公开(公告)号：US08839418B2

公开(公告)日：2014-09-16

申请号：US11335902

申请日：2006-01-18

申请人： Geoffrey John Hulten ,  Paul Stephen Rehfuss ,  Robert Rounthwaite ,  Joshua Theodore Goodman ,  Gopalakrishnan Seshadrinathan ,  Anthony P. Penta ,  Manav Mishra ,  Roderic C. Deyo ,  Elliott Jeb Haber ,  David Aaron Ward Snelling

发明人： Geoffrey John Hulten ,  Paul Stephen Rehfuss ,  Robert Rounthwaite ,  Joshua Theodore Goodman ,  Gopalakrishnan Seshadrinathan ,  Anthony P. Penta ,  Manav Mishra ,  Roderic C. Deyo ,  Elliott Jeb Haber ,  David Aaron Ward Snelling

IPC分类号： H04L9/00 ,  H04L29/06

CPC分类号： H04L63/1483 ,  G06F17/30887 ,  H04L63/08 ,  H04L63/1441

摘要： Described is a technology by which phishing-related data sources are processed into aggregated data and a given site evaluated the aggregated data using a predictive model to automatically determine whether the given site is likely to be a phishing site. The predictive model may be built using machine learning based on training data, e.g., including known phishing sites and/or known non-phishing sites. To determine whether an object corresponding to a site is likely a phishing-related object are described, various criteria are evaluated, including one or more features of the object when evaluated. The determination is output in some way, e.g., made available to a reputation service, used to block access to a site or warn a user before allowing access, and/or used to assist a hand grader in being more efficient in evaluating sites.

摘要翻译： 描述了一种将钓鱼相关数据源处理为聚合数据的技术，给定的站点使用预测模型评估聚合数据，以自动确定给定站点是否可能是钓鱼站点。 可以使用基于训练数据的机器学习来构建预测模型，例如包括已知的网络钓鱼站点和/或已知的非网络钓鱼站点。 为了确定对应于站点的对象是否可能是与钓鱼相关的对象，评估了各种标准，包括评估时对象的一个​​或多个特征。 该确定以某种方式输出，例如可用于信誉服务，用于阻止对站点的访问或在允许访问之前警告用户，和/或用于帮助平手机更有效地评估站点。

公开(公告)号：US08046832B2

公开(公告)日：2011-10-25

申请号：US10180565

申请日：2002-06-26

申请人： Joshua Theodore Goodman ,  Robert L. Rounthwaite

发明人： Joshua Theodore Goodman ,  Robert L. Rounthwaite

IPC分类号： G08B23/00

CPC分类号： G06Q10/107 ,  H04L51/12

摘要： A system and method facilitating detection of unsolicited e-mail message(s) with challenges is provided. The invention includes an e-mail component and a challenge component. The system can receive e-mail message(s) and associated probabilities that the e-mail message(s) are spam. Based, at least in part, upon the associated probability, the system can send a challenge to a sender of an e-mail message. The challenge can be an embedded code, computational challenge, human challenge and/or micropayment request. Based, at least in part, upon a response to the challenge (or lack of response), the challenge component can modify the associated probability and/or delete the e-mail message.

摘要翻译： 提供了一种有助于检测未经请求的电子邮件消息的系统和方法。 本发明包括电子邮件组件和挑战组件。 该系统可以接收电子邮件消息和电子邮件消息是垃圾邮件的关联概率。 至少部分地基于相关联的概率，系统可以向电子邮件消息的发送者发送挑战。 挑战可能是嵌入式代码，计算挑战，人为挑战和/或微支付请求。 至少部分地基于对挑战的响应（或缺乏响应），挑战组件可以修改相关联的概率和/或删除电子邮件消息。

公开(公告)号：US20110191848A1

公开(公告)日：2011-08-04

申请号：US12699073

申请日：2010-02-03

申请人： Benjamin Goth Zorn ,  Benjamin Livshits ,  Reid Borsuk ,  John Joseph Lambert ,  Matthew Ryan Miller ,  Louis Lafreniere ,  Peter Stuart Beck ,  Joshua Theodore Goodman ,  Timothy William Burrell ,  Steven Edward Lucco

