CAREER: Measuring and Supporting Creativity in Developing Software Requirements

软件行业的新产品和新公司经常与现有软件竞争。这使得更好地满足人们需求并不同于现有产品的新产品变得重要。需求工程中的创造性强调开发既新颖又适合给定应用程序的创造性需求的关键作用。现有的促进需求创造力的方法，包括多天的研讨会、工具和框架，是耗时的，通常是无效的，并且在很大程度上依赖于需求分析人员自己的创造力。此外，创造力有许多方面应该独立评估，但RE领域几乎没有好的工具来评估创造力。该项目的目标是使用一系列基于人工智能的技术来帮助评估需求创造力，并帮助软件工程师开发更具创造力的需求。这将包括开发接口，帮助工程师考虑更广泛的可能需求，以及评估RE中的创造力的框架。研究工作还将支持课程开发、对学生研究人员的指导，以及为K-12学生、教师和软件初创公司举办以创造力为重点的培训研讨会。该项目的成功将促进创意软件开发的基础知识，并通过帮助开发人员成为创意软件专业人员并增加产品和公司成功的机会来影响地方、州和国家经济。该项目主要关注两个目标。第一个主要目标是使用数据挖掘、机器学习和自然语言处理技术，基于现有的在线软件相关内容(如评论、产品描述、问题跟踪系统和论坛讨论)生成潜在的创造性想法。然后，这些技术将被集成到支持系统中，以探索和选择具有创造性的需求，并进行引导的人工干预。第二个主要目标是开发一个全面的框架，用于评估软件需求的创造力，基于公认的创造力维度，包括相关性和有效性、新颖性、优雅和起源。这些工具和量表将通过包括专业软件开发人员和RE专家在内的用户研究进行验证。该项目提供了一种自动化机制来支持创造性需求的发现，并提供了一种新的框架来评估需求的创造性，解决了阻碍创造性在软件需求工程中广泛采用的基本问题。该项目由以人为中心的计算(CC)、已建立的激励竞争研究计划(EPSCoR)和软件与硬件基金会(SHF)共同资助。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

New products and new companies in the software industry often compete against existing software. This makes it important for new products that better meet people's needs and that differ from existing ones. Creativity in Requirements Engineering (RE) emphasizes the key role of developing creative requirements that are both novel and appropriate for a given application. Existing approaches to promote requirements creativity, including multi-day workshops, tools, and frameworks, are time-consuming, often ineffective, and largely dependent on the requirements analyst's own creativity. Moreover, creativity has a number of facets that should be assessed independently, but the field of RE has few good tools for evaluating creativity. This project's goal is to use a range of artificial intelligence-based techniques both to help assess requirement creativity and to help software engineers develop more creative requirements. This will involve both developing interfaces that help engineers consider a wider range of possible requirements and a framework for evaluating creativity in RE. The research work will also support curriculum development, mentoring of student researchers, and creativity-focused training workshops for K-12 students, teachers, and software startups. The success of this project will advance fundamental knowledge in creative software development and impact local, state, and national economies by helping developers become creative software professionals and increasing the chance that products and companies succeed.The project focuses on two main objectives. The first main objective is to use data mining, machine learning, and natural language processing techniques to generate potentially creative ideas based on existing online software-related content such as reviews, product descriptions, issue tracking systems, and forum discussions. These techniques will then be integrated into a support system to explore and select creative requirements with guided human intervention. The second main objective is to develop a comprehensive framework for evaluating the creativity of software requirements, based on well-established dimensions of creativity that include relevance and effectiveness, novelty, elegance, and genesis. These tools and scales will be validated through user studies including professional software developers and experts in RE. Providing an automated mechanism to support the discovery of creative requirements and a novel framework to evaluate the creativity of requirements, this project addresses the fundamental issues that act as barriers to the wide adoption of creativity in software requirements engineering.This project is jointly funded by Human Centered Computing (HCC), the Established Program to Stimulate Competitive Research (EPSCoR), and Software & Hardware Foundations (SHF).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.