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. 发明申请

US20050071148A1 Chinese word segmentation 审中-公开
标题翻译：中文分词
公开(公告)号：US20050071148A1
公开(公告)日：2005-03-31
申请号：US10662602
申请日：2003-09-15
申请人： Chang-Ning Huang , Jianfeng Gao , Mu Li , Ashley Chang
发明人： Chang-Ning Huang , Jianfeng Gao , Mu Li , Ashley Chang
IPC分类号： G06F17/20 , G06F17/27 , G06F17/28 , G10L13/06
CPC分类号： G06F17/2755 , G06F17/277 , G06F17/2863
摘要： The present invention relates to a corpus for use in training a language model. The corpus includes a plurality of characters and a plurality of morphological tags associated with a plurality of sequences of characters. The plurality of morphological tags indicate a morphological type of an associated sequence of characters and a combination of parts forming a morphological subtype.
摘要翻译：本发明涉及一种用于训练语言模型的语料库。语料库包括与多个字符序列相关联的多个字符和多个形态标签。多个形态标签表示相关联的字符序列的形态类型和形成形态亚型的部分的组合。

2. 发明申请

US20060287848A1 Language classification with random feature clustering 审中-公开
标题翻译：语言分类与随机特征聚类
公开(公告)号：US20060287848A1
公开(公告)日：2006-12-21
申请号：US11157091
申请日：2005-06-20
申请人： Mu Li , Jianfeng Gao , Ming Zhou
发明人： Mu Li , Jianfeng Gao , Ming Zhou
IPC分类号： G06F17/27
CPC分类号： G06F16/355
摘要： An ensemble of random feature clusters is built from training data using a clustering algorithm where some randomness has been introduced. For each clustered feature space, a classifier, such as a Naïve Bayesian Classifier, is trained, realizing a classifier ensemble. The final classification decision is made by the resulting classifier ensemble.
摘要翻译：随机特征群集由训练数据构建，使用聚类算法，其中引入了一些随机性。对于每个聚类特征空间，训练一个分类器，如朴素贝叶斯分类器，实现分类器集合。最终的分类决定是由所得到的分类器集合决定的。

3. 发明授权

US07818332B2 Query speller 有权
标题翻译：查询拼写器
公开(公告)号：US07818332B2
公开(公告)日：2010-10-19
申请号：US11465023
申请日：2006-08-16
申请人： Elliott K. Olds , Gregory N. Hullender , Haoyong Zhang , Janine R. Crumb , Jianfeng Gao , Ming Zhou , Mu Li , Yajuan Lv
发明人： Elliott K. Olds , Gregory N. Hullender , Haoyong Zhang , Janine R. Crumb , Jianfeng Gao , Ming Zhou , Mu Li , Yajuan Lv
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06F17/3064
摘要： Candidate suggestions for correcting misspelled query terms input into a search application are automatically generated. A score for each candidate suggestion can be generated using a first decoding pass and paths through the suggestions can be ranked in a second decoding pass. Candidate suggestions can be generated based on typographical errors, phonetic mistakes and/or compounding mistakes. Furthermore, a ranking model can be developed to rank candidate suggestions to be presented to a user.
摘要翻译：自动生成用于纠正输入到搜索应用程序中的拼错查询条件的候选建议。可以使用第一解码通道来生成每个候选建议的得分，并且通过建议的路径可以被排列在第二解码通行证中。可以根据印刷错误，语音错误和/或复合错误生成候选建议。此外，可以开发排名模型来排列要呈现给用户的候选建议。

4. 发明授权

US07231349B2 Method and apparatus for compressing asymmetric clustering language models 有权
标题翻译：用于压缩非对称聚类语言模型的方法和装置
公开(公告)号：US07231349B2
公开(公告)日：2007-06-12
申请号：US10448498
申请日：2003-05-30
申请人： Mu Li , Jianfeng Gao
发明人： Mu Li , Jianfeng Gao
IPC分类号： G01L15/06
CPC分类号： G10L15/197 , G06F17/277 , G10L15/285
摘要： A method and data structure are provided for efficiently storing asymmetric clustering models. The models are stored by storing a first level record for a word identifier and two second level records, one for a word identifier and one for a cluster identifier. An index to the second level word record and an index to the second level cluster record are stored in the first level record. Many of the records in the data structure include both cluster sub-model parameters and word sub-model parameters.
摘要翻译：提供了一种方法和数据结构，用于有效地存储非对称聚类模型。通过存储用于字标识符的第一级记录和两个第二级记录来存储模型，一个用于字标识符，一个用于集群标识符。第二级记录的索引和第二级集群记录的索引存储在第一级记录中。数据结构中的许多记录包括集群子模型参数和单词子模型参数。

