公开(公告)号：US20170270407A1

公开(公告)日：2017-09-21

申请号：US15407470

申请日：2017-01-17

Applicant: Google Inc.

Inventor： Christopher Alberti ,  Aliaksei Severyn ,  Daniel Andor ,  Slav Petrov ,  Kuzman Ganchev Ganchev ,  David Joseph Weiss ,  Michael John Collins ,  Alessandro Presta

IPC: G06N3/08 ,  G06N3/04

CPC classification number: G06N3/08 ,  G06N3/04 ,  G06N5/003 ,  G06N7/005

Abstract: A method includes training a neural network having parameters on training data, in which the neural network receives an input state and processes the input state to generate a respective score for each decision in a set of decisions. The method includes receiving training data including training text sequences and, for each training text sequence, a corresponding gold decision sequence. The method includes training the neural network on the training data to determine trained values of parameters of the neural network. Training the neural network includes for each training text sequence: maintaining a beam of candidate decision sequences for the training text sequence, updating each candidate decision sequence by adding one decision at a time, determining that a gold candidate decision sequence matching a prefix of the gold decision sequence has dropped out of the beam, and in response, performing an iteration of gradient descent to optimize an objective function.

公开(公告)号：US09507852B2

公开(公告)日：2016-11-29

申请号：US14102087

申请日：2013-12-10

Applicant: Google Inc.

Inventor： Slav Petrov ,  Alexander Rush

IPC: G10L15/18 ,  G06F17/30 ,  G06F17/27 ,  G10L15/26

CPC classification number: G06F17/30654 ,  G06F17/2775 ,  G10L15/1822 ,  G10L15/26

Abstract: A computer-implemented method can include receiving a speech input representing a question, converting the speech input to a string of characters, and obtaining tokens each representing a potential word. The method can include determining one or more part-of-speech (POS) tags for each token and determining sequences of the POS tags for the tokens, each sequence of the POS tags including one POS tag per token. The method can include determining one or more parses for each sequence of the POS tags for the tokens and determining a most-likely parse and its corresponding sequence of the POS tags for the tokens to obtain a selected parse and a selected sequence of the POS tags for the tokens. The method can also include determining a most-likely answer to the question using the selected parse and the selected sequence of the POS tags for the tokens and outputting the most-likely answer.

Abstract translation: 计算机实现的方法可以包括接收表示问题的语音输入，将语音输入转换成一串字符，以及获得每个代表潜在词的令牌。 该方法可以包括为每个令牌确定一个或多个词性（POS）标签，并确定令牌的POS标签的序列，每个标记的每个POS标签序列包括一个POS标签。 该方法可以包括为令牌的POS标签的每个序列确定一个或多个解析，并确定用于令牌的POS标签的最有可能的解析及其相应序列，以获得所选择的解析和所选择的POS标签序列 为令牌。 该方法还可以包括使用所选择的解析和用于令牌的所选择的POS标签序列来确定问题的最可能的答案并输出最可能的答案。

公开(公告)号：US20150161996A1

公开(公告)日：2015-06-11

申请号：US14102087

申请日：2013-12-10

Applicant: Google Inc.

Inventor： Slav Petrov ,  Alexander Rush

IPC: G10L15/18

CPC classification number: G06F17/30654 ,  G06F17/2775 ,  G10L15/1822 ,  G10L15/26

Abstract: A computer-implemented method can include receiving a speech input representing a question, converting the speech input to a string of characters, and obtaining tokens each representing a potential word. The method can include determining one or more part-of-speech (POS) tags for each token and determining sequences of the POS tags for the tokens, each sequence of the POS tags including one POS tag per token. The method can include determining one or more parses for each sequence of the POS tags for the tokens and determining a most-likely parse and its corresponding sequence of the POS tags for the tokens to obtain a selected parse and a selected sequence of the POS tags for the tokens. The method can also include determining a most-likely answer to the question using the selected parse and the selected sequence of the POS tags for the tokens and outputting the most-likely answer.

Abstract translation: 计算机实现的方法可以包括接收表示问题的语音输入，将语音输入转换成一串字符，以及获得每个代表潜在词的令牌。 该方法可以包括为每个令牌确定一个或多个词性（POS）标签，并确定令牌的POS标签的序列，每个标记的每个POS标签序列包括一个POS标签。 该方法可以包括为令牌的POS标签的每个序列确定一个或多个解析，并确定用于令牌的POS标签的最有可能的解析及其相应序列，以获得所选择的解析和所选择的POS标签序列 为令牌。 该方法还可以包括使用所选择的解析和用于令牌的所选择的POS标签序列来确定问题的最可能的答案并输出最可能的答案。

公开(公告)号：US09311299B1

公开(公告)日：2016-04-12

申请号：US13955491

申请日：2013-07-31

Applicant: Google Inc.

