60results about How to "Efficient and accurate generation" patented technology

The invention provides a method and device for generating a clustering model and carrying out clustering based on the clustering model. The clustering model generating method comprises the following steps: in response to receiving a plurality of historical messages from a historical speech dialogue, extracting feature information of each historical message of the plurality of historical messages; obtaining association relationship between the plurality of historical messages; and generating the clustering model for carrying out clustering on the plurality of historical messages according to the feature information of each historical message of the plurality of historical messages and the association relationship. The clustering method comprises the following steps: in response to receiving a plurality of current messages in a current dialogue, extracting feature information of each current message of the plurality of current messages; and clustering the plurality of current messages to at least one theme group by adopting the clustering model generated through the method above and on the basis of the feature information of each current message of the plurality of current messages. According to the method and device for generating the clustering model and carrying out clustering based on the clustering model, the reliable and accurate clustering model can be generated, and the plurality of messages can be clustered and displayed based on the clustering model.

The present invention relates to a method and device for generating an ontology instance that classifies documents into structured documents and unstructured documents and automatically generates ontology instances. The method includes collecting documents corresponding to classes of an ontology from Web; if the collected documents are unstructured documents, extracting inter-entity relationship information from the unstructured documents; if the collected documents are structured documents, extracting inter-entity relationship information from the structured documents; generating ontology instances from the extracted inter-entity relationship information, and mapping the generated ontology instances to corresponding classes of the ontology.

Systems and methods for generating sound, detecting sound, and generating and detecting sound are provided. An array of audio transducers can be provided whereby each audio transducer in the array can be optimized for a narrow range of sound frequencies. When operating at or close to its resonant frequency, a transducer can generate (and / or detect) sound with a higher efficiency and less distortion as compared to other frequencies. Accordingly, sound may be divided into component signals such that each transducer is only responsible for generating (and / or detection) sound close to its resonant frequency. This sound reproduction (and / or detection) technique can increase efficiency, and therefore, can increase the total output volume that an array can generate using a given amount of input power when generating sound (and / or increase the total output power that an array can generate using a given amount of input volume when detecting sound).

A Doppler sensitivity measuring unit measures the Doppler sensitivity of Doppler spectrum data obtained by ultrasonic wave transmission / reception with respect to a predetermined region of an object to be examined. A spectrum shape model generating unit generates a spectrum shape model on the basis of a window function used when the above Doppler spectrum data is generated. A threshold setting unit sets a threshold range in which trace waveform data displaces by a predetermined amount in the frequency axis direction on the basis of the spectrum shape model and the Doppler sensitivity, and sets a predetermined number of thresholds at almost equal intervals in the threshold range. A trace data generating unit selects a threshold for the acquisition of desired trace waveform data by generating trace waveform data while sequentially updating the plurality of set thresholds.

An automatic programming apparatus 1 comprises: a tool path data generating section 12 which generates movement path data for a semifinishing tool by analyzing an NC program; a work shape data generating section 16 which computes areas over which blank is to be machined on the basis of the movement path data, and which generates shape data of a semifinished workpiece by deleting the machined areas from the shape of the blank and also generates attribute data indicating that the faces newly generated by the deletion processing are machined faces; a CL data generating section 20 which generates CL data by setting the edges of the machined faces as machining portions on the basis of the shape data of the semifinished workpiece and attribute data; and an NC program generating section 23 which converts the CL data into the NC program for performing the chamfering.

The present invention relates to a method and device for generating an ontology instance that classifies documents into structured documents and unstructured documents and automatically generates ontology instances. The method includes collecting documents corresponding to classes of an ontology from Web; if the collected documents are unstructured documents, extracting inter-entity relationship information from the unstructured documents; if the collected documents are structured documents, extracting inter-entity relationship information from the structured documents; generating ontology instances from the extracted inter-entity relationship information, and mapping the generated ontology instances to corresponding classes of the ontology.

Systems and methods for training semantic segmentation. Embodiments of the present invention include predicting semantic labeling of each pixel in each of at least one training image using a semantic segmentation model. Further included is predicting semantic boundaries at boundary pixels of objects in the at least one training image using a semantic boundary model concurrently with predicting the semantic labeling. Also included is propagating sparse labels to every pixel in the at least one training image using the predicted semantic boundaries. Additionally, the embodiments include optimizing a loss function according the predicted semantic labeling and the propagated sparse labels to concurrently train the semantic segmentation model and the semantic boundary model to accurately and efficiently generate a learned semantic segmentation model from sparsely annotated training images.