发明名称 Systems and methods for categorizing and moderating user-generated content in an online environment
摘要 Exemplary embodiments provide systems, devices and methods for computer-based categorization and moderation of user-generated content for publication of the content in an online environment. Exemplary embodiments automatically determine a probability value indicating that the user-generated content is either a positive example or a negative example of one or more unsuitable categories. If the user-generated content is determined to be a positive example of any of the unsuitable categories to a predefined degree of certainty, exemplary embodiments may automatically exclude the content from publication in the online environment.
申请公布号 US8954316(B2) 申请公布日期 2015.02.10
申请号 US201113112556 申请日期 2011.05.20
申请人 AOL, Inc. 发明人 Revesz Jeffrey;Wiggins Christopher
分类号 G06F17/30 主分类号 G06F17/30
代理机构 Finnegan, Henderson, Farabow, Garrett & Dunner LLP 代理人 Finnegan, Henderson, Farabow, Garrett & Dunner LLP
主权项 1. A computer-executable method for moderating publication of textual content in an online environment, the method comprising: receiving, at a computer system, text to be published in an online environment: calculating at the computer system, a first numeric likelihood that the text falls into a first selected category unsuitable for publication; determining, at the computer system, first and second threshold values associated with the first selected category based on numeric likelihoods assigned to content within a reference corpus, the second threshold value exceeding the first threshold value, and at least a portion of the reference corpus content corresponding to the first selected category; comparing the first numeric likelihood to a first set of threshold values associated with the first selected category and stored in a database of the computer system; and determining whether to electronically publish the text in the online environment or exclude the text from publication based on the comparison of the first numeric likelihood to the first and second threshold values.
地址 Dulles VA US