Guided Crowdsourcing for Collective Work Coordination in Corporate Environments Ioanna Lykourentzou 1 , Dimitrios J. Vergados 2 , Katerina Papadaki 3 , and Yannick Naudet 1 1 Centre de Recherche Public Henri Tudor, Luxembourg {ioanna.lykourentzou,yannick.naudet}@tudor.lu 2 Norwegian University of Science and Technology, Department of Telematics, Norway dimitrios.vergados@item.ntnu.no 3 Bank of Greece, Operational Risk Management, Greece kpapadaki@bankofgreece.gr Abstract. Crowdsourcing is increasingly gaining attention as one of the most promising forms of large-scale dynamic collective work. However current crowd- sourcing approaches do not offer guarantees often demanded by consumers, for example regarding minimum quality, maximum cost or job accomplishment time. The problem appears to have a greater impact in corporate environments because in this case the above-mentioned performance guarantees directly affect its viabil- ity against competition. Guided crowdsourcing can be an alternative to overcome these issues. Guided crowdsourcing refers to the use of Artificial Intelligence methods to coordinate workers in crowdsourcing settings, in order to ensure col- lective performance goals such as quality, cost or time. In this paper, we inves- tigate its potential and examine it on an evaluation setting tailored for intra and inter-corporate environments. Keywords: crowdsourcing, crowd coordination, resource allocation. 1 Introduction Crowdsourcing is a new form of user involvement on the Web. It has recently emerged as a new paradigm of collective work and as a natural result of the Web’s evolution course, from a purely non-participatory system, with users in the place of content con- sumers, to a virtual space of full user involvement, since the Web 2.0 era and beyond. Crowdsourcing refers to the splitting of a large, human-intelligence job into smaller micro-tasks and dynamically “outsourcing” these, not to specific individuals, but to an unknown crowd of web workers. Examples of jobs often accomplished through crowd- sourcing include the translation of large corpuses of small sentences from one language to the other, the recognition of captchas, the transcription of audio files to text, but also the collective creation of articles in Wikipedia and the development of open source software artifacts by several distributed programmers [4]. The crowdsourcing technology increases rapidly. Having started only a few years ago, it is already being used at large-scale by commercial players, academics and in- dividuals, who benefit from its ability to involve millions of users worldwide and to C. B˘ adic˘ a, N.T. Nguyen, and M. Brezovan (eds.): ICCCI 2013, LNAI 8083, pp. 90–99, 2013. c  Springer-Verlag Berlin Heidelberg 2013