Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5444713 | Energy Procedia | 2017 | 19 Pages |
Abstract
Quality Assurance Management is imperative in nuclear engineering. And quality assurance language is a rigorous and highly logical language in workplaces. We build machine-learning models of natural language process about quality assurance management activities to extract valuable information from texts for management intelligently. As technological means, the primary purpose of NLP tools here is that converting massive unstructured data (text) to structured data (data attribute relationship).The tasks include event classification, named entity recognition (NER) of engineering, event domain judgment, automatic summarization and event similarity computing. We focus on Labeled-LDA and SVMs algorithms to perform short text classification. And using them as primary content, we can perform more advanced nuclear quality assurance management in future.
Keywords
NLPCRFClassification algorithmsNuclear SafetyExperience feedbackPattern recognitionERPEnterprise Resource PlanningQuality assuranceChinese word segmentationunstructured dataExpert systemEvent classificationText classificationSVMTopic modelsKnowledge managementNuclear engineeringartificial intelligenceMachine learning
Related Topics
Physical Sciences and Engineering
Energy
Energy (General)
Authors
Yongqing Guan, Yan Sun, Zeyu Wang, Qiyuan Zheng,