{"title":"未见事物的证据:在关于除草剂“橙剂”的非结构化文本中映射情感","authors":"S. Hopton","doi":"10.1145/2666216.2666233","DOIUrl":null,"url":null,"abstract":"Sentiment analysis is a new, relatively unexplored, but potentially helpful method technical communicators can measure shifts in feeling, values, attitudes and beliefs, which drive behavior. Sentiment analysis is a relatively new method so there is little research in technical communications that address best practices. This poster demonstrates the process by which I mined sentiment from a corpus of unstructured text using a dictionary, categorization model and co-location algorithms. Preliminary results are consistent with empirical and historical observations, showing an uptick in negative emotion during moments of greatest political and scientific controversy. The implications of this research suggest new programs and greater access to digitized texts offer the practitioner and scholar new sites and tools with which to conduct research wider in scale and scope, but that the disambiguation of texts continues to prevent total automation of such processes.","PeriodicalId":393730,"journal":{"name":"Proceedings of the 32nd ACM International Conference on The Design of Communication CD-ROM","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2014-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Evidence of Things Not Seen: Mapping Sentiment in Unstructured Texts about the Herbicide 'Agent Orange'\",\"authors\":\"S. Hopton\",\"doi\":\"10.1145/2666216.2666233\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Sentiment analysis is a new, relatively unexplored, but potentially helpful method technical communicators can measure shifts in feeling, values, attitudes and beliefs, which drive behavior. Sentiment analysis is a relatively new method so there is little research in technical communications that address best practices. This poster demonstrates the process by which I mined sentiment from a corpus of unstructured text using a dictionary, categorization model and co-location algorithms. Preliminary results are consistent with empirical and historical observations, showing an uptick in negative emotion during moments of greatest political and scientific controversy. The implications of this research suggest new programs and greater access to digitized texts offer the practitioner and scholar new sites and tools with which to conduct research wider in scale and scope, but that the disambiguation of texts continues to prevent total automation of such processes.\",\"PeriodicalId\":393730,\"journal\":{\"name\":\"Proceedings of the 32nd ACM International Conference on The Design of Communication CD-ROM\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2014-09-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 32nd ACM International Conference on The Design of Communication CD-ROM\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2666216.2666233\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 32nd ACM International Conference on The Design of Communication CD-ROM","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2666216.2666233","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Evidence of Things Not Seen: Mapping Sentiment in Unstructured Texts about the Herbicide 'Agent Orange'
Sentiment analysis is a new, relatively unexplored, but potentially helpful method technical communicators can measure shifts in feeling, values, attitudes and beliefs, which drive behavior. Sentiment analysis is a relatively new method so there is little research in technical communications that address best practices. This poster demonstrates the process by which I mined sentiment from a corpus of unstructured text using a dictionary, categorization model and co-location algorithms. Preliminary results are consistent with empirical and historical observations, showing an uptick in negative emotion during moments of greatest political and scientific controversy. The implications of this research suggest new programs and greater access to digitized texts offer the practitioner and scholar new sites and tools with which to conduct research wider in scale and scope, but that the disambiguation of texts continues to prevent total automation of such processes.