公开(公告)号：US20140270497A1

公开(公告)日：2014-09-18

申请号：US14292294

申请日：2014-05-30

Applicant: Microsoft Corporation

Inventor： Anitha Kannan ,  Partha Pratim Talukdar ,  Nikhil Rasiwasia ,  Qifa Ke ,  Rakesh Agrawal

IPC: G06K9/62

CPC classification number: G06K9/6268 ,  G06K9/6227 ,  G06K9/6293

Abstract: Product images are used in conjunction with textual descriptions to improve classifications of product offerings. By combining cues from both text and image descriptions associated with products, implementations enhance both the precision and recall of product description classifications within the context of web-based commerce search. Several implementations are directed to improving those areas where text-only approaches are most unreliable. For example, several implementations use image signals to complement text classifiers and improve overall product classification in situations where brief textual product descriptions use vocabulary that overlaps with multiple diverse categories. Other implementations are directed to using text and images “training sets” to improve automated classifiers including text-only classifiers. Certain implementations are also directed to learning a number of three-way image classifiers focused only on “confusing categories” of the text signals to improve upon those specific areas where text-only classification is weakest.

Abstract translation: 产品图像与文本描述结合使用，以改进产品分类。 通过结合来自与产品相关的文本和图像描述的提示，实现在基于网络的商业搜索的上下文中增强了产品描述分类的精度和回收。 几个实现旨在改进那些仅文本方法最不可靠的领域。 例如，在简短的文本产品描述使用与多个不同类别重叠的词汇的情况下，多个实现使用图像信号来补充文本分类器并改进整体产品分类。 其他实现涉及使用文本和图像“训练集”来改进自动分类器，包括纯文本分类器。 某些实现也针对学习一些三维图像分类器，仅针对文本信号的“混淆类别”，以改善文本分类最弱的特定领域。

公开(公告)号：US08843519B2

公开(公告)日：2014-09-23

申请号：US13929917

申请日：2013-06-28

Applicant: Microsoft Corporation

Inventor： Christopher Armstrong ,  Alan Keister ,  Lark Dunham ,  Jennifer Granito ,  Steven Greenberg ,  Rakesh Agrawal ,  Michael Sommers ,  Carol A. Glover

IPC: G06F17/30 ,  H04L12/58 ,  H04L29/08 ,  G06Q50/00 ,  G06Q10/10

CPC classification number: H04L67/306 ,  G06F3/0482 ,  G06F17/30867 ,  G06F17/3089 ,  G06Q10/107 ,  G06Q50/01 ,  H04L51/04 ,  H04L51/32 ,  H04L65/403 ,  H04L67/02 ,  H04L67/26

Abstract: Collecting and distributing information related to recent content publication activity of an instant messaging (IM) user provides other users in a network with timely, relevant information about people known to the user or within the same social network. A user participating in a social network can quickly and efficiently perceive new information related to other users (referred to as co-users) in a social network by reviewing the co-users' recent content publication activity. A user may be made able to do so without requiring the co-user to send a communication directly to the user regarding the new facts or new content, and also without requiring the user to actively browse or request information about the co-user.

Abstract translation: 收集和分发与即时消息（IM）用户的近期内容发布活动相关的信息，向网络中的其他用户提供关于用户已知或在相同社交网络内的人的及时，相关的信息。 参与社交网络的用户可以通过审查合作者最近的内容出版活动来快速有效地感知在社交网络中与其他用户（称为共同用户）相关的新信息。 可以使用户能够这样做，而不需要共同用户直接向用户发送关于新事实或新内容的通信，并且还可以不要求用户主动地浏览或请求关于共同用户的信息。

公开(公告)号：US09720914B2

公开(公告)日：2017-08-01

申请号：US14523404

申请日：2014-10-24

Applicant: MICROSOFT CORPORATION

Inventor： Rakesh Agrawal ,  Krishnaram Kenthapadi ,  Anitha Kannan ,  Sreenivas Gollapudi

IPC: G06F17/30

CPC classification number: G06F17/30014 ,  G06F17/30687

Abstract: Systems, methods, and computer storage media are provided for generating rich navigational study aids for electronic books. For a particular section of interest in a document, one or more related sections for providing additional context to the particular section are determined. The related sections are ranked based on a score indicating significance to the particular section. Based on a user's information processing preference, a set of ranked navigational links to each related section is presented to the user for additional context related to the particular section.

