摘要:
Most recent research of scalable inductive learning on very large streaming dataset focuses on eliminating memory constraints and reducing the number of sequential data scans. However, state-of-the-art algorithms still require multiple scans over the data set and use sophisticated control mechanisms and data structures. There is discussed herein a general inductive learning framework that scans the dataset exactly once. Then, there is proposed an extension based on Hoeffding's inequality that scans the dataset less than once. The proposed frameworks are applicable to a wide range of inductive learners.
摘要:
The invention provides methods for rapidly synthesizing heteroatom containing zeolites including Sn-Beta, Si-Beta, Ti-Beta, Zr-Beta and Fe-Beta. The methods for synthesizing heteroatom zeolites include using well-crystalline zeolite crystals as seeds and using a fluoride-free, caustic medium in a seeded dry-gel conversion method. The Beta zeolite catalysts made by the methods of the invention catalyze both isomerization and dehydration reactions.
摘要:
In the field of a liquid crystal on silicon (LCoS) display, a charge sharing system and a charge sharing method of an LCoS display are provided. The system includes: a column driving circuit, a row driving circuit, a pixel matrix, a control circuit, a gamma reference voltage circuit, and a first switching module. The column driving circuit includes a second switching module, a buffer, a shift register, a latch, and a digital-to-analog (D/A) converter. The second switching module is serially connected between the D/A converter and the buffer. The control circuit simultaneously outputs a signal for controlling turn-off of the buffer, a signal for controlling turn-off of a switch in the second switching module, and a signal for controlling turn-off of the gamma reference voltage circuit. The control circuit outputs a signal for controlling turn-on of a switch in the first switching module. The control circuit outputs a signal for controlling the row driving circuit. The control circuit simultaneously outputs a signal for controlling turn-on of the buffer, a signal for controlling turn-on of the switch in the second switching module, and a signal for controlling turn-on of the gamma reference voltage circuit.
摘要:
A heating apparatus for regulating/controlling the surface temperature of a substrate is provided. At least a thermal pyrolytic graphite (TPG) layer is embedded in the heater to diffuse the temperature difference of the various components in the heating apparatus and provide temporal and spatial control of the surface temperature of the substrate, for a relatively uniform substrate temperature with the difference between the maximum and minimum temperature points on the substrate of less than 10° C.
摘要:
A self-powered in-pipe fluid meter to be mounted inside of a pipe carrying a fluid therein. The fluid meter comprises at least one sensing unit capable of measuring one or more parameters of the fluid inside of the pipe; a telemetric data transmission unit capable of telemetrically transmitting data including a measured fluid parameter to a host terminal and/or another fluid meter; and at least one fluid-driven power source unit capable of generating power from the fluid flow within the pipe and supplying power to the sensing unit and/or the transmission unit.
摘要:
Techniques for classifying structural data with skewed distribution are disclosed. By way of example, a method classifying structural input data comprises a computer system performing the following steps. Multiple classifiers are constructed, wherein each classifier is constructed on a subset of training data, using one or more selected composite features from the subset of training data. A consensus among the multiple classifiers is computed in accordance with a voting scheme such that at least a portion of the structural input data is assigned to a particular class in accordance with the computed consensus. Such techniques for structured data classification are capable of handling skewed class distribution and partial feature coverage issues.
摘要:
A commodity product distribution plan is used to instruct source sites as to how commodity products are to be distributed among target sites. Where the commodity products are to be sold at auction, a wide range of auction prices can be expected due to mixed models, model years, commodity attributes such as color or optional features, economic conditions, and the auction site location itself. Additional factors that contribute to realized auction prices include depreciation and interest rate costs as well as constraints on shipments and auction site capacities. The present invention provides forecast auction prices for the commodity products, taking these various factors into consideration. In this way, an optimized distribution planned aimed at maximizing the potential profit for the commodity products to be sold at auction is generated.
摘要:
A general framework for mining concept-drifting data streams using weighted ensemble classifiers. An ensemble of classification models, such as C4.5, RIPPER, naive Bayesian, etc., is trained from sequential chunks of the data stream. The classifiers in the ensemble are judiciously weighted based on their expected classification accuracy on the test data under the time-evolving environment. Thus, the ensemble approach improves both the efficiency in learning the model and the accuracy in performing classification. An empirical study shows that the proposed methods have substantial advantage over single-classifier approaches in prediction accuracy, and the ensemble framework is effective for a variety of classification models.
摘要:
In connection with the mining of time-evolving data streams, a general framework that mines changes and reconstructs models from a data stream with unlabeled instances or a limited number of labeled instances. In particular, there are defined herein statistical profiling methods that extend a classification tree in order to guess the percentage of drifts in the data stream without any labelled data. Exact error can be estimated by actively sampling a small number of true labels. If the estimated error is significantly higher than empirical expectations, there preferably re-sampled a small number of true labels to reconstruct the decision tree from the leaf node level.
摘要:
A system for the real-time and in-situ macro and micro measurement of in-plane deformations of a microelectronic package or the like comprises a closed environmental chamber (3) within which a test sample may be subjected to thermal cycle loading and/or humidity loading, an incoherent white light source (6) for illuminating the sample, a long-working-distance microscope (2) and image acquisition means (7) for capturing speckle patterns from the surface of the sample during loading, and a control (8) for automating the co-ordination of the various components and for analysing the speckle images using digital image speckle correlation.