing 7(1):76–80
Malinowski J, Weitzel T, Keim T (2008) Decision support for team staffing: an automated relational recommendation approach. Decision Support Systems 45(3):429–447
Montaner M, López B, de La Rosa JL (2003) A taxonomy of recommender agents on the internet. Artificial Intelligence Review 19(4):285–330
Mulvenna MD, Anand SS, Buchner AG (2000) Personalization on the net using web mining: introduction. Communications of the ACM 43(8):122–125
Murray G (1999) The portal is the desktop. Intraspect Inc, Los Altos
Pazzani MJ (1999) A framework for collaborative, content-based and demographic filtering. Artificial Intelligence Review 13(5–6):393–408
Pazzani MJ, Billsus D (2007) Content-based recommendation systems. In: Brusilovsky P, Kobsa A, Nejdl W (Hrsg) The adaptive web. Springer, Heidelberg, S 325–341
Peffers K, Tuunanen T, Rothenberger M, Chatterjee S (2007) A design science research methodology for information systems research. Journal of Management Information Systems 24(3):45–77
Remus U (2007) Critical success factors for implementing enterprise portals: a comparison with ERP implementations. Business Process Management Journal 13(4):538–552
Resnick P, Iacovou N, Suchak M, Bergstrom P, Riedl J (1994) GroupLens: an open architecture for collaborative filtering of netnews. In: Proceedings of the 1994 ACM conference on computer supported cooperative work (CSCW ’94). ACM, New York, S 175–186
Runte M (2000) Personalisierung im Internet: Individualisierte Angebote mit Collaborative Filtering. PhD thesis, Christian-Albrechts-Universität zu Kiel
Sarwar BM, Konstan JA, Borchers A, Herlocker JL, Miller B, Riedl J (1998) Using filtering agents to improve prediction quality in the GroupLens research collaborative filtering system. In: Proceedings of the 1998 ACM conference on computer supported cooperative work (CSCW ’98). ACM, New York, S 345–354
Sarwar B, Karypis G, Konstan J, Reidl J (2001) Item-based collaborative filtering recommendation algorithms. In: Proceedings of the 10th international conference on world wide web (WWW ’01). ACM, New York, S 285–295
Schafer JB, Frankowski D, Herlocker J, Sen S (2007) Collaborative filtering recommender systems. In: Brusilovsky P, Kobsa A, Nejdl W (Hrsg) The adaptive web. Springer, Heidelberg, S 291–324
Shardanand U, Maes P (1995) Social information filtering: algorithms for automating “word of mouth”. In: Proceedings of the ACM conference on human factors in computing systems (CHI ’95), S 210–217
Shilakes CC, Tylman J (1998) Enterprise information portals. Merrill Lynch in-depth report
Sullivan D (2003) Proven portals: best practices for planning, designing, and developing enterprise portals. Addison-Wesley, Boston
Terra JC, Gordon C (2003) Realizing the promise of corporate portals: leveraging knowledge for business success. Butterworth-Heinemann, Amsterdam
Vo HTK (2007) Engineering corporate portals. PhD thesis, Universität Karlsruhe (TH), Karlsruhe
Weiser M, Brown J (1996) Designing calm technology. PowerGrid Journal 1(1):75–85
White C (2003) Is the portal dead? Information Management Magazine (July)
White M (2000) Enterprise information portals. The Electronic Library 18(5):354–362
Wisneski C, Ishii H, Dahley A, Gorbet M, Brave S, Ullmer B, Yarin P (1998) Ambient displays: turning architectural space into an interface between people and digital information. In: Streitz NA, Konomi S, Burkhardt H-J (Hrsg) Cooperative buildings: integrating information, organization, and architecture. Springer, Heidelberg, S 22–32
Zanker M, Jannach D, Gordea S, Jessenitschnig M (2007) Comparing recommendation strategies in a commercial context. IEEE Intelligent Systems 22(3):69–73
Zhen L, Huang G, Jiang Z (2009) Recommender system based on workflow. Decision Support Systems 48(1):237–245
Zhen L, Huang G, Jiang Z (2010) An inner-enterprise knowledge recommender system. Expert Systems with Applications 37(2):1703–1712
Ziegler CN, Lausen G (2004) Analyzing correlation between trust and user similarity in online communities. In: Proceedings of 2nd international conference on trust management, S 251–265
Author information
Authors and Affiliations
Corresponding author
Additional information
Angenommen nach zwei Überarbeitungen durch Prof. Dr. Bichler.
This article is also available in English via http://www.springerlink.com and http://www.bise-journal.org: Elsner H, Krämer J (2013) Managing Corporate Portal Usage with Recommender Systems. Bus Inf Syst Eng. doi: 10.1007/s12599-013-0275-3.
Zusätzliche Information
Ergänzendes Onlinematerial kann unter folgendem Link abgerufen werden.
Rights and permissions
About this article
Cite this article
Elsner, H., Krämer, J. Nutzungsmanagement von Unternehmensportalen mithilfe von Empfehlungssystemen. Wirtschaftsinf 55, 205–219 (2013). https://doi.org/10.1007/s11576-013-0370-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11576-013-0370-6