公开(公告)号：US20140359485A1

公开(公告)日：2014-12-04

申请号：US14461587

申请日：2014-08-18

Applicant: Microsoft Corporation

Inventor： Christopher Armstrong ,  Alan Keister ,  Lark Dunham ,  Jennifer Granito ,  Steven Greenberg ,  Rakesh Agrawal ,  Michael Sommers ,  Carol A. Glover

IPC: H04L29/06 ,  G06F3/0482

CPC classification number: H04L67/306 ,  G06F3/0482 ,  G06F16/9535 ,  G06F16/958 ,  G06Q10/107 ,  G06Q50/01 ,  H04L51/04 ,  H04L51/32 ,  H04L65/403 ,  H04L67/02 ,  H04L67/26

Abstract: Collecting and distributing information related to recent content publication activity of an instant messaging (IM) user provides other users in a network with timely, relevant information about people known to the user or within the same social network. A user participating in a social network can quickly and efficiently perceive new information related to other users (referred to as co-users) in a social network by reviewing the co-users' recent content publication activity. A user may be made able to do so without requiring the co-user to send a communication directly to the user regarding the new facts or new content, and also without requiring the user to actively browse or request information about the co-user.

Abstract translation: 收集和分发与即时消息（IM）用户的近期内容发布活动相关的信息，向网络中的其他用户提供关于用户已知或在相同社交网络内的人的及时，相关的信息。 参与社交网络的用户可以通过审查合作者最近的内容出版活动来快速有效地感知在社交网络中与其他用户（称为共同用户）相关的新信息。 可以使用户能够这样做，而不需要共同用户直接向用户发送关于新事实或新内容的通信，并且还可以不要求用户主动地浏览或请求关于共同用户的信息。

公开(公告)号：US09384233B2

公开(公告)日：2016-07-05

申请号：US13693040

申请日：2012-12-04

Applicant: Microsoft Corporation

Inventor： Ariel Fuxman ,  Hoa Nguyen ,  Juliana Freire de Lima e Silva ,  Stelios Paparizos ,  Rakesh Agrawal ,  Zhimin Chen ,  Lawrence William Colagiovanni ,  Prakash Sikchi

IPC: G06F17/30 ,  G06Q30/02 ,  G06Q30/06

CPC classification number: G06F17/30386 ,  G06Q30/0281 ,  G06Q30/0603

Abstract: Methods and systems for automatically synthesizing product information from multiple data sources into an on-line catalog are disclosed, and in particular, for automatically synthesizing the product information based on attribute-value pairs. Information for a product may be obtained, via entity extraction, feed ingestion, and other mechanisms, from a plurality of structured and unstructured data sources having different taxonomies and schemas. Product information may additionally or alternatively be obtained or derived based on popularity data. The product information may be cleansed, segmented and normalized. The product information may be clustered so closest products, attribute names and attribute values are associated. A representative value for an attribute name may be determined, and the on-line catalog may be updated so that entries are comprehensive, meaningful and useful to a catalog user. Updates from at least 500 million different data sources may be scheduled to occur as frequently as several times daily.

公开(公告)号：US20140324982A1

公开(公告)日：2014-10-30

申请号：US13872175

申请日：2013-04-29

Applicant: MICROSOFT CORPORATION

Inventor： Rakesh Agrawal ,  James A. Cook ,  Krishnaram Kenthapadi ,  Nina Mishra

IPC: H04L29/06

CPC classification number: H04L65/403 ,  H04L12/1831 ,  H04L51/04 ,  H04L51/16

Abstract: Text messages over some period of time are collected. Topic identifiers, such as hashtags, are extracted from the text messages. The text messages associated with each topic identifier are processed to identify which topic identifiers are associated with group chats based on information associated with the text messages such as the times when the text messages were generated and whether the text messages identify user accounts. The topic identifiers that are determined to be associated with the group chats are incorporated into applications that allow users to search for group chats, and to view text messages from past group chats.

