公开(公告)号：US20130246383A1

公开(公告)日：2013-09-19

申请号：US13423243

申请日：2012-03-18

申请人： Ryen William White ,  Georg LW Buscher ,  Susan T. Dumais ,  Jeff Huang ,  Kuansan Wang ,  Abdigani M. Diriye

发明人： Ryen William White ,  Georg LW Buscher ,  Susan T. Dumais ,  Jeff Huang ,  Kuansan Wang ,  Abdigani M. Diriye

IPC分类号： G06F17/30

CPC分类号： G06F16/9535

摘要： The subject disclosure is directed towards using cursor activity with respect to search result pages to enhance search engine operation. Data associated with the cursor activity may be translated into cursor events representing user interactions with a search result page. Based on the cursor events, user behavior indicia may be identified via a user intent prediction model corresponding to various search result page related user actions. The user behavior indicia and/or the user intent prediction model may be used to produce search result pages for current search queries from the user.

摘要翻译： 本发明涉及使用关于搜索结果页面的光标活动来增强搜索引擎的操作。 与游标活动相关联的数据可以被转换成表示用户与搜索结果页面的交互的游标事件。 基于光标事件，可以经由与各种搜索结果页面相关的用户动作对应的用户意图预测模型来识别用户行为标记。 用户行为标记和/或用户意图预测模型可以用于产生来自用户的当前搜索查询的搜索结果页面。

公开(公告)号：US08527534B2

公开(公告)日：2013-09-03

申请号：US12726358

申请日：2010-03-18

申请人： Kuansan Wang ,  Bo-June Hsu ,  Xiaolong Li

发明人： Kuansan Wang ,  Bo-June Hsu ,  Xiaolong Li

IPC分类号： G06F17/30

CPC分类号： G06F17/30654 ,  G06F17/30731 ,  G06F17/30976

摘要： Architecture that employs a modeling technique based on language modeling to estimate a probability of a document matching the user need as expressed in the query. The modeling technique is based on the data mining results that various portions of a document (e.g., body, title, URL, anchor text, user queries) use different styles of human languages. Thus, the results based on a language can be adapted individually to match the language of query. Since the approach is based on adaptation, the framework also provides a natural means to progressively revise the model as user data are collected. Different styles of languages in a document can be recognized and adapted individually. Background language models are also employed that offer a fallback approach in case the document has incomplete fields of data, and can utilize topical or semantic hierarchy of the knowledge domain.

摘要翻译： 采用基于语言建模的建模技术的架构来估计在查询中表达的与用户匹配的文档的概率。 建模技术基于数据挖掘结果，文档的各个部分（例如，主体，标题，URL，锚文本，用户查询）使用不同风格的人类语言。 因此，基于语言的结果可以单独调整以匹配查询语言。 由于该方法是基于适应性的，所以该框架还提供了一种在收集用户数据时逐步修改模型的自然方法。 文档中不同风格的语言可以单独识别和修改。 还使用背景语言模型，在文档具有不完整的数据领域的情况下提供回退方法，并且可以利用知识领域的局部或语义层次结构。

公开(公告)号：US20120323839A1

公开(公告)日：2012-12-20

申请号：US13162563

申请日：2011-06-16

申请人： Emre Kiciman ,  Abulimiti Aji ,  Kuansan Wang

发明人： Emre Kiciman ,  Abulimiti Aji ,  Kuansan Wang

IPC分类号： G06N5/02

CPC分类号： G06N99/005

摘要： A classifier that disambiguates among entities based on a dictionary, such as corpus of documents about those entities, is built by incorporating probabilities that an entity exists that is not in the dictionary. Given a document it is associated by the classifier with an entity. By incorporating out of collection probabilities into the classifier, a higher level of confidence in the match between an entity and a document is achieved.

摘要翻译： 基于字典的实体之间消歧的分类器，例如关于这些实体的文档语料库，是通过并入不在字典中的实体存在的概率来构建的。 给定一个文档，它由分类器与实体相关联。 通过将收集概率合并到分类器中，实现了实体和文档之间的匹配的更高的置信度。

公开(公告)号：US20120246133A1

公开(公告)日：2012-09-27

申请号：US13069526

申请日：2011-03-23

申请人： Bo-June Hsu ,  Kuansan Wang ,  Huizhong Duan

发明人： Bo-June Hsu ,  Kuansan Wang ,  Huizhong Duan

IPC分类号： G06F17/30

CPC分类号： G06F17/273 ,  G06F17/276

摘要： Online spelling correction/phrase completion is described herein. A computer-executable application receives a phrase prefix from a user, wherein the phrase prefix includes a first character sequence. A transformation probability is retrieved responsive to receipt of the phrase prefix, wherein the transformation probability indicates a probability that a second character sequence has been transformed into a first character sequence. A search is then executed over a trie to locate a most probable phrase completion based at least in part upon the transformation probability.

