Analysis of evolution of software project risk research based on text analysis

oleh: Ying QU, Moran FAN, Bin LIU, Shugui CAO

Format: Article
Diterbitkan: Hebei University of Science and Technology 2020-02-01

Deskripsi

Software project is an engineering process that aims to meet the demand side of software scientifically, covering various elements such as personnel, technology, management, etc. For a long time, due to some uncertain factors of software projects, many problems such as design, cost, progress and function change are often encountered in the implementation of software projects. Although the development technology is constantly improving, the management problems always exist, and the risk problem is inevitable, making it being the focus of the software industry. Correctly understanding and managing the software project risk can improve the success rate of software project development and reduce the probability of risk occurrence. 河北科技大学学报 2020年 第1期 瞿〓英,等:基于文本分析的软件项目风险研究演化脉络解析 Software project risk management is one of the key issues in software project management. With the development of software industry, software project risk also presents new characteristics, and the content of risk management also has new changes. In order to track the context and development trend of software project risk management research, this paper adopts the following methods: using web crawler technology, 3 129 domestic and foreign software project risk related literature are obtained; using word segmentation and statistical analysis technology, the author, key words, main topics, etc. of the literature are extracted and word frequency analysis is carried out, and the spatial distribution of the research subjects is mined; from the perspective of the published paper number and time, this paper compares and analyzes the study of Chinese and foreign scholars, and combs out the evolution processes of software project risk concept definition, research stage, research methods, etc. Through the analysis of the literature topic, the paper summarizes the research direction and trend of software project risk. Through the above text analysis ideas, we can get the mainstream research methods and development trends in this field. It can be inferred from the change of literature number that there are many researches on software project risk homogenization, and it is difficult to explore the innovative breakthrough of the research, resulting in the decrease of the number of Chinese articles and stagnation. By observing the word cloud visualization of the research team, two main research directions are found: computer and management. How to conduct risk research from the perspective of management has always been the focus of this field. Based on the time sequence literature topic summary, it can be clearly seen that the software project risk research is a project management oriented and risk management-oriented research, and the project management and risk management are integrated to define the concept of software project as the main body. After the theoretical research of software project risk management is relatively mature, related technical researches such as risk assessment, risk identification and risk control begin to rise, then risk management validation research is conducted from the perspective of application. Up to now, risk management still takes assessment as the main research method. Through the co-occurrence analysis of high-frequency words, this paper can find relatively new research directions in lexical association. for example, the process of risk identification is transforming from artificial subjective identification to machine automatic identification, the transformation of attribute characteristics of risk management objects is from static to dynamic, and other cutting-edge researches such as the establishment of risk knowledge base caused by big data technology, the construction of intelligent risk management system, etc. The exploration of research trends will provide new solutions for software project risk prevention and control, improve the success rate of software projects, and provide reference methods for the related research of software project risk.