362results about How to "Accurate resolution" patented technology

Entity disambiguation resolves which names, words, or phrases in text correspond to distinct persons, organizations, locations, or other entities in the context of an entire corpus. The invention is based largely on language-independent algorithms. Thus, it is applicable not only to unstructured text from arbitrary human languages, but also to semi-structured data, such as citation databases and the disambiguation of named entities mentioned in wire transfer transaction records for the purpose of detecting money-laundering activity. The system uses multiple types of context as evidence for determining whether two mentions correspond to the same entity and it automatically learns the weight of evidence of each context item via corpus statistics. The invention uses multiple search keys to efficiently find pairs of mentions that correspond to the same entity, while skipping billions of unnecessary comparisons, yielding a system with very high throughput that can be applied to truly massive data.

A hyperspectral imager that achieves accurate spectral and spatial resolution by using a micro-lens array as a series of field lenses, with each lens distributing a point in the image scene received through an objective lens across an area of a detector array forming a hyperspectral detector super-pixel. Each sub-pixel within a super-pixel has a bandpass or other type filter to pass a different band of the image spectrum. The micro-lens spatially corrects the focused image point to project the same image scene point onto all sub-pixels within a super-pixel. A color separator can be used to split the image into sub-bands, with each sub-band image projected onto a different spatially corrected detector array. A shaped limiting aperture can be used to isolate the image scene point within each super-pixels and minimize energy coupling to adjacent super-pixels.

A communication system is disclosed for providing dedicated bandwidth to at least one subscriber location for transmitting to a common point of distribution via an HFC network. In an embodiment of the invention, the communication system includes a channel interface module and at least one gateway coupled across the HFC network. The channel interface module is located at the point of distribution and includes a transmitter that transmits a windowing signal via the HFC network. The gateway is located at a subscriber location and includes a processor that encapsulates subscriber data into data cells suitable for burst transmission, receive logic that receives the windowing signal, timing logic that indicates burst transmission times only at programmed time slots within each of repeating transmission windows based on the windowing signal and a predetermined transmission timing offset, and a burst transmitter that burst transmits subscriber data cells in a predetermined upstream frequency channel when indicated by the timing logic.

The present disclosure describes techniques for entity classification and data enrichment of data sets. A data enrichment system is disclosed that can extract, repair, and enrich datasets, resulting in more precise entity resolution and classification for purposes of subsequent indexing and clustering. Disclosed techniques may include performing entity recognition to identify segments of interest that relate to an entity. Related data may be analyzed for classification, which can be used to transform the data for enrichment to its users.

A method is provided for representing and updating the state of a dialog involving a series of queries and commands to an artificial intelligence system. Each statement within the dialogue may be modeled as a relational tree spanning nodes corresponding to named entities within the statement. A data structure may be used to store each of these trees and to modify them as the dialog progresses. A subsequent statement in the dialog may be parsed and its contents used to update an ongoing search initiated within that dialog. Statements may be used for the update process despite being fragmentary or not corresponding to any predetermined grammar. An algorithm is disclosed for updating the trees within the data structure after a new statement is parsed.

A method is provided for monitoring a network that includes providing a path selector that can select between a first input and a second input, where the first input is coupled to a customer premises equipment interface that is operable to receive signals from customer premises equipment and the second input is coupled to a test signal generator that is operable to generate test signals. The method also includes determining whether operating customer premises equipment is coupled to an input of the customer premises equipment interface that is coupled to the first input of the path selector. If not, then an idle state is entered into. While in the idle state, it is determined whether a signal is received from the customer premises equipment interface at the first input of the path selector. The test signal received at the second input of the path selector is selected unless a signal is received from the customer premises equipment interface at the first input. The method also includes communicating the test signal from the path selector to the network.

Software management methods and apparatus for performing software configuration management operations and for supporting build processes are described. Both implied and explicit dependency relationship information is captured and intelligently used in an automated fashion avoiding or reducing the need for human tracking of such relationships. Temporary or non-persistent data elements used during various phases of the software development process are automatically detected, including for example environment variables and their values and the versions of tools used to operate on the software components. A persistent association is created and stored between the normally temporary data element(s) and the software component(s) they affect. Implicit dependencies between data elements and software components are also automatically updated and revision-controlled, making historical and current implicit dependencies persistent. A software management information data base including implicit dependency relationship and explicit dependency relationship information is created. The information reflects explicit or implied relationships between various software components.

The present invention relates to a computer-implemented method of providing a landslide hazard assessment, including: inputting a region from which to retrieve geographic information from a database or from a user in real-time, using an input device of a computer system; retrieving geographic information from the database, on the region; retrieving satellite-based remote sensing data from the external database in real-time or near real-time from the region, the external database being accessed by the computer system over the internet; creating a susceptibility map which is divided into a plurality of categories that represent relative susceptibility to landslides; showing the plurality of categories on the susceptibility map on a display of the computer system, using a defining characteristic; comparing a current daily rainfall accumulation from the external database to a daily rainfall threshold; and issuing a moderate- or high-hazard level landslide nowcast, depending on a result of the comparison.

