专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US08489590B2 Cross-market model adaptation with pairwise preference data 有权
标题翻译：跨市场模型适应与成对偏好数据
公开(公告)号：US08489590B2
公开(公告)日：2013-07-16
申请号：US12966983
申请日：2010-12-13
申请人： Yi Chang , Zhaohui Zheng , Fernando David Diaz , Jing Bai
发明人： Yi Chang , Zhaohui Zheng , Fernando David Diaz , Jing Bai
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06F17/30864
摘要： Embodiments are directed towards generating market-specific ranking models by leveraging target market specific pairwise preference data. The pairwise preference data includes market-specific training examples, while a ranking model from another market captures the common characteristics of the resulting ranking model. In one embodiment, the ranking model is trained by applying a Tree Based Ranking Function Adaptation (TRADA) algorithm to multi-grade labeled training data, such as editorially generated training data. Then, contradictions between the TRADA generated ranking model and target-market specific pairwise preference data are identified. For each identified contradiction, new training data is generated to correct the contradiction. Then, in one embodiment, an algorithm such as TRADA is applied to the existing ranking model and the new training data to generate a new ranking model.
摘要翻译：实施例旨在通过利用目标市场特定的成对偏好数据来产生市场特定的排名模型。成对偏好数据包括市场特定的培训示例，而来自另一个市场的排名模型捕获了所得到的排名模型的共同特征。在一个实施例中，通过将基于树的排序函数适应（TRADA）算法应用于诸如编辑生成的训练数据的多等级标记的训练数据来训练排名模型。然后，确定了TRADA产生的排名模型和目标市场特定成对偏好数据之间的矛盾。对于每个确定的矛盾，产生新的训练数据以纠正矛盾。然后，在一个实施例中，诸如TRADA的算法被应用于现有的排名模型和新的训练数据以生成新的排名模型。

2. 发明申请

US20120150855A1 CROSS-MARKET MODEL ADAPTATION WITH PAIRWISE PREFERENCE DATA 有权
标题翻译：交叉市场模型适配与配对偏好数据
公开(公告)号：US20120150855A1
公开(公告)日：2012-06-14
申请号：US12966983
申请日：2010-12-13
申请人： Yi Chang , Zhaohui Zheng , Fernando David Diaz , Jing Bai
发明人： Yi Chang , Zhaohui Zheng , Fernando David Diaz , Jing Bai
IPC分类号： G06F17/30
CPC分类号： G06F17/30864
摘要： Embodiments are directed towards generating market-specific ranking models by leveraging target market specific pairwise preference data. The pairwise preference data includes market-specific training examples, while a ranking model from another market captures the common characteristics of the resulting ranking model. In one embodiment, the ranking model is trained by applying a Tree Based Ranking Function Adaptation (TRADA) algorithm to multi-grade labeled training data, such as editorially generated training data. Then, contradictions between the TRADA generated ranking model and target-market specific pairwise preference data are identified. For each identified contradiction, new training data is generated to correct the contradiction. Then, in one embodiment, an algorithm such as TRADA is applied to the existing ranking model and the new training data to generate a new ranking model.
摘要翻译：实施例旨在通过利用目标市场特定的成对偏好数据来产生市场特定的排名模型。成对偏好数据包括市场特定的培训示例，而来自另一个市场的排名模型捕获了所得到的排名模型的共同特征。在一个实施例中，通过将基于树的排序函数适应（TRADA）算法应用于诸如编辑生成的训练数据的多等级标记的训练数据来训练排名模型。然后，确定了TRADA产生的排名模型和目标市场特定成对偏好数据之间的矛盾。对于每个确定的矛盾，产生新的训练数据以纠正矛盾。然后，在一个实施例中，诸如TRADA的算法被应用于现有的排名模型和新的训练数据以生成新的排名模型。