5. 发明申请

US20070078644A1 Detecting segmentation errors in an annotated corpus 审中-公开
标题翻译：检测注释语料库中的分段错误
公开(公告)号：US20070078644A1
公开(公告)日：2007-04-05
申请号：US11241037
申请日：2005-09-30
申请人： Chang-Ning Huang , Jianfeng Gao , Mu Li
发明人： Chang-Ning Huang , Jianfeng Gao , Mu Li
IPC分类号： G06F17/27
CPC分类号： G06F17/2863 , G06F17/277 , G06F17/278
摘要： Segmentation error candidates are detected using segmentation variations found in an annotated corpus.
摘要翻译：使用在注释语料库中发现的分段变异来检测分段错误候选。

6. 发明申请

US20060282255A1 Collocation translation from monolingual and available bilingual corpora 审中-公开
标题翻译：单语和双语语料库的翻译
公开(公告)号：US20060282255A1
公开(公告)日：2006-12-14
申请号：US11152540
申请日：2005-06-14
申请人： Yajuan Lu , Jianfeng Gao , Ming Zhou , John Chen , Mu Li
发明人： Yajuan Lu , Jianfeng Gao , Ming Zhou , John Chen , Mu Li
IPC分类号： G06F17/28
CPC分类号： G06F17/2827
摘要： A system and method of extracting collocation translations is presented. The methods include constructing a collocation translation model using monolingual source and target language corpora as well as bilingual corpus, if available. The collocation translation model employs an expectation maximization algorithm with respect to contextual words surrounding collocations. The collocation translation model can be used later to extract a collocation translation dictionary. Optional filters based on context redundancy and/or bi-directional translation constrain can be used to ensure that only highly reliable collocation translations are included in the dictionary. The constructed collocation translation model and the extracted collocation translation dictionary can be used later for further natural language processing, such as sentence translation.
摘要翻译：提出了一种提取搭配翻译的系统和方法。这些方法包括使用单语源语言和目标语言语料库以及双语语料库（如果可用）来构建搭配翻译模型。搭配翻译模型采用围绕搭配的上下文单词的期望最大化算法。搭配翻译模型可以随后用于提取搭配翻译字典。可以使用基于上下文冗余和/或双向转换约束的可选过滤器来确保字典中仅包含高度可靠的并置转换。构建的搭配翻译模型和提取的搭配翻译词典可以稍后用于进一步的自然语言处理，如句子翻译。

7. 发明授权

US07493251B2 Using source-channel models for word segmentation 有权
标题翻译：使用源通道模型进行分词
公开(公告)号：US07493251B2
公开(公告)日：2009-02-17
申请号：US10448644
申请日：2003-05-30
申请人： Jianfeng Gao , Mu Li , Chang-Ning Huang , Jian Sun , Lei Zhang , Ming Zhou
发明人： Jianfeng Gao , Mu Li , Chang-Ning Huang , Jian Sun , Lei Zhang , Ming Zhou
IPC分类号： G06F17/27 , G10L11/00
CPC分类号： G06F17/2755 , G06F17/277
摘要： A method and apparatus for segmenting text is provided that identifies a sequence of entity types from a sequence of characters and thereby identifies a segmentation for the sequence of characters. Under the invention, the sequence of entity types is identified using probabilistic models that describe the likelihood of a sequence of entities and the likelihood of sequences of characters given particular entities. Under one aspect of the invention, organization name entities are identified from a first sequence of identified entities to form a final sequence of identified entities.
摘要翻译：提供了用于分割文本的方法和装置，其从字符序列识别实体类型的序列，从而识别字符序列的分割。在本发明下，使用描述实体序列的可能性的概率模型和给定特定实体的字符序列的可能性来识别实体类型的序列。在本发明的一个方面，从识别的实体的第一序列识别组织名称实体，以形成所识别实体的最终序列。

