Technologies to Support Online Knowledge Sharing and Retrieval for Feature-Rich Software Applications

点击翻译按钮获取中文摘要

Feature-rich software applications (e.g., Photoshop, Fusion 360, AutoCAD, Excel) enable a wide range of tasks through extended command and tool sets. Their widespread functionality and domain-specific vocabulary, however, can make it difficult for users to get started and to expand their skills. A relatively new reality of software learning is the loosely connected and distributed nature of help resources, with the Internet now ripe with materials (e.g., tutorials, discussion forums, Q&A sites) authored by other users. There are currently unparalleled opportunities for user communities to contribute resources that address the unique abilities, contexts and desired outcomes of an application's diverse user base. There is also the potential for other users to leverage these tailored resources to grow their skills. At the same time, current tools and platforms are not necessarily eliciting the perspectives of all user groups, and finding appropriate help resources can be a significant source of user frustration. My long-term objective is to inform and develop technologies that make contributing to and navigating this complex help ecosystem a seamless and rewarding activity for all feature-rich software users. For the research in this funding cycle, I focus on three short-term objectives: 1) Creating technologies that widen the pool of software users contributing their knowledge and experiences online, emphasizing two underrepresented groups (women and children). 2) Creating technologies for exploratory learners, who want to find materials that expand their software skills. 3) Investigating new evaluation approaches that increase and broaden participant pools, and capture new measures of software learning. Following established best practices in Human-Computer Interaction (HCI) research, my HQP and I will conduct mixed-method formative studies to create novel technologies that support users in contributing to and finding appropriate online external resources for feature-rich software. In developing our underlying algorithms, we will investigate machine learning, natural language processing, and collaborative filtering. We will validate our approaches and techniques through empirical evaluations with the intended target populations both in the lab and in the field. The proposed research will involve training approximately 14 HQP, providing them with skills in user-centred interaction design, Web technologies, applied artificial intelligence, and quantitative/qualitative analysis techniques. Our research will produce novel prototypes and design guidelines to inform the next generation of feature-rich software and knowledge-sharing communities. The proposed research will also address important dimensions of computing diversity by illustrating ways to consider women and child audiences upfront in the design of new technologies.