摘要翻译： 本文描述了在线拼写校正/短语完成。 计算机可执行应用程序从用户接收短语前缀，其中短语前缀包括第一字符序列。 响应于短语前缀的接收来检索变换概率，其中变换概率表示第二字符序列已被变换为第一字符序列的概率。 然后，通过一个搜索来执行搜索以至少部分地基于转换概率来定位最可能的短语完成。

公开(公告)号：US20110010323A1

公开(公告)日：2011-01-13

申请号：US12499092

申请日：2009-07-08

申请人： Kuansan Wang ,  Xiaodong Li ,  Nikolas Clemens Gloy

发明人： Kuansan Wang ,  Xiaodong Li ,  Nikolas Clemens Gloy

IPC分类号： G06N5/04 ,  G06F3/048 ,  G06F17/30

CPC分类号： G06F9/451

摘要： A system described herein includes a receiver component that receives action sequence data pertaining to a plurality of selectable graphical items displayed to a user on a computer display screen, wherein the action sequence data is indicative of an order in which the user performed an action with respect to a first subset of graphical items in the plurality of selectable graphical items. The system also includes a determiner component that infers a view sequence with respect to a second subset of graphical items in the plurality of graphical items, wherein the view sequence is an order that the user viewed graphical items in the second subset of graphical items, and wherein the second subset of graphical items comprises the first subset of graphical items and at least one other graphical item in the plurality of graphical items.

摘要翻译： 本文描述的系统包括接收器组件，其接收与在计算机显示屏幕上向用户显示的多个可选图形项目有关的动作序列数据，其中动作序列数据指示用户执行相关动作的顺序 涉及多个可选图形项目中的图形项目的第一子集。 所述系统还包括确定器组件，所述确定器组件相对于所述多个图形项目中的图形项目的第二子集推断视图序列，其中所述视图序列是所述用户在图形项目的第二子集中观看图形项目的顺序，以及 其中图形项目的第二子集包括多个图形项目中的图形项目的第一子集和至少一个其他图形项目。

公开(公告)号：US07409349B2

公开(公告)日：2008-08-05

申请号：US09960229

申请日：2001-09-20

申请人： Kuansan Wang ,  Hsiao-Wuen Hon

发明人： Kuansan Wang ,  Hsiao-Wuen Hon

IPC分类号： G10L21/00

CPC分类号： G10L15/30 ,  G06F3/16 ,  G06F17/218 ,  G10L15/24 ,  H04M1/271 ,  H04M1/72561 ,  H04M3/493 ,  H04M3/4936 ,  H04M2207/40

摘要： A markup language for execution on a client device in a client/server system includes instructions to unify at least one of recognition-related events, GUI events and telephony events on non-display, voice input based client device and a multimodal based client for a web server interacting with each of the client devices. A recognition server for receiving data indicative of inputted data provided to a client device and an indication of a grammar to use for recognition is also provided.

摘要翻译： 用于在客户端/服务器系统中的客户端设备上执行的标记语言包括用于将非显示器，基于语音输入的客户端设备和基于多模式的客户机上的识别相关事件，GUI事件和电话事件中的至少一个统一的指令，用于 Web服务器与每个客户端设备进行交互。 还提供一种识别服务器，用于接收指示提供给客户端设备的输入数据的数据和用于识别的语法指示。

公开(公告)号：US20080177534A1

公开(公告)日：2008-07-24

申请号：US11656606

申请日：2007-01-23

申请人： Kuansan Wang ,  Nirav Girish Shah

发明人： Kuansan Wang ,  Nirav Girish Shah

IPC分类号： G10L19/14

CPC分类号： H04M7/125 ,  G10L15/26 ,  H04M3/002 ,  H04M3/2236 ,  H04M3/301

摘要： Described is automatically testing the quality of an audio channel between a caller and a callee that includes a device under test, such as a VoIP or other gateway. An analyzer receives timestamps from a caller and callee during a calling session, including timestamps for when the callee initially provides audio (e.g., speech) to the caller, when the caller initially detects sound, when the caller initially provides audio to the callee, and when the callee initially detects sound. The analyzer uses the relative timing of the timestamps and the speech recognizer's outcome to determine whether the audio channel is experiencing interference or echo. When the audio includes speech, a confidence level corresponding to accuracy of speech recognition also may establish the audio channel's quality. Random selection and timing of output may be employed, such as to vary the testing patterns during repetitive tests.