Inventor： Slav Petrov ,  Dipanjan Das ,  Ryan McDonald ,  Joakim Nivre ,  Oscar Tackstrom

IPC: G06F17/28 ,  G06F17/27

CPC classification number: G06F17/28 ,  G06F17/271 ,  G06F17/2785 ,  G06F17/2827

Abstract: A method and system are provided for a part-of-speech tagger that may be particularly useful for resource-poor languages. Use of manually constructed tag dictionaries from dictionaries via bitext can be used as type constraints to overcome the scarcity of annotated data in some instances. Additional token constraints can be projected from a resource-rich source language via word-aligned bitext. Several example models are provided to demonstrate this such as a partially observed conditional random field model, where coupled token and type constraints may provide a partial signal for training. The disclosed method achieves a significant relative error reduction over the prior state of the art.

Abstract translation: 为可能对资源贫乏的语言特别有用的词性标签器提供了一种方法和系统。 通过bitext使用手工构建的字典字典可用作类型约束来克服某些情况下注释数据的稀缺性。 额外的令牌约束可以从资源丰富的源语言通过字对齐的bitext进行投影。 提供了几个示例模型来证明这一点，例如部分观察到的条件随机场模型，其中耦合的令牌和类型约束可以提供用于训练的部分信号。 所公开的方法相对于现有技术的现有技术实现了显着的相对误差减小。

公开(公告)号：US09305544B1

公开(公告)日：2016-04-05

申请号：US14594900

申请日：2015-01-12

Applicant: Google Inc.

Inventor： Slav Petrov ,  Ryan McDonald ,  Keith Hall

IPC: G06F17/20 ,  G06F17/28 ,  G06F17/27 ,  G06F17/21 ,  G10L15/00 ,  G10L17/00 ,  G10L21/00 ,  G10L15/06

CPC classification number: G10L15/06 ,  G06F17/2705 ,  G06F17/2827 ,  G06F17/2854 ,  G10L15/1822

Abstract: A source language sentence is tagged with non-lexical tags, such as part-of-speech tags and is parsed using a lexicalized parser trained in the source language. A target language sentence that is a translation of the source language sentence is tagged with non-lexical labels (e.g., part-of speech tags) and is parsed using a delexicalized parser that has been trained in the source language to produce k-best parses. The best parse is selected based on the parse's alignment with lexicalized parse of the source language sentence. The selected best parse can be used to update the parameter vector of a lexicalized parser for the target language.

公开(公告)号：US08914279B1

公开(公告)日：2014-12-16

申请号：US13624280

申请日：2012-09-21

Applicant: Google Inc.

Inventor： Slav Petrov ,  Alexander Rush

IPC: G06F17/27 ,  G06F17/20 ,  G06F17/21

CPC classification number: G06F17/2715 ,  G06F17/271

Abstract: A dependency parsing method can include determining an index set of possible head-modifier dependencies for a sentence. The index set can include inner arcs and outer arcs, inners arcs representing possible dependency between words in the sentence separated by a distance less than or equal to a threshold and outer arcs representing possible dependency between words in the sentence separated by a distance greater than the threshold. The index set can be pruned to include: (i) each specific inner arc when a likelihood that the specific inner arc is appropriate is greater than a first threshold, and (ii) the outer arcs when a likelihood that there exists any possible outer arc that is appropriate is greater than the first threshold. The method can include further pruning the pruned index set based on a second parsing algorithm, and determining a most-likely parse for the sentence from the pruned index set.

Abstract translation: 依赖性解析方法可以包括确定句子的可能的头修饰符依赖性的索引集合。 索引集可以包括内弧和外弧，内圆弧表示句子中的词之间的距离小于或等于阈值的可能依赖性，外弧表示句子中的词之间的距离大于 阈。 索引集可以被修剪为包括：（i）当特定内弧适合的可能性大于第一阈值时的每个特定内弧，以及（ii）当存在任何可能的外弧的可能性时的外弧 这是适当的大于第一阈值。 该方法可以包括基于第二解析算法进一步修剪修剪的索引集合，并且从修剪的索引集合确定对于该句子的最有可能的解析。