Abstract translation: 收集一段时间内的短信。 从文本消息中提取主题标识符（如主题标签）。 处理与每个主题标识符相关联的文本消息，以基于与文本消息相关联的信息（例如，生成文本消息的时间）以及文本消息是否识别用户帐户来识别哪些主题标识符与组聊天相关联。 被确定为与组聊天相关联的主题标识符被合并到允许用户搜索组聊天并且从过去的组聊天中查看文本消息的应用中。

公开(公告)号：US20130290431A1

公开(公告)日：2013-10-31

申请号：US13929917

申请日：2013-06-28

Applicant: Microsoft Corporation

Inventor： Christopher Armstrong ,  Alan Keister ,  Lark Dunham ,  Jennifer Granito ,  Steven Greenberg ,  Rakesh Agrawal ,  Michael Sommers ,  Carol A. Glover

IPC: H04L29/08

CPC classification number: H04L67/306 ,  G06F3/0482 ,  G06F17/30867 ,  G06F17/3089 ,  G06Q10/107 ,  G06Q50/01 ,  H04L51/04 ,  H04L51/32 ,  H04L65/403 ,  H04L67/02 ,  H04L67/26

Abstract: Collecting and distributing information related to recent content publication activity of an instant messaging (IM) user provides other users in a network with timely, relevant information about people known to the user or within the same social network. A user participating in a social network can quickly and efficiently perceive new information related to other users (referred to as co-users) in a social network by reviewing the co-users' recent content publication activity. A user may be made able to do so without requiring the co-user to send a communication directly to the user regarding the new facts or new content, and also without requiring the user to actively browse or request information about the co-user.

Abstract translation: 收集和分发与即时消息（IM）用户的近期内容发布活动相关的信息，向网络中的其他用户提供关于用户已知或在相同社交网络内的人的及时，相关的信息。 参与社交网络的用户可以通过审查合作者最近的内容出版活动来快速有效地感知在社交网络中与其他用户（称为共同用户）相关的新信息。 可以使用户能够这样做，而不需要共同用户直接向用户发送关于新事实或新内容的通信，并且还可以不要求用户主动地浏览或请求关于共同用户的信息。

公开(公告)号：US20130091159A1

公开(公告)日：2013-04-11

申请号：US13693040

申请日：2012-12-04

Applicant: MICROSOFT CORPORATION

Inventor： Ariel Fuxman ,  Hoa Nguyen ,  Juliana Freire de Lima e Silva ,  Stelios Paparizos ,  Rakesh Agrawal ,  Zhimin Chen ,  Lawrence William Colagiovanni ,  Prakash Sikchi

IPC: G06F17/30

CPC classification number: G06F17/30386 ,  G06Q30/0281 ,  G06Q30/0603

Abstract: Methods and systems for automatically synthesizing product information from multiple data sources into an on-line catalog are disclosed, and in particular, for automatically synthesizing the product information based on attribute-value pairs. Information for a product may be obtained, via entity extraction, feed ingestion, and other mechanisms, from a plurality of structured and unstructured data sources having different taxonomies and schemas. Product information may additionally or alternatively be obtained or derived based on popularity data. The product information may be cleansed, segmented and normalized. The product information may be clustered so closest products, attribute names and attribute values are associated. A representative value for an attribute name may be determined, and the on-line catalog may be updated so that entries are comprehensive, meaningful and useful to a catalog user. Updates from at least 500 million different data sources may be scheduled to occur as frequently as several times daily.

Abstract translation: 公开了用于将产品信息从多个数据源自动合成到在线目录中的方法和系统，特别地，用于基于属性值对自动合成产品信息。 可以通过实体提取，饲料摄取和其他机制从具有不同分类和模式的多个结构化和非结构化数据源获得信息。 产品信息可以另外地或替代地基于流行度数据获得或导出。 产品信息可以被清洁，分段和归一化。 产品信息可能被聚集，因此最接近的产品，属性名称和属性值相关联。 可以确定属性名称的代表值，并且可以更新在线目录，使得条目对目录用户是全面的，有意义的和有用的。 可能会安排从至少5亿个不同数据源进行更新，频繁发生，每天多次。