International Business Machines Corporation

Microsoft Technology Licensing, LLC

Shixia Liu

Nan Cao

Wei Hong Qian

Li Ni Zhang

Li Ping Yang

Michelle Xue Zhou

Shimei Pan

Wei Cai

Li Tan

Yangqiu Song

Fei Wang

Zhong Su

Qiang Zhang

Yang Song

Lei Shi

Guo Tong Xie

Zhuo Zhang

Hong Fei Zhou

Yue Pan

Martin Miles Wattenberg

Jimeng Sun

Hao Lu

Zhao Ming Qiu

Shih-Gong Li

Zhongyuan Wang

Xi Jun Ma

Haixun Wang

Hongsong Li

Xiao Xiao Lian

A system, method or computer readable storage device to derive a taxonomy from keywords is described herein. A domain-dependent taxonomy from a set of keywords may be automatically derived by leveraging both a general knowledgebase and keyword search. For example, concepts may be deduced with the technique of conceptualization, and context information may be extracted from a search engine. Then, the taxonomy may be constructed using a tree algorithm.

Automatic taxonomy construction from keywords

Visualizing content change of a data collection over time. A topic may be split into multiple linear, non-overlapping sub-topics along a timeline by satisfying a diverse set of semantic, temporal, and visualization constraints simultaneously. For each derived sub-topic, a set of representative keywords may be automatically selected to summarize the main content of the sub-topic.

Temporal topic segmentation and keyword selection for text visualization

The invention provides a method and system for visualization of a data set, the method comprises: dividing the data set into a plurality of information layers based on different information dimensions; and visually processing the plurality of information layers based on different information dimensions, respectively, in order to present respective views of the plurality of information layers. In the present invention, by visualizing the data set through presenting different overviews of the data set from different information dimensions, respectively, the presentation of comprehensive information of the data set to a data set analyst is ensured while distortion of presented contents as well as visual clutter are prevented.

Method and system for visualization of data set

A method, apparatus and system for linking documents, the method comprising the steps of: providing a plurality of clusters in an enterprise intranet, each cluster consists of one or more documents; building a cluster page for each cluster to present the documents in the cluster; and building links between the cluster pages, between the documents, and between the cluster page and the document, based on analysis of the contents of the clusters and the documents. The present invention is useful for building the links between separate documents and may apply a link analyzing algorithm to the search for these documents to implement better search performance within the enterprise intranet.

Method, apparatus and system for linking documents

A layout method in a display area for multiple graph components in a dynamic network. The multiple graph components are sorted according to their importance into a first subset Sand a second subset Saccording to the order of their importance. The first subset Sis divided into an upper subset Cincluding only the most important graph component and a lower subset C. The display area is divided into display portions S' and S&#x2032; proportionally according to the importance values of the first subset Sand the second subset S. Display portion S&#x2032; is divided into display portions C&#x2032; and C&#x2032; proportionally according to the importance values of the upper subset Cand the lower subset C. The first division and the second division and the corresponding division of the display area are executed iteratively until the aspect ratio of display portion C&#x2032; is close to 1.

Layout method and system in a display area for disconnected dynamic networks

The invention discloses a method for displaying a weighted tree based on hyperbolic geometry, comprising: obtaining weighted tree data defining a weighted tree structure; positioning the root node in a negatively curved space as a center, based on the weighted tree defining data; based on the weighted tree defining data, performing layout processing for each parent node and all its direct child nodes from top to bottom recursively, until all the nodes in the tree have been laid out in the negatively curved space, wherein the higher the weight of an edge connecting the parent node and each of its direct child nodes is, the shorter the distance between the parent node and the child node is; mapping the layout data from the negative curved space to an Euclidean plane; and displaying the mapped data in display.

Method and computer system for displaying a weighted tree based on hyperbolic geometry

A method and an apparatus for identifying sub-graphs similar to a sample sub-graph from a network, in particular for identifying similar community structures from a network, are disclosed. Specifically, when a sample sub-graph is a community structure, it usually contains a plurality of nodes densely associated with each other. The method proposes first screening the network via key nodes of community structures so as to find sub-graph candidates that are potentially similar to the sample sub-graph, then strictly comparing the sub-graph candidates with the sample sub-graph to find similar sub-graphs of the sample sub-graph.

