1 - 4 of 4
ItemCapturing the Forest or the Trees: Designing for Granularity in Data Crowdsourcing( 2020-01-07)Crowdsourcing is a method of completing a task by engaging a large group of heterogeneous contributors. Data crowdsourcing is crowdsourcing of data collection. In this paper, we demonstrate how data crowdsourcing projects can be differentiated along five dimensions: (1) the extent to which tasks are well-defined; (2) the duration of the task; (3) the type of value generated by the consumers of crowdsourcing data; (4) the variety of contribution allowed when completing the task; and (5) the relative value of each contribution. We argue that the quality of information created by a crowd depends on the granularity of contributions contributors are able to make. Finally, we propose a set of principles for designing crowdsourcing system to align the level of granularity of contributions with project objectives.
ItemFrom Voice to Knowledge: A Proposal for a Voice Annotation System to Support Collaborative Engineering Design Processes( 2020-01-07)This paper describes a novel voice interaction mechanism for capturing and managing design knowledge within a collaborative Computer-Aided Design (CAD) environment. We present a software module for speech recognition that integrates with a CAD application to allow the automatic creation of textual annotations in a 3D model directly from voice data. Audio is transcribed automatically, resulting in a textual note that is searchable and available to other users via a Product Data Management (PDM) system, providing an intuitive mechanism to document modeling processes and design knowledge. The system consists of three functional blocks: (1) audio recording, (2) speech recognition, and (3) query management against a cloud-based service. In this paper, we justify the need for our system from a human-computer interaction standpoint and discuss the rationale of its design and implementation in the context of collaborative design communication. Finally, we discuss some application spaces that demonstrate the capability of voice annotations for capturing knowledge.
ItemTowards Trust-Aware Human-Automation Interaction: An Overview of the Potential of Computational Trust Models( 2020-01-07)Several computational models have been proposed to quantify trust and its relationship to other system variables. However, these models are still under-utilised in human-machine interaction settings due to the gap between modellers’ intent to capture a phenomenon and the requirements for employing the models in a practical context. Our work amalgamates insights from the system modelling, trust, and human-autonomy teaming literature to address this gap. We explore the potential of computational trust models in the development of trust-aware systems by investigating three research questions: 1- At which stages of development can trust models be used by designers? 2- how can trust models contribute to trust-aware systems? 3- which factors should be incorporated within trust models to enhance models’ effectiveness and usability? We conclude with future research directions.