3. 发明申请

US20110093459A1 Incorporating Recency in Network Search Using Machine Learning 有权
标题翻译：使用机器学习在网络搜索中结合新近度
公开(公告)号：US20110093459A1
公开(公告)日：2011-04-21
申请号：US12579855
申请日：2009-10-15
申请人： Anlei Dong , Yi Chang , Ruiqiang Zhang , Zhaohui Zheng , Gilad Avraham Mishne , Jing Bai , Karolina Barbara Buchner , Ciya Liao , Shihao Ji , Gilbert Leung , Georges-Eric Albert Marie Robert Dupret , Ling Liu
发明人： Anlei Dong , Yi Chang , Ruiqiang Zhang , Zhaohui Zheng , Gilad Avraham Mishne , Jing Bai , Karolina Barbara Buchner , Ciya Liao , Shihao Ji , Gilbert Leung , Georges-Eric Albert Marie Robert Dupret , Ling Liu
IPC分类号： G06F17/30
CPC分类号： G06F17/30867
摘要： In one embodiment, access a set of recency ranking data comprising one or more recency search queries and one or more recency search results, each of the recency search queries being recency-sensitive with respect to a particular time period and being associated with a query timestamp representing the time at which the recency search query is received at a search engine, each of the recency search results being generated by the search engine for one of the recency search queries and comprising one or more recency network resources. Construct a plurality of recency features from the set of recency ranking data. Train a first ranking model via machine learning using at least the recency features.
摘要翻译：在一个实施例中，访问包括一个或多个新近度搜索查询和一个或多个新近度搜索结果的一组新近度排序数据，每个新近度搜索查询相对于特定时间段是新近敏感度，并且与查询时间戳相关联表示在搜索引擎处接收到新近度搜索查询的时间，每个新近度搜索结果由搜索引擎为新近搜索查询之一生成，并且包括一个或多个新近网络资源。从新近度排序数据集合构建多个新近特征。通过机器学习，至少使用新特性来训练第一个排名榜。

4. 发明授权

US08886641B2 Incorporating recency in network search using machine learning 有权
标题翻译：使用机器学习将新近度纳入网络搜索
公开(公告)号：US08886641B2
公开(公告)日：2014-11-11
申请号：US12579855
申请日：2009-10-15
申请人： Anlei Dong , Yi Chang , Ruiqiang Zhang , Zhaohui Zheng , Gilad Avraham Mishne , Jing Bai , Karolina Barbara Buchner , Ciya Liao , Shihao Ji , Gilbert Leung , Georges-Eric Albert Marie Robert Dupret , Ling Liu
发明人： Anlei Dong , Yi Chang , Ruiqiang Zhang , Zhaohui Zheng , Gilad Avraham Mishne , Jing Bai , Karolina Barbara Buchner , Ciya Liao , Shihao Ji , Gilbert Leung , Georges-Eric Albert Marie Robert Dupret , Ling Liu
IPC分类号： G06F17/30
CPC分类号： G06F17/30867
摘要： In one embodiment, access a set of recency ranking data comprising one or more recency search queries and one or more recency search results, each of the recency search queries being recency-sensitive with respect to a particular time period and being associated with a query timestamp representing the time at which the recency search query is received at a search engine, each of the recency search results being generated by the search engine for one of the recency search queries and comprising one or more recency network resources. Construct a plurality of recency features from the set of recency ranking data. Train a first ranking model via machine learning using at least the recency features.
摘要翻译：在一个实施例中，访问包括一个或多个新近度搜索查询和一个或多个新近度搜索结果的一组新近度排序数据，每个新近度搜索查询相对于特定时间段是新近敏感度，并且与查询时间戳相关联表示在搜索引擎处接收到新近度搜索查询的时间，每个新近度搜索结果由搜索引擎为新近搜索查询之一生成，并且包括一个或多个新近网络资源。从新近度排序数据集合构建多个新近特征。通过机器学习，至少使用新特性来训练第一个排名榜。

5. 发明授权

US08751511B2 Ranking of search results based on microblog data 有权
标题翻译：基于微博数据的搜索结果排名
公开(公告)号：US08751511B2
公开(公告)日：2014-06-10
申请号：US12749972
申请日：2010-03-30
申请人： Anlei Dong , Pranam Kolari , Ruiqiang Zhang , Jing Bai , Yi Chang , Zhaohui Zheng
发明人： Anlei Dong , Pranam Kolari , Ruiqiang Zhang , Jing Bai , Yi Chang , Zhaohui Zheng
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06Q10/06 , G06F17/30864 , G06Q50/01
摘要： An information retrieval system is described herein that monitors a microblog data stream that includes microblog posts to discover and index fresh resources for searching by a search engine. The information retrieval system also uses data from the microblog data stream as well as data obtained from a microblog subscription system to compute novel and effective features for ranking fresh resources which would otherwise have impoverished representations. An embodiment of the present invention advantageously enables a search engine to produce a fresher set of resources and to rank such resources for both relevancy and freshness in a more accurate manner.
摘要翻译：本文描述了一种信息检索系统，其监视包括微博帖子的微博数据流，以发现和索引新的资源以供搜索引擎搜索。信息检索系统还使用来自微博数据流的数据以及从微博订阅系统获得的数据来计算新颖而有效的特征来排列否则将具有贫困表示的新鲜资源。本发明的一个实施例有利地使得搜索引擎能够产生更新鲜的资源集合，并且以更准确的方式对相关性和新鲜度进行排名。

