复杂产品协同设计流程的多视图自然语言文本生成  

Multi-view natural language text generation method for complex product design process

作　　者：周长红[1,2] 原桂远 曾庆田 段华 ZHOU Changhong 1,2 , YUAN Guiyuan 2, ZENG Qingtian 2, DUAN Hua 3(1. College of Economics and Management, Shandong University of Science and Technology,Qingdao 266590, China;2. College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, China;3. College of Mathematics and System Science, Shandong University of Science and Technology, Qingdao 266590, Chin)

机构地区：[1]山东科技大学经济管理学院,山东青岛266590 [2]山东科技大学计算机科学与工程学院,山东青岛266590 [3]山东科技大学数学与系统科学学院,山东青岛266590

出　　处：《计算机集成制造系统》2018年第7期1838-1849,共12页Computer Integrated Manufacturing Systems

基　　金：国家自然科学基金资助项目(71704096;61602278;61602279;61472229;31671588);山东省科技发展计划资助项目(2014GGX101035;2016ZDJS02A11);山东省自然科学基金资助项目(BS2014DX013;ZR2015FM013;ZR2017MF027);同济大学嵌入式系统与服务计算教育部重点实验室开放课题资助项目(ESSCKF201403);山东省博士后创新专项资金资助项目(201603056);山东科技大学领军人才与优秀科研团队计划资助项目(2015TDJH102);教育部人文社会科学研究项目(16YJCZH012)~~

摘　　要：为了业务人员更好地理解复杂产品设计过程的业务流程建模标注(BPMN)模型,以便与系统分析人员进行有效沟通,提出一种基于BPMN模型的复杂产品设计过程多视图自然语言文本生成方法。该方法能够根据业务人员的不同需求对流程模型进行结构投影,进而使用句法分析和语义角色标注解析BPMN流程模型的文本信息,并使用流程结构树(RPST)算法解析BPMN流程模型的结构,从而将流程模型划分成具有层次的子流程片断得到流程结构树。解析的文本信息与不同视图下的流程结构树合并,形成带注释的多视图流程结构树,然后使用语法树由带注释的流程结构树自动生成自然语言文本,最终生成多视图下的自然语言文本。所提方法已在系统中实现,并通过测试集进行实验验证了所提方法的有效性。To facilitate the business staff to better understand Business Process Model and Notation（BPMN） model of complex product design process and to communicate effectively with system analysts, a multi-view natural language text generation method of complex product design process was presented based on BPMN model, which could create the projection of process model according to the different requirements of business staff. The text information of BPMN process model could be annotated by using syntactic and semantic roles, and the structure of BPMN process model could be analyzed by using Refined Process Structure Tree （RPST） algorithm. The process model was divided into hierarchical sub-process fragments to get the process structure tree. The parsed text information merged with the process tree to form the annotated multi-view process structure tree, and then the syntax tree was used to automatically generate the natural language text from the annotated process structure tree. The multi-view natural language text was generated. The method had been implemented in the system, and the test set had been used to verify the effectiveness of the proposed method.

关 键 词：协同设计 业务流程建模标注 多视图 自然语言文本生成 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]