会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 7. 发明授权
    • Developing and assuring policy documents through a process of refinement and classification
    • 通过细化和分类的过程开发和确保政策文件
    • US07552472B2
    • 2009-06-23
    • US10324502
    • 2002-12-19
    • Paul T. BaffesJohn Michael GarrisonMichael GilfixAllan HsuTyron Jerrod Stading
    • Paul T. BaffesJohn Michael GarrisonMichael GilfixAllan HsuTyron Jerrod Stading
    • G06F15/177
    • H04L63/1416H04L63/20
    • A system and method for developing network policy document and assuring up-to-date monitoring and automated refinement and classification of the network policy. The system administrator defines an initial policy document that is provided as the initial symbolic classifier. The classification rules remain in human readable form throughout the process. Network system data is fed through the classifier, which labels the data according to whether a policy constraint is violated. The labels are tagged to the data. The user then reviews the labels to determine whether the classification is satisfactory. If the classification of the data is satisfactory, the label is unaltered; However, if the classification is not satisfactory, the data is re-labeled. The re-labeled data is then introduced into a refinement algorithm, which determines what policy must be modified to correct classification of network events in accordance with the re-labeling. The network administrator then inspects the resulting new policy and modifies it if necessary. An updated classifier replaces the previous classifier.
    • 一种用于开发网络策略文档并确保最新监控和网络策略的自动细化和分类的系统和方法。 系统管理员定义作为初始符号分类器提供的初始策略文档。 分类规则在整个过程中保持以人类可读的形式。 网络系统数据通过分类器馈送,分类器根据是否违反策略约束来标记数据。 标签被标记为数据。 用户然后审查标签以确定分类是否令人满意。 如果数据分类令人满意,则标签不变; 然而,如果分类不令人满意,则重新标记数据。 然后将重新标记的数据引入到细化算法中,该算法确定根据重新标记来确定哪些策略必须被修改以校正网络事件的分类。 然后,网络管理员检查生成的新策略,并在必要时对其进行修改。 更新的分类器将替换以前的分类器。
    • 9. 发明授权
    • Method and apparatus for detecting password attacks using modeling techniques
    • 使用建模技术检测密码攻击的方法和装置
    • US07386892B2
    • 2008-06-10
    • US10621928
    • 2003-07-17
    • Michael GilfixFoluso Olaiya OkunseindeTyron Jerrod Stading
    • Michael GilfixFoluso Olaiya OkunseindeTyron Jerrod Stading
    • G06F7/04H04K1/00
    • G06F21/31
    • Provided is an apparatus and method for detecting fraudulent passwords so that computer break-in attempts can be distinguished from authorized users incorrectly entering their passwords. An actual password is mapped against a computer keyboard and the resultant data is stored in memory. The profile of an entered password is compared to the stored profile. If the profile of the entered password differs significantly from the stored profile, then the login attempt is flagged as an attempted intrusion. In one embodiment of the current invention, passwords are mapped according to the distance subsequent keystrokes are from each other. Different embodiments may have different mapping schemes. For example, mapping data may correspond to statistical data that corresponds to the likelihood that a particular character is typed by mistake when another character is intended.
    • 提供了一种用于检测欺诈口令的装置和方法,使得计算机入侵尝试可以与授权用户不正确地输入其密码区分开来。 实际的密码与计算机键盘映射,结果数据存储在内存中。 将输入的密码的配置文件与存储的配置文件进行比较。 如果输入的密码的配置文件与存储的配置文件显着不同,则登录尝试将被标记为尝试的入侵。 在本发明的一个实施例中,密码根据随后的击键彼此之间的距离被映射。 不同的实施例可以具有不同的映射方案。 例如,映射数据可以对应于对应于当预期另一个角色时特定字符被错误地键入的可能性的统计数据。
    • 10. 发明申请
    • Apparatus for Detecting Password Attacks Using Modeling Techniques
    • 使用建模技术检测密码攻击的装置
    • US20090328199A1
    • 2009-12-31
    • US12111494
    • 2008-04-29
    • Michael GilfixFoluso Olaiva OkunseindeTyron Jerrod Stading
    • Michael GilfixFoluso Olaiva OkunseindeTyron Jerrod Stading
    • H04L9/32G06F12/14
    • G06F21/31
    • Provided is an apparatus for detecting fraudulent passwords so that computer break-in attempts can be distinguished from authorized users incorrectly entering their passwords. An actual password is mapped against a computer keyboard and the resultant data is stored in memory. The profile of an entered password is compared to the stored profile. If the profile of the entered password differs significantly from the stored profile. then the login attempt is flagged as an attempted intrusion. In one embodiment of the current invention, passwords are mapped according to the distance subsequent keystrokes arc from each other. Different embodiments may have different mapping schemes. For example, mapping data may correspond to statistical data that corresponds to the likelihood that a particular character is typed by mistake when another character is intended.
    • 提供了一种用于检测欺诈密码的装置,使得可以将计算机入侵尝试与授权用户不正确地输入其密码区分开来。 实际的密码与计算机键盘映射,结果数据存储在内存中。 将输入的密码的配置文件与存储的配置文件进行比较。 如果输入的密码的配置文件与存储的配置文件显着不同。 那么登录尝试被标记为尝试入侵。 在本发明的一个实施例中,密码根据随后击键的距离相互映射。 不同的实施例可以具有不同的映射方案。 例如,映射数据可以对应于对应于当预期另一个角色时特定字符被错误地键入的可能性的统计数据。