The invention provides a method of resolving a location in a second digital map from an ordered list of location reference points determined from a first digital map. The method involves identifying candidate lines and nodes in the second digital map, and using curvature, height and gradient information associated with the location reference points to identify the most likely candidate nodes or lines in the second digital map corresponding to the nodes represented by the location reference points and to lines emanating from or incident at the node in the first digital map. The method involves carrying out a route search between the most likely identified candidate node or line identified for one location reference point, and the corresponding node or line associated with the next reference point in the list, and repeating this step for consecutive pairs of reference points until the final location reference point is reached.

A method for monitoring fluid flow through a downhole device, comprises a) providing an acoustic tube wave in fluid in the device; b) measuring the acoustic tube wave after it has passed through the fluid in the device; and c) assessing the permeability of the device by measuring the attenuation of the acoustic signal. Changes in velocity of the acoustic signal may also be measured. The device may be a permeable downhole device such as a sand screen the measurements in step b) are made using a plurality of sensors deployed in the hole. The method may further including the step of cross-correlating a signal received at a first receiver with signals received at additional sensors so as to obtain an effective response as if the signal had been emitted from a source at the position of said first receiver.

The invention relates to a normal-temperature normal-pressure femto-second CARS (Coherent Anti-stokes Raman Spectroscopy) time-resolved spectrum measuring system which relates to the technical field of nonlinear optics and solves the problem of multiple limiting factors of traditional femto-second CARS time-resolved spectrum measuring experiment. Beams output by a femto-second laser are adjusted through a series of reflectors and optical delay lines by the system to form three bundles of beams which have approximate energy and are respectively positioned on three peaks of a square on the vertical direction of the beams, the beams are focused to a sample pool and then emit a new beam along a specific angle, i.e. a CARS signal, the CARS signal is filtered through a filter plate, then received by a probe and input to a monochromator, the data acquisition of an electrical signal converted by a photomultiplier is carried out by utilizing BOCCAR, and the data are input to a computer for data processing. The invention can carry out femto-second CARS time-resolved spectrum measurement under the experimental conditions of normal temperature and normal pressure and is applicable for the femto-second CARS spectrum measurement of gas samples and liquid samples in a static sample pool.

A method and system for bringing together online and offline advertising uses anonymous links that are associated with consumer data. The anonymous links allow processing without personally identifiable information (PII) in a secure environment. Data is matched using the anonymous links, and further using identifiers that are encrypted for use in connection with individual match distribution partners. The method and system allows a marketer to utilize offline data to precisely target advertisements without the use of PII, and to perform analytics concerning the use of the online advertisements to more precisely determine the effectiveness of multichannel marketing efforts.

The invention relates to a dwell-time-based moving object semantic behavior pattern mining method, which comprises the following steps of: (1) collecting semantic track data of moving objects, and storing semantic information of the moving objects; (2) for each moving object, mining all frequent semantic behavior patterns of each moving object; (3) designing a time-weight-based semantic behavior pattern similarity degree measurement method, and calculating the similarity degree between the frequent semantic behavior patterns; and (4) according to the similarity degree between the frequent semantic behavior patterns, using a pruning strategy for hierarchical clustering, and mining all moving object clustering with the similar semantic behavior patterns. The method has the advantage that high accuracy and high efficiency of the moving object semantic behavior pattern mining can be ensured. The final result can be provided for a user in relevant fields to be used, such as a friend recommending system, the track case cracking field and the individualized service field; the accurate finding on the similar semantic behavior patterns of mobile object groups can be supported; and the error rate is reduced.

The invention discloses a communication method of underwater digital voice and relates to a communication method. The communication method comprises the following steps of: at a communication transmitting end, sampling input simulated voice by using an analogue to digital (A/D) converter so as to obtain digital voice, performing mixed excitation linear prediction (MELP) low-speed compression coding on the digital voice so as to obtain voice data, grouping and performing Turbo coding on a voice data bit stream so as to obtain different groups of data, performing orthogonal frequency division multiplexing (OFDM) multicarrier modulation on every group of data and converting a modulated signal into a sound wave through an underwater acoustic transducer for transmitting in a sea underwater sound channel; and at a communication receiving end, performing underwater sound OFDM multicarrier modulation on a received signal after synchronization so as to obtain demodulation data, performing Turbo decoding on the demodulation data, performing MELP decoding on the data so as to synthesize the digital voice, converting the digital voice into simulated voice by using a digital to analogue (D/A) converter and outputting the simulated voice through a loudspeaker.