8. 发明申请

US20080046405A1 QUERY SPELLER 有权
公开(公告)号：US20080046405A1
公开(公告)日：2008-02-21
申请号：US11465023
申请日：2006-08-16
申请人： Elliott K. Olds , Gregory N. Hullender , Haoyong Zhang , Janine R. Crumb , Jianfeng Gao , Ming Zhou , Mu Li , Yajuan Lv
发明人： Elliott K. Olds , Gregory N. Hullender , Haoyong Zhang , Janine R. Crumb , Jianfeng Gao , Ming Zhou , Mu Li , Yajuan Lv
IPC分类号： G06F17/30
CPC分类号： G06F17/3064
摘要： Candidate suggestions for correcting misspelled query terms input into a search application are automatically generated. A score for each candidate suggestion can be generated using a first decoding pass and paths through the suggestions can be ranked in a second decoding pass. Candidate suggestions can be generated based on typographical errors, phonetic mistakes and/or compounding mistakes. Furthermore, a ranking model can be developed to rank candidate suggestions to be presented to a user.
摘要翻译：自动生成用于纠正输入到搜索应用程序中的拼错查询条件的候选建议。可以使用第一解码通道来生成每个候选建议的得分，并且通过建议的路径可以被排列在第二解码通行证中。可以根据印刷错误，语音错误和/或复合错误生成候选建议。此外，可以开发排名模型来排列要呈现给用户的候选建议。

9. 发明申请

US20050060150A1 Unsupervised training for overlapping ambiguity resolution in word segmentation 审中-公开
标题翻译：用于重叠模糊度分辨率的无监督训练
公开(公告)号：US20050060150A1
公开(公告)日：2005-03-17
申请号：US10662502
申请日：2003-09-15
申请人： Mu Li , Jianfeng Gao
发明人： Mu Li , Jianfeng Gao
IPC分类号： G06F17/27 , G06F17/28 , G10L15/00
CPC分类号： G06F17/2863 , G06F17/2775
摘要： A method for resolving overlapping ambiguity strings in unsegmented languages such as Chinese. The methodology includes segmenting sentences into two possible segmentations and recognizing overlapping ambiguity strings in the sentences. One of the two possible segmentations is selected as a function of probability information. The probability information is derived from unsupervised training data. A method of constructing a knowledge base containing probability information needed to select one of the segmentation is also provided.
摘要翻译：用于解析诸如中文的未分段语言中的重叠歧义字符串的方法。该方法包括将句子分割成两个可能的分段，并识别句子中的重叠歧义字符串。作为概率信息的函数选择两个可能的分段中的一个。概率信息是从无监督的训练数据导出的。还提供了构建包含选择分割之一所需的概率信息的知识库的方法。

10. 发明授权

US08645289B2 Structured cross-lingual relevance feedback for enhancing search results 有权
标题翻译：结构化的跨语言相关性反馈，以增强搜索结果
公开(公告)号：US08645289B2
公开(公告)日：2014-02-04
申请号：US12970879
申请日：2010-12-16
申请人： Paul Nathan Bennett , Jianfeng Gao , Jagadeesh Jagarlamudi , Kristen Patricia Parton
发明人： Paul Nathan Bennett , Jianfeng Gao , Jagadeesh Jagarlamudi , Kristen Patricia Parton
IPC分类号： G06F15/18
CPC分类号： G06F17/30669 , G06F17/30675
摘要： A “Cross-Lingual Unified Relevance Model” provides a feedback model that improves a machine-learned ranker for a language with few training resources, using feedback from a more complete ranker for a language that has more training resources. The model focuses on linguistically non-local queries, such as “world cup” (English language/U.S. market) and “copa mundial” (Spanish language/Mexican market), that have similar user intent in different languages and markets or regions, thus allowing the low-resource ranker to receive direct relevance feedback from the high-resource ranker. Among other things, the Cross-Lingual Unified Relevance Model differs from conventional relevancy-based techniques by incorporating both query- and document-level features. More specifically, the Cross-Lingual Unified Relevance Model generalizes existing cross-lingual feedback models, incorporating both query expansion and document re-ranking to further amplify the signal from the high-resource ranker to enable a learning to rank approach based on appropriately labeled training data.
摘要翻译： “跨语言统一相关性模型”提供了一种反馈模型，可以为少数培训资源的语言改进机器学习游戏者，使用更完整的游戏者的反馈来获得具有更多培训资源的语言。该模式侧重于语言上的非本地查询，例如“世界杯”（英语/美国市场）和“复合世界”（西班牙语/墨西哥市场），在不同语言和市场或区域具有类似的用户意图，因此允许低资源游击队员接收来自高资源队员的直接相关反馈。其中，跨语言统一相关性模型与传统的相关性技术不同，包括查询和文档级功能。更具体地说，跨语言统一相关性模型概括了现有的跨语言反馈模型，其中包括查询扩展和文档重新排序，以进一步放大来自高资源游戏者的信号，以使学习能够基于适当标记的训练进行排名数据。

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