6. 发明申请

US20110246457A1 RANKING OF SEARCH RESULTS BASED ON MICROBLOG DATA 有权
标题翻译：基于MICROBLOG数据的搜索结果排名
公开(公告)号：US20110246457A1
公开(公告)日：2011-10-06
申请号：US12749972
申请日：2010-03-30
申请人： Anlei Dong , Pranam Kolari , Ruiqiang Zhang , Jing Bai , Yi Chang , Zhaohui Zheng
发明人： Anlei Dong , Pranam Kolari , Ruiqiang Zhang , Jing Bai , Yi Chang , Zhaohui Zheng
IPC分类号： G06F17/30 , G06Q99/00 , G06F15/18
CPC分类号： G06Q10/06 , G06F17/30864 , G06Q50/01
摘要： An information retrieval system is described herein that monitors a microblog data stream that includes microblog posts to discover and index fresh resources for searching by a search engine. The information retrieval system also uses data from the microblog data stream as well as data obtained from a microblog subscription system to compute novel and effective features for ranking fresh resources which would otherwise have impoverished representations. An embodiment of the present invention advantageously enables a search engine to produce a fresher set of resources and to rank such resources for both relevancy and freshness in a more accurate manner.
摘要翻译：本文描述了一种信息检索系统，其监视包括微博帖子的微博数据流，以发现和索引新的资源以供搜索引擎搜索。信息检索系统还使用来自微博数据流的数据以及从微博订阅系统获得的数据来计算新颖而有效的特征来排列否则将具有贫困表示的新鲜资源。本发明的一个实施例有利地使得搜索引擎能够产生更新鲜的资源集合，并且以更准确的方式对相关性和新鲜度进行排名。

7. 发明申请

US20130044786A1 SYSTEMS AND METHODS FOR PEAK JUNCTION TEMPERATURE SENSING AND THERMAL SAFE OPERATING AREA PROTECTION 有权
标题翻译：峰值温度传感和热安全操作区域保护的系统和方法
公开(公告)号：US20130044786A1
公开(公告)日：2013-02-21
申请号：US13563484
申请日：2012-07-31
申请人： Anindya Bhattacharya , Tim Green , Jing Bai
发明人： Anindya Bhattacharya , Tim Green , Jing Bai
IPC分类号： G01K3/00
CPC分类号： G01K3/10 , G01K3/02 , G01K7/427
摘要： A peak junction temperature monitoring system for a semiconductor device includes a peak power dissipation sensor for sensing the peak power dissipation in the device. A temperature sensor senses an average temperature of the device, and a peak junction temperature computation circuit generates a signal representative of a peak junction temperature based on input from the peak power dissipation sensor and the temperature sensor.
摘要翻译：用于半导体器件的峰值结温监测系统包括峰值功耗传感器，用于感测器件中的峰值功耗。温度传感器感测器件的平均温度，并且峰值结温度计算电路基于来自峰值功耗传感器和温度传感器的输入产生表示峰值结温的信号。

8. 发明申请

US20140284092A1 SPLIT PAD FOR CIRCUIT BOARD 审中-公开
标题翻译：电路板分接盘
公开(公告)号：US20140284092A1
公开(公告)日：2014-09-25
申请号：US14178303
申请日：2014-02-12
申请人： Jing Bai , Yin Guo , Shayan Zhang , Yanyan Zhang
发明人： Jing Bai , Yin Guo , Shayan Zhang , Yanyan Zhang
IPC分类号： H05K1/02 , H05K1/11
CPC分类号： H05K1/112 , H05K1/0268 , H05K3/225 , H05K3/3421 , H05K2201/09663 , H05K2203/176
摘要： An electronic device such as a circuit board has a contact pad for connection to a contact of a component, and a pad portion interconnection. The contact pad has physically separate pad portions. The pad portion interconnection electrically connects the pad portions of the contact pad, independently of any mounted connection on the pad portions. Providing multiple pad portions for a single contact pad allows the contact pad to function even if one of the pad portions is damaged such as by peeling off. An example application is an EMC (Electromagnetic Compatibility) and/or ESD (Electro-Static Discharge) test circuit board.
摘要翻译：诸如电路板的电子装置具有用于连接到部件的触点和焊盘部分互连的接触焊盘。接触垫具有物理上分离的垫部分。焊盘部分互连电连接接触焊盘的焊盘部分，独立于焊盘部分上的任何安装的连接。为单个接触垫提供多个焊盘部分允许接触焊盘即使其中一个焊盘部分被损坏（例如通过剥离）也起作用。示例应用是EMC（电磁兼容性）和/或ESD（静电放电）测试电路板。