发明人： Benjamin Goth Zorn ,  Benjamin Livshits ,  Reid Borsuk ,  John Joseph Lambert ,  Matthew Ryan Miller ,  Louis Lafreniere ,  Peter Stuart Beck ,  Joshua Theodore Goodman ,  Timothy William Burrell ,  Steven Edward Lucco

IPC分类号： G06F11/00 ,  G06F9/45

CPC分类号： G06F11/00

摘要： A method disclosed herein includes acts of receiving code at a Just-in-Time compiler executing in an application on a computing device and compiling the code to generate machine code and causing the machine code to be placed on at least one page that is accessible by at least one processor on the computing device, wherein the Just-in-Time compiler compiles the code utilizing at least one technique for preventing a Just-in-Time spraying attack.

摘要翻译： 本文公开的方法包括在计算设备上的应用程序中执行的即时编译器接收代码的动作，并编译代码以生成机器代码，并使机器代码放置在至少一个可由 所述计算设备上的至少一个处理器，其中所述即时编译器使用至少一种用于防止即时喷涂攻击的技术来编译代码。

公开(公告)号：US07266492B2

公开(公告)日：2007-09-04

申请号：US11465102

申请日：2006-08-16

申请人： Joshua Theodore Goodman

发明人： Joshua Theodore Goodman

IPC分类号： G06F17/28

CPC分类号： G06N99/005 ,  G10L15/063

摘要： A system and method facilitating training machine learning systems utilizing sequential conditional generalized iterative scaling is provided. The invention includes an expected value update component that modifies an expected value based, at least in part, upon a feature function of an input vector and an output value, a sum of lambda variable and a normalization variable. The invention further includes an error calculator that calculates an error based, at least in part, upon the expected value and an observed value. The invention also includes a parameter update component that modifies a trainable parameter based, at least in part, upon the error. A variable update component that updates at least one of the sum of lambda variable and the normalization variable based, at least in part, upon the error is also provided.

摘要翻译： 提供了一种利用连续条件广义迭代缩放来促进训练机器学习系统的系统和方法。 本发明包括至少部分地基于输入向量的特征函数和输出值，λ变量和归一化变量的和来修改期望值的期望值更新组件。 本发明还包括误差计算器，该误差计算器至少部分地基于期望值和观测值来计算误差。 本发明还包括至少部分地基于错误来修改可训练参数的参数更新组件。 还提供了至少部分地基于错误来更新λ变量和归一化变量之和中的至少一个的可变更新组件。

公开(公告)号：US07107207B2

公开(公告)日：2006-09-12

申请号：US10175430

申请日：2002-06-19

申请人： Joshua Theodore Goodman

发明人： Joshua Theodore Goodman

IPC分类号： G06F17/28

CPC分类号： G06N99/005 ,  G10L15/063

摘要： A system and method facilitating training machine learning systems utilizing sequential conditional generalized iterative scaling is provided. The invention includes an expected value update component that modifies an expected value based, at least in part, upon a feature function of an input vector and an output value, a sum of lambda variable and a normalization variable. The invention further includes an error calculator that calculates an error based, at least in part, upon the expected value and an observed value. The invention also includes a parameter update component that modifies a trainable parameter based, at least in part, upon the error. A variable update component that updates at least one of the sum of lambda variable and the normalization variable based, at least in part, upon the error is also provided.

摘要翻译： 提供了一种利用连续条件广义迭代缩放来促进训练机器学习系统的系统和方法。 本发明包括至少部分地基于输入向量的特征函数和输出值，λ变量和归一化变量的和来修改期望值的期望值更新组件。 本发明还包括误差计算器，该误差计算器至少部分地基于期望值和观测值来计算误差。 本发明还包括至少部分地基于错误来修改可训练参数的参数更新组件。 还提供了至少部分地基于错误来更新λ变量和归一化变量之和中的至少一个的可变更新组件。