摘要翻译： 描述的是自动测试呼叫者和被叫者之间的音频频道的质量，包括被测设备，如VoIP或其他网关。 分析仪在呼叫会话期间从呼叫者和被呼叫者接收时间戳，包括呼叫者最初在呼叫者最初检测到声音时最初向呼叫者提供音频（例如，语音）的时间戳，当呼叫者最初向被叫方提供音频时，以及 当被调用者最初检测到声音时。 分析仪使用时间戳和语音识别器的结果的相对定时来确定音频信道是否正在经历干扰或回波。 当音频包括语音时，对应于语音识别的准确性的置信水平也可以建立音频通道的质量。 可以采用随机选择和输出定时，例如在重复测试期间改变测试模式。

公开(公告)号：US20060070086A1

公开(公告)日：2006-03-30

申请号：US10977106

申请日：2004-10-29

申请人： Kuansan Wang

发明人： Kuansan Wang

IPC分类号： G06F13/00

CPC分类号： G06F8/00 ,  G06Q10/10

摘要： A unified programming environment allows application developers to work with declarative, procedural and service model based logic. In one aspect, instructions on a computer readable medium include at least one of a declarative logic module adapted to define an object and a procedural logic module adapted to define an object. A service model based logic module is adapted to define actions to be performed on the object with a reference to said at least one of the declarative and procedural logic module.

摘要翻译： 统一的编程环境允许应用开发人员使用基于声明，过程和服务模型的逻辑。 在一个方面，计算机可读介质上的指令包括适于定义对象的声明性逻辑模块和适于定义对象的程序逻辑模块中的至少一个。 基于服务模型的逻辑模块适于通过参考所述至少一个声明性和过程性逻辑模块来定义要对对象执行的动作。

公开(公告)号：US09449078B2

公开(公告)日：2016-09-20

申请号：US12243937

申请日：2008-10-01

申请人： Kuansan Wang ,  Toby H. Walker ,  Zijian Zheng ,  Frederic H. Behr, Jr. ,  Yu Chen ,  Robert C. Wang

发明人： Kuansan Wang ,  Toby H. Walker ,  Zijian Zheng ,  Frederic H. Behr, Jr. ,  Yu Chen ,  Robert C. Wang

IPC分类号： G06F17/30

CPC分类号： G06F17/30648 ,  G06F17/30722 ,  G06F17/30864

摘要： The ranking quality of a ranked list may be evaluated. In an example embodiment, a method is implemented by a system to access log data, ascertain which entries of a ranked list are skipped, and determine a ranking quality metric from the skipped entries. More specifically, log data that reflects user interactions with a ranked list having multiple entries is accessed. The user interactions include at least indications of which of the multiple entries are selected entries. It is ascertained which entries of the multiple entries of the ranked list are skipped entries based on the selected entries. The ranking quality metric for the ranked list is determined responsive to the skipped entries.

摘要翻译： 可以评估排名列表的排名质量。 在一个示例性实施例中，系统通过系统实现访问日志数据，确定跳过排名列表的哪些条目并从跳过的条目确定排序质量度量的方法。 更具体地，访问反映与具有多个条目的排名列表的用户交互的日志数据。 用户交互包括至少指示多个条目中的哪一个是选择的条目。 基于所选择的条目，确定排序列表的多个条目的哪些条目被跳过条目。 响应于跳过的条目来确定排名列表的排名质量度量。

公开(公告)号：US08515950B2

公开(公告)日：2013-08-20

申请号：US12243678

申请日：2008-10-01

申请人： Jianfeng Gao ,  Kuansan Wang

发明人： Jianfeng Gao ,  Kuansan Wang

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/30867

摘要： Log-based rankers and document-based rankers may be combined for searching. In an example embodiment, there is a method for combining rankers to perform a search operation. A count of query instances in log data is ascertained based on a query. A search for the query is performed to produce a set of search results. The set of search results is ranked by relevance score with a document-based ranker and a log-based ranker using a weighting factor that is adapted responsive to the count of the query instances in the log data.

摘要翻译： 基于日志的排名者和基于文档的排名者可以组合进行搜索。 在示例性实施例中，存在用于组合等级以执行搜索操作的方法。 基于查询确定日志数据中的查询实例的计数。 执行查询的搜索以产生一组搜索结果。 搜索结果的集合通过使用基于文档的游客和基于日志的游戏者的相关性得分进行排序，该游戏者使用响应于日志数据中的查询实例的计数的加权因子。