9. 发明申请

US20130327897A1 METHOD FOR IMPROVING OPERATION DENSITY OF RAIL VEHICLES AND PREVENTING HEAD-ON COLLISION AND REAR-ENDING COLLISION 有权
标题翻译：改善铁路车辆运行密度和防止碰撞和后期碰撞的方法
公开(公告)号：US20130327897A1
公开(公告)日：2013-12-12
申请号：US13825262
申请日：2011-08-09
申请人： Wei Bai , Jing Bai , Qing Bai , Baolong Feng
发明人： Wei Bai , Jing Bai , Qing Bai , Baolong Feng
IPC分类号： B61L23/26
CPC分类号： B61L23/26 , B61L21/10 , B61L23/14 , B61L23/18 , B61L23/30
摘要： The present invention provides a method for improving operation density of rail vehicles and for preventing head-on collision and rear-ending collision. Said method divides a rail line into equidistant electronic zones, the length of a zone being greater than the shortest safe distance between two running vehicles. Said method installs a locomotive passing detection alarm device in each zone, when a locomotive travels at high speed on the rail, the locomotive passing detection alarm device corresponding to the zone occupied by the locomotive itself will simultaneously access adjacent front and back zones, and determine whether the two adjacent zones are simultaneously occupied by locomotives. If the two adjacent. zones are simultaneously occupied by locomotives, the locomotive passing alarm device will send an alarm signal to the locomotives to warn or otherwise take measures. The aforesaid method can avoid locomotive head-on collision and rear-end collision and increase transportation density according to the vehicle speed and distance at the same time, thus improving the transportation efficiency.
摘要翻译：本发明提供了一种提高轨道车辆的运行密度和防止碰撞和后端碰撞的方法。所述方法将铁路线划分成等距离的电子区域，区域的长度大于两辆运行车辆之间的最短安全距离。所述方法在每个区域安装机车通过检测报警装置，当机车在轨道上高速行驶时，对应于机车本身占据的区域的机车通过检测报警装置将同时访问相邻的前后区域，并确定两个相邻区域是否同时被机车占据。如果两个相邻。机车同时占用机车，机车通过报警装置将向机车发出报警信号，以警告或采取措施。上述方法可以避免机车前后碰撞和后端碰撞，同时根据车速和距离增加运输密度，从而提高运输效率。

10. 发明申请

US20090018451A1 Dynamic Sampling System and Method for In Vivo Fluorescent Molecular Imaging 审中-公开
标题翻译：动态采样系统和体内荧光分子成像方法
公开(公告)号：US20090018451A1
公开(公告)日：2009-01-15
申请号：US12053355
申请日：2008-03-21
申请人： Jing Bai , Xiaolei Song , Gang Hu , Junjie Yao , Zhun Xu , Daifa Wang , Jianchang Ou , Guoliang Huang , Yonghong Zhang
发明人： Jing Bai , Xiaolei Song , Gang Hu , Junjie Yao , Zhun Xu , Daifa Wang , Jianchang Ou , Guoliang Huang , Yonghong Zhang
IPC分类号： A61B5/00
CPC分类号： G01N21/6456 , A61B5/0073 , A61B6/037 , A61B6/4417 , A61B6/5247 , A61B2503/40 , G01N21/6428 , G01N2021/1787
摘要： A dynamic dada sampling system and method is disclosed for in vivo small animal fluorescence molecular imaging and dual-modality molecular imaging. The system comprises a computer, a rotation stage for animal suspension driven by a motor, and a fluorescence excitation-detection apparatus. The fluorescence excitation-detection apparatus comprises a fluorescence excitation module and a fluorescence detection module. The CCD device of the fluorescence detection module is connected to a computer through an interface controller. The motor is connected to a computer through RS232 interface. The process of dynamic data acquisition is as follows: a fluorescent probe is injected into a small animal in order to target specific cells or tissues; a small animal is vertically hung on a rotation stage after anesthesia; the fluorescence imaging detection module acquires the emitting light continuously. The present invention can provide 360 degree imaging quickly, efficiently, and non-invasively.
摘要翻译：公开了用于体内小动物荧光分子成像和双重模态分子成像的动态dada采样系统和方法。该系统包括计算机，由电动机驱动的用于动物悬架的旋转台和荧光激发检测装置。荧光激发检测装置包括荧光激发模块和荧光检测模块。荧光检测模块的CCD器件通过接口控制器连接到计算机。电机通过RS232接口连接到电脑。动态数据采集过程如下：将荧光探针注射到小动物中以靶向特定细胞或组织; 麻醉后，一只小动物垂直悬挂在旋转台上; 荧光成像检测模块连续获取发光。本发明可以快速，有效和非侵入地提供360度成像。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式