Requirement Management plays an important role in software development. But a large number of stakeholders make the traditional steps of requirement elicitation, filtration, and documentation extremely costly and time-consuming. The trend shifts towards collecting requirements from the Crowd. Crowdsourcing cuts the cost of outsourcing and decreases the workload. However, getting high-quality requirements from the crowd can be challenging. There is still a lot of work to do to transform crowdsourced requirements to high-quality ones. What we want to do is to accelerate automation of this process by classifying, rewriting, and prioritizing etc.
Natural language processing is one of the hottest and fundamental challenge in artificial intelligence. Some of the most commonly researched tasks in NLP are Part-of-speech tagging, Parsing, Word segmentation, Machine translation, Named entity recognition (NER), Natural language generation, Natural language understanding, Automatic summarization, Co-reference resolution and so on. But my interest is not in solving these basic NLP tasks. My recent projects have focused primarily on developing unsupervised and semi-supervised machine learning techniques, as well as utilizing exiting NLP techniques for solving domain specific applications, such as Statute/Advertisement Recommendation, News Sentiment Classification, and Constructing Knowledge Graph of Automobile Industry, etc.
Business Process Management (BPM) is a study of supporting business processes using methods, techniques, and software to design, enact, control, and analyze operational processes involving humans, organizations, applications, documents and other sources of information. Process mining is the technique for enhancing the BPM by extracting knowledge from the feedbacks. Concretely, we focused on process discovery, conformance checking and enhancement in BPM. For software process management, we are aiming at serving software reuse better by developing a hybrid software process management platform, including managing crowdsourcing requirements, traceability from requirements to design artifacts and code.
项目名称: 面向软件复用的众包需求管理与分析技术研究, 南京大学-计算机软件新技术国家重点实验室, 主持.
2020.06 - 2022.05. (Management and analysis of crowdsourcing requirements for software reuse)
项目名称: 面向中英文自然语言文本的夸张计算, 南京大学-计算机软件新技术国家重点实验室, 主持.
2021.03 - 2021.11. (Exaggeration Computing for Chinese and English)
项目名称: 基于行为网络的item表征学习的算法研究, 腾讯（数据平台部/数据应用中心/营销平台组）, 主持.
2020.01 - 2020.12. (Research on algorithm of item representation learning based on behavioral network)
项目名称: 华夏幸福-知识推理算法研发合作, 华夏幸福基业股份有限公司, 主持.
2019.10 - 2020.12. (Research & Development Cooperation on Algorithm of Knowledge Mining and Reasoning)
项目名称: 众包需求工程中面向软件复用的需求管理与分析技术研究, 国家自然科学青年基金, 主持.
2019.01 - 2021.12. (Research on Requirements Management and Analysis Technology for Software Reuse in Crowdsourcing Requirements Engineering)
项目名称: 众包需求工程中面向软件复用的需求管理与分析技术研究, 中央高校科研经费, 主持.
2018.03 - 2018.10. (Research on Requirements Management and Analysis Technology for Software Reuse in Crowdsourcing Requirements Engineering)
项目名称: 基于事件日志的业务过程并行挖掘研究, 南京大学-计算机软件新技术国家重点实验室, 主持.
2017.09 - 2019.08. (Parallel Mining Algorithm based on Business Process Event Logs)