公开(公告)号：US08700673B2

公开(公告)日：2014-04-15

申请号：US13398792

申请日：2012-02-16

申请人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

发明人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

IPC分类号： G06F7/00

CPC分类号： G06F17/30404

摘要： Metadata search is enhanced by utilizing relationship data indicating relationships between metadata items. A server generates an index mapping metadata items to terms associated with the metadata items and a graph describing relationships between each of the metadata items. When the server receives a search request, the server locates a candidate set of the metadata items based on the search term(s) and the index. The server performs a link analysis of the graph to determine a relationship score for each metadata item. For each particular metadata item in the candidate set of the metadata items, the server calculates a ranking score based at least on the relationship score for the particular metadata item. The server generates a ranked result set based on comparing the ranking scores for the candidate set of metadata items. The server then provides information indicating the ranked result set in response to the search request.

摘要翻译： 通过利用表示元数据项之间关系的关系数据来增强元数据搜索。 服务器生成将元数据项映射到与元数据项相关联的术语的索引，以及描述每个元数据项之间的关系的图。 当服务器接收到搜索请求时，服务器根据搜索项和索引来定位元数据项的候选集。 服务器执行图形的链接分析，以确定每个元数据项目的关系得分。 对于元数据项的候选集合中的每个特定元数据项，服务器至少基于特定元数据项的关系得分来计算排名得分。 服务器基于比较候选元数据集合的排名分数来生成排名结果集。 然后，服务器响应于搜索请求提供指示排名结果集的信息。

公开(公告)号：US08682932B2

公开(公告)日：2014-03-25

申请号：US13398794

申请日：2012-02-16

申请人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

发明人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

IPC分类号： G06F7/00

CPC分类号： G06F17/30651 ,  G06F17/30112 ,  G06F17/30958

摘要： Highly relevant search results for unstructured queries are efficiently identified from normalized data. A server graphs relationships between each of the data objects. For each search term, the server identifies a candidate set of data objects mapped to the term. The server calculates priority scores for the candidate data objects based at least in part on one or more of: a link analysis of the graph; or an analysis of metadata describing structural constraints upon the candidate data objects. Based on the graph, the server identifies one or more search result subgraphs, each comprising at least one data object from each of the candidate sets. The server looks for subgraphs in an order that is based on the priority scores for the candidate data objects. The server may terminate its search early, in part because prioritizing the candidate data objects increases the likelihood of receiving relevant search results first.

摘要翻译： 从规范化数据有效地识别非结构化查询的高度相关的搜索结果。 服务器显示每个数据对象之间的关系。 对于每个搜索项，服务器识别映射到该术语的候选数据对象集合。 服务器至少部分地基于图形的链接分析中的一个或多个来计算候选数据对象的优先级分数; 或描述对候选数据对象的结构约束的元数据的分析。 基于该图，服务器识别一个或多个搜索结果子图，每个包括来自每个候选集的至少一个数据对象。 服务器按照候选数据对象的优先级得分的顺序查找子图。 服务器可以提前终止其搜索，部分原因是优先级候选数据对象增加了首先接收相关搜索结果的可能性。

公开(公告)号：US20130218899A1

公开(公告)日：2013-08-22

申请号：US13398794

申请日：2012-02-16

申请人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

发明人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

IPC分类号： G06F17/30

CPC分类号： G06F17/30651 ,  G06F17/30112 ,  G06F17/30958

摘要： Highly relevant search results for unstructured queries are efficiently identified from normalized data. A server graphs relationships between each of the data objects. For each search term, the server identifies a candidate set of data objects mapped to the term. The server calculates priority scores for the candidate data objects based at least in part on one or more of: a link analysis of the graph; or an analysis of metadata describing structural constraints upon the candidate data objects. Based on the graph, the server identifies one or more search result subgraphs, each comprising at least one data object from each of the candidate sets. The server looks for subgraphs in an order that is based on the priority scores for the candidate data objects. The server may terminate its search early, in part because prioritizing the candidate data objects increases the likelihood of receiving relevant search results first.

摘要翻译： 从规范化数据有效地识别非结构化查询的高度相关的搜索结果。 服务器显示每个数据对象之间的关系。 对于每个搜索项，服务器识别映射到该术语的候选数据对象集合。 服务器至少部分地基于图形的链接分析中的一个或多个来计算候选数据对象的优先级分数; 或描述对候选数据对象的结构约束的元数据的分析。 基于该图，服务器识别一个或多个搜索结果子图，每个包括来自每个候选集的至少一个数据对象。 服务器按照候选数据对象的优先级得分的顺序查找子图。 服务器可以提前终止其搜索，部分原因是优先级候选数据对象增加了首先接收相关搜索结果的可能性。

公开(公告)号：US20130218898A1

公开(公告)日：2013-08-22

申请号：US13398792

申请日：2012-02-16

申请人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

发明人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

IPC分类号： G06F17/30

CPC分类号： G06F17/30404

摘要： Metadata search is enhanced by utilizing relationship data indicating relationships between metadata items. A server generates an index mapping metadata items to terms associated with the metadata items and a graph describing relationships between each of the metadata items. When the server receives a search request, the server locates a candidate set of the metadata items based on the search term(s) and the index. The server performs a link analysis of the graph to determine a relationship score for each metadata item. For each particular metadata item in the candidate set of the metadata items, the server calculates a ranking score based at least on the relationship score for the particular metadata item. The server generates a ranked result set based on comparing the ranking scores for the candidate set of metadata items. The server then provides information indicating the ranked result set in response to the search request.

摘要翻译： 通过利用表示元数据项之间关系的关系数据来增强元数据搜索。 服务器生成将元数据项映射到与元数据项相关联的术语的索引，以及描述每个元数据项之间的关系的图。 当服务器接收到搜索请求时，服务器根据搜索项和索引来定位元数据项的候选集。 服务器执行图形的链接分析，以确定每个元数据项目的关系得分。 对于元数据项的候选集合中的每个特定元数据项，服务器至少基于特定元数据项的关系得分来计算排名得分。 服务器基于比较候选元数据集合的排名分数来生成排名结果集。 然后，服务器响应于搜索请求提供指示排名结果集的信息。

公开(公告)号：US08484208B1

公开(公告)日：2013-07-09

申请号：US13398799

申请日：2012-02-16

申请人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

发明人： Nikhil Raghavan ,  Ravi Murthy ,  Aman Naimat

IPC分类号： G06F17/30 ,  G06F7/00

CPC分类号： G06F17/30554 ,  G06F17/30941

摘要： Techniques are disclosed for presenting semi-structured sets of search results comprising two or more differently-structured subsets of search results. The search results are divided into two or more groups of similarly-structured results. The search results are displayed in their respective groups rather than as a single set. Each group is displayed using a different display structure, in an order determined by a group ranking mechanism. The search results within a group ordered by a result ranking mechanism. Techniques are also disclosed for enhancing a display of similarly structured data by emphasizing highly relevant result fields. The highly relevant result fields may be identified based on metadata ranking mechanisms, uniqueness of their constituent values, historical feedback, keyword location, and/or other mechanisms. The fields are emphasized using, without limitation, highlighting, reordering, and filtering of unemphasized fields from the display.