首页> 外文会议>International Conference on Software Engineering Research and Practice >Technical debt elicitation in text using natural language processing techniques
【24h】

Technical debt elicitation in text using natural language processing techniques

机译:使用自然语言处理技术的文本中的技术债务诱导

获取原文

摘要

In this article we explore a methodology for knowledge elicitation from textual information in software development communities. Our goal is to evaluate and predict whether technical depth is addressed in texts related to software development. Technical debt accumulation is a result of sub-optimal decisions taken during the life cycle of a software project and it is difficult to identify in textual information due to its pervasive nature. Our methodology used natural language processing techniques to possibilistically evaluate the relevance of a text to several aspects of software project management such as technical debt, architecture, code style, testing, design, organization governing, or documentation. We use data fusion to expand the inferred knowledge and the Choquet integral to evaluate the relevance of knowledge of each component of information. Further we identify a model of predicting technical debt using a linear combination of the other aspects. To test our approach we use sample text from IBM Developer Works website.
机译:在本文中,我们探索从软件开发社区中的文本信息中获取知识诱因的方法。我们的目标是评估和预测技术深度是否在与软件开发相关的文本中解决。技术债务积累是软件项目生命周期中采取的次优的决策的结果,并且由于其普遍的性质,难以识别文本信息。我们的方法使用自然语言处理技术来衡量文本与软件项目管理的几个方面的相关性,例如技术债务,架构,代码样式,测试,设计,组织管理或文档。我们使用数据融合来扩展推断知识和Choquet积分,以评估信息的每个组件的知识的相关性。此外,我们使用其他方面的线性组合来确定预测技术债务的模型。要测试我们的方法,我们使用IBM Developer Works网站的示例文本。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号