Method and apparatus for identifying similar sub-graphs in a network

A method and an apparatus for animating transition among a dynamic graph series including a preceding graph and a succeeding graph which are temporally sequentially displayed. The method includes: selecting an unchanged node which is the nearest to a removed node and merging through animation the removed node with the unchanged node, thereby animating removal of the removed node; and selecting an unchanged node which is the nearest to an added node and splitting through animation the added node from the unchanged node, thereby animating addition of the added node. During the animation, the moving distances of the nodes can be shortened, and the crossing of movement paths and the crossing of graph edges can be prevented during the movement of the nodes, thereby visualizing smoothly and clearly transition among dynamic graph series.

Method and apparatus for animating transition among dynamic graph series

A method and system of producing a visual text summarization. According to the method, a plurality of topics may be extracted from a collection of text documents, where each of the topics comprises a distribution of topic keywords. An importance ranking for each of the topics may be determined, and an importance ranking for each of the topic keywords of a topic may also be determined. A graph may be displayed having a plurality of stacked layers representing a group of topics selected from the plurality of topics based on the importance ranking of the topics. A keyword cloud within each layer of the graph may also be displayed, where the one keyword cloud is a group of topic keywords selected from the extracted topic keywords based on the importance ranking of the extracted topic keywords.

Producing a visual summarization of text documents

A method and computer system for constructing a link structure for T documents. An initial link structure Gbetween the T documents is generated. For each document d of D documents of the T documents, a loop of no more than M iterations is performed. In each iteration: a search engine is used to generate a ranking order of T' documents consisting of the T documents except document d; the initial link structure Gis modified according to the ranking order to generate a modified link structure G; if a condition is satisfied then the loop is exited and the modified link structure Gis outputted, otherwise Gis set equal to Gand a next iteration of the loop is performed. The condition is that a measure of a difference between Gand Gis less than a specified threshold or the M iterations were performed.

Method and apparatus for constructing a link structure between documents

The visibility of a treemap is improved by offsetting the area of a parent node relative to a bounding box containing all of its child nodes while constructing the treemap, so that a part of the bounding box lies outside the area of the parent node, and the area of each child node in the bounding box is located entirely or partially within the area of the parent node. The present invention highlights the parent-child relationships in the treemap with a cascading effect to make the structural information therein more apparent and easier to recognize.

Method and apparatus for improving the visibility of a treemap

A system and a method for automatically generating new concepts based on existing ontologies are provided. The system creates new concepts in existing ontologies based on new concept descriptions expressed in a natural language format, and comprises: a normalizer for receiving and parsing the new concept descriptions so as to transform them into normalized one and output them, wherein said normalized concept description includes one or more description part(s) having kernel concepts, said description parts can only contain the terms which can be identified in said existing ontologies, and each of said kernel concept contains a headword and zero or more property(s); and a new concept factory for, based on the normalized description of the new concept, identifying the kernel concepts in each normalized concept description part, and extracting the identified kernel concepts, related properties, and the relations among the kernel concepts for a user to create new concepts according to existing ontologies. Therefore, ontologies will be easier to be managed, reused and expanded, while keeping smaller size and less complexity of the ontologies.

System and method for generating new concepts based on existing ontologies

A visual method for enhancing search result navigation, comprising: obtaining a first search result from a search engine; clustering the first search result to get clustering information; calculating the correlations between the clustering information and the ranked list of the first search result, and performing visualization processing on the clustering information; and displaying the visual cluster hierarchy and the ranked list of the first search result in a joint manner based on the correlations. When a cluster is selected, further searching is performed and the search result is clustered again. Using the present invention, through combining a traditional ranked list of search results and the visual cluster hierarchy of these search results to display them in a joint manner, a convenient way is provided for the web user to find the potential correlations between the visual cluster hierarchy and the ranked list of the search results. Besides, through dynamically searching and clustering more search results again, the web user may get more detailed and more accurate search results easily.

Growing community of inventors

Beijing, China

Shixia Liu