UNCORRECTED PROOF 1 2 Construction of interaction observation systems for collaboration analysis 3 in groupware applications 4 Rafael Duque a, * , Manuel Noguera b , Crescencio Bravo a , José Luis Garrido b , María Luisa Rodríguez b 5 a Department of Information Technologies and Systems, School of Computer Science Engineering, University of Castilla – La Mancha, 6 Paseo de la Universidad 4, 13071 Ciudad Real, Spain 7 b Department of Software Engineering, School of Computer Science and Telecommunication Engineering, University of Granada, 8 Calle Periodista Daniel Saucedo Aranda s/n, 18071 Granada, Spain 9 11 article info 12 Article history: 13 Received 22 September 2008 14 Received in revised form 19 November 2008 15 Accepted 19 January 2009 16 Available online xxxx 17 Keywords: 18 Groupware 19 Interaction and collaboration analysis 20 Ontologies 21 22 abstract 23 Interaction observation systems for groupware applications capture and process all the actions per- 24 formed by users engaged in workgroups. These actions are then stored in log documents that enable 25 the work process carried out by the users to be analyzed and the interaction between users to be studied. 26 This article proposes an approach, based on ontological models, which is devised to help the developer of 27 an observation system for a groupware application to structure and record user actions. In order to 28 achieve this aim, we present a specific ontology that shapes the collaborative work process of the users 29 so as to obtain an XML-based log document that stores all the actions carried out by the users and facil- 30 itates the subsequent analysis of the system usage and users’ behavior. This approach has been used to 31 improve communication and collaboration capabilities in the COLLECE groupware application. 32 Ó 2009 Elsevier Ltd. All rights reserved. 33 34 35 1. Introduction 36 The discipline of Computer-Supported Cooperative Work 37 (CSCW) examines how to develop groupware applications that 38 support the work of users collaborating to accomplish common 39 tasks. Since these groupware applications support the collabora- 40 tion of users who play different roles in the same social organiza- 41 tion, groupware applications must therefore offer the users a 42 shared workspace where they can interact. 43 Experience in the development and subsequent use of group- 44 ware systems, however, demonstrates that many of these systems 45 fail to provide useful tools for effective collaboration [18]. Conse- 46 quently (and in addition to technological issues), it becomes neces- 47 sary to consider the social dimension of the collaborative work by 48 analyzing how groups work and evolve [17]. 49 In this context, there emerges the challenge of designing inter- 50 action observation systems [36] that log all the actions performed 51 by the users of groupware applications [2], and consequently store 52 information characterizing the action accomplishment during a 53 collaborative activity. The information in the log documents typi- 54 cally contains variables, called descriptors, used to represent as- 55 pects related to user actions (e.g., the actor who carried out the 56 action, the artifact used in the action, etc.). These log documents re- 57 spond to the need for sources of information to facilitate analysis 58 and evaluation of the interaction between users and the collabora- 59 tive work processes carried out [25,35]. 60 This process, known as collaboration and interaction analysis, is 61 typically carried out in three phases [5]: (i) an initial observation 62 phase, in which all the information necessary for evaluating spe- 63 cific aspects of the collaborative work is collected; (ii) abstraction 64 phase, where the data stored in the observation phase are pro- 65 cessed and transformed to infer analysis indicators about charac- 66 teristics of the collaborative work processes; and (iii) a final 67 intervention phase, in which the inferred analysis indicators are 68 interpreted and different forms of actions are proposed to improve 69 the collaborative work (e.g., adapting the user interface, providing 70 advice automatically, etc.). 71 This article focuses on the first phase and proposes an ontolog- 72 ical approach for the development of interaction observation sys- 73 tems for groupware applications. This proposal is based firstly on 74 an ontological model that describes the collaborative work pro- 75 cesses to be carried out. Secondly, a set of descriptors are used 76 for recording in real time the actions performed by the users 77 through user interface components. Herein, actions are the most 78 basic unit of work supported by a groupware application. The 79 observation system subsequently captures the actions carried out 80 during the work process and records them in log documents. 81 This proposal presents three fundamental aims: (i) to use action 82 descriptors and analysis indicators in order to identify the work 83 carried out by each user within a workgroup; (ii) to register inter- 84 action traces in order to reconstruct the collaboration process be- 85 tween users and the interaction process with the groupware 0965-9978/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.advengsoft.2009.01.028 * Corresponding author. Tel.: +34 926 29 53 00; fax: +34 926 29 53 54. E-mail address: Rafael.Duque@uclm.es (R. Duque). Advances in Engineering Software xxx (2009) xxx–xxx Contents lists available at ScienceDirect Advances in Engineering Software journal homepage: www.elsevier.com/locate/advengsoft ADES 1404 No. of Pages 10, Model 5G 2 April 2009 Disk Used ARTICLE IN PRESS Please cite this article in press as: Duque R et al. Construction of interaction observation systems for collaboration analysis in groupware applications. Adv Eng Softw (2009), doi:10.1016/j.advengsoft.2009.01.028