{"title":"基于知识图谱的化工安全智能问答系统研究","authors":"Zhenzhen Liu, Sheng Zheng, Xi Shi","doi":"10.1109/AICIT55386.2022.9930319","DOIUrl":null,"url":null,"abstract":"In recent years, the chemical industry has developed rapidly, and the production usage of chemical has continued to increase. In the face of variety of chemicals and ever-changing chemical reactions, people’s cognition of knowledge and damage characteristics lags behind, leading to frequent chemical accidents. For facilitating and assisting people to make reasonable judgments in daily work or when accidents happen, this paper utilizes the existing structured data of chemical safety constructs an intelligent question answering system based on knowledge graph, which is applied to chemical safety for the first time. Firstly, to deal with the problem of user intent understanding in question answering system, a multi-classifier combining NB (Naive Bayes) algorithm and Bert-BiLSTM-CRF model is constructed to complete the task of question classification. Secondly, according to the classification result, the most similar question template is matched. Finally, the semantic information of the template and the intention words in the question are mapped to the chemical safety knowledge graph to retrieve the answers. In addition, this paper utilizes SVM (Support Vector Machines) to compare the effect of question classification, and utilizes BiLSTM-CRF model to compare the effect of feature words recognition. The experimental results demonstrate that 91% of the questions are answered correctly, indicating that question answering system can answer the questions related to chemical safety in real time and accurately.","PeriodicalId":231070,"journal":{"name":"2022 International Conference on Artificial Intelligence and Computer Information Technology (AICIT)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Research on Intelligent Question Answering System for Chemical Safety Based on Knowledge Graph\",\"authors\":\"Zhenzhen Liu, Sheng Zheng, Xi Shi\",\"doi\":\"10.1109/AICIT55386.2022.9930319\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In recent years, the chemical industry has developed rapidly, and the production usage of chemical has continued to increase. In the face of variety of chemicals and ever-changing chemical reactions, people’s cognition of knowledge and damage characteristics lags behind, leading to frequent chemical accidents. For facilitating and assisting people to make reasonable judgments in daily work or when accidents happen, this paper utilizes the existing structured data of chemical safety constructs an intelligent question answering system based on knowledge graph, which is applied to chemical safety for the first time. Firstly, to deal with the problem of user intent understanding in question answering system, a multi-classifier combining NB (Naive Bayes) algorithm and Bert-BiLSTM-CRF model is constructed to complete the task of question classification. Secondly, according to the classification result, the most similar question template is matched. Finally, the semantic information of the template and the intention words in the question are mapped to the chemical safety knowledge graph to retrieve the answers. In addition, this paper utilizes SVM (Support Vector Machines) to compare the effect of question classification, and utilizes BiLSTM-CRF model to compare the effect of feature words recognition. The experimental results demonstrate that 91% of the questions are answered correctly, indicating that question answering system can answer the questions related to chemical safety in real time and accurately.\",\"PeriodicalId\":231070,\"journal\":{\"name\":\"2022 International Conference on Artificial Intelligence and Computer Information Technology (AICIT)\",\"volume\":\"31 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-09-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 International Conference on Artificial Intelligence and Computer Information Technology (AICIT)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/AICIT55386.2022.9930319\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 International Conference on Artificial Intelligence and Computer Information Technology (AICIT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AICIT55386.2022.9930319","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Research on Intelligent Question Answering System for Chemical Safety Based on Knowledge Graph
In recent years, the chemical industry has developed rapidly, and the production usage of chemical has continued to increase. In the face of variety of chemicals and ever-changing chemical reactions, people’s cognition of knowledge and damage characteristics lags behind, leading to frequent chemical accidents. For facilitating and assisting people to make reasonable judgments in daily work or when accidents happen, this paper utilizes the existing structured data of chemical safety constructs an intelligent question answering system based on knowledge graph, which is applied to chemical safety for the first time. Firstly, to deal with the problem of user intent understanding in question answering system, a multi-classifier combining NB (Naive Bayes) algorithm and Bert-BiLSTM-CRF model is constructed to complete the task of question classification. Secondly, according to the classification result, the most similar question template is matched. Finally, the semantic information of the template and the intention words in the question are mapped to the chemical safety knowledge graph to retrieve the answers. In addition, this paper utilizes SVM (Support Vector Machines) to compare the effect of question classification, and utilizes BiLSTM-CRF model to compare the effect of feature words recognition. The experimental results demonstrate that 91% of the questions are answered correctly, indicating that question answering system can answer the questions related to chemical safety in real time and accurately.