III: Small: Collaborative Research: Mining and Optimizing Ad Hoc Workflows

自组织工作流在服务行业、科学研究以及日常生活中无处不在，如客户服务、故障排除、信息搜索等工作流，因此优化自组织工作流具有显著的社会效益。目前，自组织工作流的执行基于人的决策，其中误解、缺乏经验和无效处理并不少见，导致操作效率低下。本研究项目的目标是设计和开发挖掘和优化自组织工作流的基本模型、概念和算法。本课题在以下几个关键领域进行了创新性的研究：(1)网络建模与结构挖掘。建立了一个网络模型，该模型统计地捕捉了自组织工作流的执行特征，并针对不同的优化目标进行了优化，以提高新工作流的执行效率。(2)工作流伪像挖掘。建立在工作流执行的网络模型基础上，利用工作流人工挖掘技术对网络模型进行扩展，实现了既能利用执行又能利用文本内容的优化系统。(3)角色发现和关系评估。为了进一步优化工作流，建立了一个计算框架来分析代理在自组织工作流执行中的角色和关系，该项目的进步包括表示自组织工作流的模型、挖掘隐藏协作模型的算法以及优化自组织工作流处理的技术。该项目连接了两个新兴的研究领域：服务科学和网络科学，丰富了数据挖掘的原理和技术。通过来自不同领域(数据挖掘、数据库和网络)的团队成员的合作，也加强了研究基础设施。这项研究通过学生指导和课程开发与教育紧密结合。该项目产生的出版物、软件和课程材料将在项目网站上传播：网址：http://www.cs.ucsb.edu/~xyan/smartflow.htm

Ad hoc workflows are everywhere in service industry, scientificresearch, as well as daily life, such as workflows of customerservice, trouble shooting, information search, etc. Optimizing adhoc workflows thus has significant benefits to the society.Currently the execution of ad hoc workflows is based on humandecisions, where misinterpretation, inexperience, and ineffectiveprocessing are not uncommon, leading to operation inefficiency.The goal of this research project is to design and developfundamental models, concepts, and algorithms to mine and optimize adhoc workflows. The project includes novel research on the followingkey areas: (1) Network Modeling and Structure Mining. A network modelis built that statistically captures the execution characteristicsof ad hoc workflows, and is optimized to improve the execution ofnew workflows with respect to different optimization objectives.(2) Workflow Artifact Mining. The network model built on workflowexecutions is then extended with workflow artifact mining to realizean optimization system that is able to take advantage of bothexecutions and text contents. (3) Role Discovery and RelationAssessment. A computational framework is built to analyze the rolesand relationships of agents involved in ad hoc workflow executionsin order to further optimize workflows.Advances from this project include models to represent ad hocworkflows, algorithms for mining hidden collaborative models, andtechniques that optimize ad hoc workflow processing. The projectbridges two emerging research areas: service science and networkscience, and enriches the principles and technologies of data mining.It also enhances research infrastructure through the collaboration ofteam members from different areas (data mining, database, andnetwork). This research is tightly integrated with education throughstudent mentoring and curriculum development. Publications, software and course materials that arisefrom this project will be disseminated on the project website:URL: http://www.cs.ucsb.edu/~xyan/smartflow.htm