Karan Bhat, Vaibhavi Ghumare, Siddhesh Khadake, H. Gadade
{"title":"文本的词法简化的Web扩展","authors":"Karan Bhat, Vaibhavi Ghumare, Siddhesh Khadake, H. Gadade","doi":"10.1109/CONIT55038.2022.9847720","DOIUrl":null,"url":null,"abstract":"Lexical simplification means the process of providing alternatives to the complex words in the sentence with texts that are much more simpler to understand, while also preserving the context and grammar of the original text to make the whole sentence more easier to understand. All of the recent work involving lexical simplification relies on unsupervised tasks to learn simpler alternatives of complex words. But the drawback of most of these researches has been the fact that they provide simpler words without taking the context of the complex word in the sentence in account. In this paper, we are proposing a lexical simplifier which is based on contextual learnings from the sentence. We have applied the pre-trained representation model, BERT. It is a very powerful tool which can make use of the wider context of the sentence in both forward and backward direction. We have also taken the word frequency indicator from the Subtlex list, to produce results that will be more correct both semantically and grammatically. We have also added a web extension for the simplification of the text on the webpage, which takes the input from the user, processes the text on the server end, and gives the result in return after computation is over.","PeriodicalId":270445,"journal":{"name":"2022 2nd International Conference on Intelligent Technologies (CONIT)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Web Extension for Lexical Simplification of Text\",\"authors\":\"Karan Bhat, Vaibhavi Ghumare, Siddhesh Khadake, H. Gadade\",\"doi\":\"10.1109/CONIT55038.2022.9847720\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Lexical simplification means the process of providing alternatives to the complex words in the sentence with texts that are much more simpler to understand, while also preserving the context and grammar of the original text to make the whole sentence more easier to understand. All of the recent work involving lexical simplification relies on unsupervised tasks to learn simpler alternatives of complex words. But the drawback of most of these researches has been the fact that they provide simpler words without taking the context of the complex word in the sentence in account. In this paper, we are proposing a lexical simplifier which is based on contextual learnings from the sentence. We have applied the pre-trained representation model, BERT. It is a very powerful tool which can make use of the wider context of the sentence in both forward and backward direction. We have also taken the word frequency indicator from the Subtlex list, to produce results that will be more correct both semantically and grammatically. We have also added a web extension for the simplification of the text on the webpage, which takes the input from the user, processes the text on the server end, and gives the result in return after computation is over.\",\"PeriodicalId\":270445,\"journal\":{\"name\":\"2022 2nd International Conference on Intelligent Technologies (CONIT)\",\"volume\":\"3 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-06-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 2nd International Conference on Intelligent Technologies (CONIT)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CONIT55038.2022.9847720\",\"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 2nd International Conference on Intelligent Technologies (CONIT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CONIT55038.2022.9847720","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Lexical simplification means the process of providing alternatives to the complex words in the sentence with texts that are much more simpler to understand, while also preserving the context and grammar of the original text to make the whole sentence more easier to understand. All of the recent work involving lexical simplification relies on unsupervised tasks to learn simpler alternatives of complex words. But the drawback of most of these researches has been the fact that they provide simpler words without taking the context of the complex word in the sentence in account. In this paper, we are proposing a lexical simplifier which is based on contextual learnings from the sentence. We have applied the pre-trained representation model, BERT. It is a very powerful tool which can make use of the wider context of the sentence in both forward and backward direction. We have also taken the word frequency indicator from the Subtlex list, to produce results that will be more correct both semantically and grammatically. We have also added a web extension for the simplification of the text on the webpage, which takes the input from the user, processes the text on the server end, and gives the result in return after computation is over.