摘要
来自弗吉尼亚州亚历山大市的Robotics&Machine Learning Daily News的新闻记者兼新闻编辑报道说，发明者Mihai，Adrian（都柏林，IE），O'Gorm An，Eoin（伦敦，GB）于2022年5月6日提交的专利于6月11日，20日24日在网上公布。专利号12008304的受让人是iCIMS公司。（美国新杰西霍尔姆德尔）。新闻编辑从发明人提供的背景信息中获得了以下引文：“简历文档段落的分类在许多应用中都是有用的。例如，简历文档段落的分类对于公司的人力资源团队、网络社交平台和/或类似工具都是有用的。然而，对简历文档段落的分类已知方法，目前用于分析非结构化文本信息的方法依赖于基于规则的机制，这些机制对技术领域、语言和拼写错误敏感，这种方法和设备不能很好地扩展，当使用上下文被改变或转移(例如，从简历解析到工作描述解析)时，这些方法和设备不能达到它们的极限，这些上下文被用于不同的语言，用于新的工业领域，出现拼写错误或歧义消除等。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – From Alexandria, Virginia, NewsRx jour nalists report that a patent by the inventors Mihai, Adrian (Dublin, IE), O’Gorm an, Eoin (London, GB), filed on May 6, 2022, was published online on June 11, 20 24. The patent’s assignee for patent number 12008304 is iCIMS Inc. (Holmdel, New Jer sey, United States). News editors obtained the following quote from the background information suppli ed by the inventors: “Classification of paragraphs of resume documents can be us eful in many applications. For example, classification of paragraphs of resume d ocuments can be helpful for a company’s human resources team, a networking socia l platform, and/or the like. Known methods of classification of paragraphs of re sume documents, however, are labor-intensive and can be costly. Current approach es used in parsing of information in unstructured text rely on rules-based mecha nisms that are sensitive to technical domains, language and spelling errors, and rely on long lists of binary comparisons that are run against pre-defined sets of ad-hoc rules on top of customized taxonomies and vocabularies. Such methods a nd devices do not scale well and reach their limits when a use context is change d or shifted (e.g., from resume parsing to job-description parsing), used in a d ifferent language, used on a new industrial field, spelling errors or disambigua tion occurs, and/or the like.
NETL
NSTL