-
公开(公告)号:US20150356107A1
公开(公告)日:2015-12-10
申请号:US14828685
申请日:2015-08-18
CPC分类号: G06F17/30109 , G06F17/2276 , G06F17/2705 , G06F17/30011 , G06F17/30286 , G06F17/30557 , G06F17/30566 , G06F17/30598 , G06F17/3071
摘要: An information management system creates data structures based entirely on the content of source files, then compares these data structures to discover synergies and commonalities. In one embodiment, the system accepts a first collection of source files, and extracts text from each source file. The text is compared to tags in one or more dictionaries, which comprise hierarchical listing of tags. Tags matching the text are associated with each source file. The system then generates a virtual relational network in which each source file having matching tags is a node. Tags associated with two or more source files are links between the nodes. This virtual relational network may be compared with another virtual relational network to discover common nodes or links. Source files later added to a collection are massively linked by associating all tags from all source files with the newly added source file, and vice versa.
摘要翻译: 信息管理系统完全基于源文件的内容创建数据结构,然后比较这些数据结构以发现协同效应和共同点。 在一个实施例中,系统接受源文件的第一集合,并从每个源文件中提取文本。 将文本与一个或多个字典中的标签进行比较,其中包括标签的分层列表。 与文本匹配的标签与每个源文件相关联。 然后,系统生成虚拟关系网络,其中具有匹配标签的每个源文件是一个节点。 与两个或多个源文件相关联的标签是节点之间的链接。 该虚拟关系网络可以与另一虚拟关系网络进行比较,以发现公共节点或链路。 通过将所有源文件中的所有标签与新添加的源文件相关联,后来添加到集合中的源文件将被大量链接,反之亦然。
-
公开(公告)号:US09575980B2
公开(公告)日:2017-02-21
申请号:US14828685
申请日:2015-08-18
CPC分类号: G06F17/30109 , G06F17/2276 , G06F17/2705 , G06F17/30011 , G06F17/30286 , G06F17/30557 , G06F17/30566 , G06F17/30598 , G06F17/3071
摘要: An information management system creates data structures based entirely on the content of source files, then compares these data structures to discover synergies and commonalities. In one embodiment, the system accepts a first collection of source files, and extracts text from each source file. The text is compared to tags in one or more dictionaries, which comprise hierarchical listing of tags. Tags matching the text are associated with each source file. The system then generates a virtual relational network in which each source file having matching tags is a node. Tags associated with two or more source files are links between the nodes. This virtual relational network may be compared with another virtual relational network to discover common nodes or links. Source files later added to a collection are massively linked by associating all tags from all source files with the newly added source file, and vice versa.
摘要翻译: 信息管理系统完全基于源文件的内容创建数据结构,然后比较这些数据结构以发现协同效应和共同点。 在一个实施例中,系统接受源文件的第一集合,并从每个源文件中提取文本。 将文本与一个或多个字典中的标签进行比较,其中包括标签的分层列表。 与文本匹配的标签与每个源文件相关联。 然后,系统生成虚拟关系网络,其中具有匹配标签的每个源文件是一个节点。 与两个或多个源文件相关联的标签是节点之间的链接。 该虚拟关系网络可以与另一虚拟关系网络进行比较,以发现公共节点或链路。 通过将所有源文件中的所有标签与新添加的源文件相关联,后来添加到集合中的源文件将被大量链接,反之亦然。
-