Abstract:
In accordance with an example embodiment, large scale category classification based on sequence semantic embedding and parallel learning is described. In one example, one or more closest matches are identified by comparison between (i) a publication semantic vector that corresponds to at least part of the publication, the publication semantic vector based on a first machine-learned model that projects the at least part of the publication into a semantic vector space, and (ii) a plurality of category vectors corresponding to respective categories from a plurality of categories.
Abstract:
Data having some similarities and some dissimilarities may be clustered or grouped according to the similarities and dissimilarities. The data may be clustered using agglomerative clustering techniques. The clusters may be used as suggestions for generating groups where a user may demonstrate certain criteria for grouping. The system may learn from the criteria and extrapolate the groupings to readily sort data into appropriate groups. The system may be easily refined as the user gains an understanding of the data.
Abstract:
A computer-implemented method is provided that is stored on computer readable non-transitory media. One or more data fields are accessed within a file. Accessed data field, are mapped mapping on a display computer system. The accessed one or more data fields are from one or more data sources that relate to situations from clustering messages received from managed infrastructure. The mapping being performed based on a input of the situation summaries using a graphical user interface. Displayed on the display computer system are one or more dashboards of situations relative to summaries from clustering messages received from managed infrastructure. The one or more dashboards include at least one of actions that a user can take relative to clustered messages.
Abstract:
An event clustering system includes an extraction engine in communication with an infrastnjcture. The extraction engine receives data from the infrastructure and produces events. An alert engine receives the events and creates alerts mapped into a matrix, M. A signalizer engine includes one or more of an NMF engine, a k-means clustering engine and a topology proximity engine. The signalizer engine determines one or more common steps from events and produces clusters relating to the alerts and or events.
Abstract:
A system for analyzing data to determine an activity around a product is provided. The system comprises a user interface configured to enable one or more data analysts to provide input data and an acquisition module coupled to user interface and configured to retrieve social media data in response to the input data. The social media data is received from one more social media platforms. The system further comprises processing circuitry coupled to the acquisition module and comprises an analysis module configured to analyze the social media data to generate processed data and classify the processed data based on a plurality of criteria and a visualization module coupled to the analysis module and configured to generate a plurality of visual representations of classified data.
Abstract:
Systems and methods for massive model visualization in product data management (PDM) systems comprises: - storing (505) a hierarchical product data structure that includes a plurality of occurrence nodes and component nodes; - creating (515) an occurrence equivalency table (410) from the hierarchical product data structure, that identifies at least one anchor occurrence node and at least one equivalent occurrence node, wherein product component corresponding to the equivalent occurrence node is spatially located within a specified distance threshold (δ) of product component corresponding to the anchor occurrence node; - creating (515) an anchor occurrence table corresponding to the hierarchical product data structure that lists a plurality of unique occurrence chain represented by the hierarchical product data structure, where each equivalent occurrence node is replaced by its corresponding anchor occurrence node, and that associates each listed unique occurrence chain with an associated cell index value.
Abstract:
A three-dimensional data displaying method includes: rendering a three-dimensional matrix image by arranging meta information representing data entries at coordinates composed of a horizontal axis component, a vertical axis component, and an azimuth component, based on arrangement information, with respect to structured data entries including the meta information and the arrangement information; and displaying the three-dimensional matrix on a screen. The meta information may include a topic word and a visual display, and the arrangement information may include a first classification attribute corresponding to the horizontal axis component, a second classification attribute corresponding to the vertical axis component, and a third classification attribute corresponding to the azimuth component.
Abstract:
A three-dimensional data displaying method includes: rendering a three-dimensional matrix image by arranging meta information representing data entries at coordinates composed of a horizontal axis component, a vertical axis component, and an azimuth component, based on arrangement information, with respect to structured data entries including the meta information and the arrangement information; and displaying the three-dimensional matrix on a screen. The meta information may include a topic word and a visual display, and the arrangement information may include a first classification attribute corresponding to the horizontal axis component, a second classification attribute corresponding to the vertical axis component, and a third classification attribute corresponding to the azimuth component.
Abstract:
A system (10) and method (50) for displaying relationships between concepts (14c, 14d) to provide classification suggestions via inclusion is provided. A set of reference concepts (14d) each associated with a classification code is designated. One or more of the reference concepts (14d) are combined with a set of uncoded concepts (14c). Clusters of the uncoded concepts (14c) and the one or more reference concepts (14d) are generated. Relationships between the uncoded concepts (14c) and the one or more reference concepts (14d) in at least one cluster (173) are visually depicted as suggestions for classifying the uncoded concepts (14c) in that cluster (173).