Chithralekha Balamurugan

CrowdUtility: know the crowd that works for you

Crowdsourcing platforms aim to leverage the collective intelligence of a largely distributed Internet workforce to solve a wide range of tasks. Crowd workers (unlike in a typical organization), exhibit varying work patterns, expertise, and performance - with little or no control that can be imposed on them. Requesters (e.g. enterprises) also exhibit diverse requirements in terms of the size, complexity and timings of the tasks, as well as SLAs (performance expectations). Clearly, the heterogeneity makes the choice of a platform suited for a given task difficult for the user. This paper highlights this problem and proposes CrowdUtility - a first-of-a-kind statistical machine learning approach, which models the dynamic behavioral characteristics of crowdsourcing platforms and uses them to recommend the best platform for the enterprise task(s). Initial results from real-world experiments suggest that the proposed system provides an attractive solution to this erstwhile unsolved problem

Sustainable employment in India by crowdsourcing enterprise tasks

This paper explains how crowdsourcing would constitute for a promising and a successful alternate business model in India, especially at the juncture where the advantages of outsourcing is beginning to fade. The success of this alternate business model depends on the solutions for two challenges -- by taking work to people to leverage large educated population in India and making enough work available for the workforce to work and earn. Though solution to the first challenge is implicitly available through increasing penetration of Information and Communication Technologies (ICT) in India, the second challenge requires enough tasks to be available by enabling business organizations to adopt crowdsourcing. Since enterprise tasks are not readily crowdsourcable owing to security, compliance and contractual reasons, this paper proceeds to describe an end to end system which encompasses technical solutions that could help crowdsourcing business tasks, by tactfully overcoming the existing business constraints. For business tasks, we consider Insurance Claim Form Digitization which is one of the most common tasks taken up by outsourcing enterprises.

Service Assurance Framework for Enterprise Task Crowdsourcing

Crowd sourcing platforms enable enterprise requesters to leverage the online workforce to process voluminous enterprise tasks on a regular basis. Web services provided by these platforms facilitate requesters to post tasks, retrieve results and incentivize crowd workers. However, service assurance associated with task execution by crowd workers is not provided by these platforms. Owing to the flexible, uncommitted, discretionary working patterns of online crowd workers, service assurance for task execution is considered to be beyond the service assurance offerings of existing crowd sourcing platforms. Enterprises however, require these guarantees to be able to adopt crowd sourcing in a profound manner. In this paper, we propose a Service Assurance Framework as a crowd sourcing platform augmenting service, that provides service assurance for task execution by crowd workers. The framework helps enterprise requesters to identify and engage with workers who possess suitable service assurance attributes. To the best of our knowledge, this work is a first of its kind, in providing service assurance associated with task execution by crowd workers, with respect to enterprise tasks. We implemented the proposed framework and conducted a four-week long, large scale crowd sourcing experiment involving digitization of forms posted by an enterprise requester. Our results validate the efficacy of the proposed service assurance framework for enterprise crowd sourcing and advocate its adoption.

Human computer interaction paradigm for business process task crowdsourcing

In conventional systems like the web-based systems, device-based systems, or desktop systems, the Human Computer Interaction (HCI) spectrum encompasses of user interaction methods, interaction modes, user interface design, etc. that strive to facilitate a user to seamlessly and intuitively interact with the system to accomplish his system goals. This spectrum usually considers aspects pertaining to bringing the user to initiate interaction with the system and retaining the users interest to keep interacting with the system as an outcome of the usability/user experience that a user obtains while interacting with the system. In the context of crowdsourcing, these aspects have individual significance and not completely user experience based. Hence, it is essential to widen the spectrum of HCI for crowdsourcing to embrace these additional aspects explicitly and provide for a comprehensive HCI paradigm for crowdsourcing In this paper, the conceptualization and definition of HCI paradigm for crowdsourcing has been described. The application of every aspect of the proposed paradigm is illustrated with respect to crowdsourcing of a typical form digitization task. The defined aspects could also serve as HCI evaluation parameters for business process task crowdsourcing