Abstract: Disclosed is an electromechanical transducer and the method for manufacturing the same, which can detect or control deformation and vibration of a structure and flow of fluids by applying controlling forces. The electromechanical transducer of the present invention comprises a base substrate to which initial stress is applied; an electro-active material layer attached on the base substrate; and electrodes installed on the top and bottom side of the electro-active material layer for actuating the electro-active material layer, the base substrate and the electro-active material layer which is deformed when initial stress is removed from the base substrate so that the base substrate and the electro-active material layer have curvatures.

Abstract: A method for using artificial neural networks as a universal computing device to model the relationship between the training inputs and corresponding outputs and to solve all problems with estimation, classification, and ranking tasks in their nature. Raw data related to problems is obtained and a subset of that data is processed and distilled for application to this universal computing device. The training data includes inputs and their corresponding results, which values could be continuous, categorical, or binary. The goal of this universal computing device is to solve problems by the universal approximation property of artificial neural networks. In this invention, a practical solution is created to resolve the issues of local minima and generalization, which have been the obstacles to the use of artificial neural networks for decades. This universal computing device uses an efficient and effective search algorithm, Retreat and Turn, to escape local minima and approach the best solutions.

Abstract: A resin composition consists of a resin mixture, a light diffusing agent and a fluorescent-whitening agent. The resin mixture consists of 50˜99.9 wt % of polystyrene (PS) and 0.1˜50 wt % of polycarbonate (PC). The light diffusing agent is 0.1˜20 wt % of the resin mixture. The fluorescent-whitening agent is 0.00001˜0.02 wt % of the resin mixture. Therefore, the present invention has low cost because of PS and also possesses high thermal resistance and high impact resistance because of PC. Resin plates made from the resin composition, such as diffusion plates, have improved thermal and impact resistances.

Abstract: The present invention relates, in general, to technology applied to entertainment information devices, such as video game machines or computer game programs, information devices for representing geo-spatial information, or information devices for performing three-dimensional rendering using computer graphics and, more particularly, to a protruded displacement mapping method, which represents an object as protruding from a surface. In the image-based protruded displacement mapping method of the present invention, an intersection point search step of searching for an intersection point of a surface of a protruding object, defined by the displacement information, and an observers sight line, and displaying the shape of the protruding object on the polygonal surface, is performed.

Abstract: A method for using machine learning to solve problems having either a “positive” result (the event occurred) or a “negative” result (the event did not occur), in which the probability of a positive result is very low and the consequences of the positive result are significant. Training data is obtained and a subset of that data is distilled for application to a machine learning system. The training data includes some records corresponding to the positive result, some nearest neighbors from the records corresponding to the negative result, and some other records corresponding to the negative result. The machine learning system uses a co-evolution approach to obtain a rule set for predicting results after a number of cycles. The machine system uses a fitness function derived for use with the type of problem, such as a fitness function based on the sensitivity and positive predictive value of the rules. The rules are validated using the entire set of training data.

Abstract: A method for using machine learning to solve problems having either a “positive” result (the event occurred) or a “negative” result (the event did not occur), in which the probability of a positive result is very low and the consequences of the positive result are significant. Training data is obtained and a subset of that data is distilled for application to a machine learning system. The training data includes some records corresponding to the positive result, some nearest neighbors from the records corresponding to the negative result, and some other records corresponding to the negative result. The machine learning system uses a co-evolution approach to obtain a rule set for predicting results after a number of cycles. The machine system uses a fitness function derived for use with the type of problem, such as a fitness function based on the sensitivity and positive predictive value of the rules. The rules are validated using the entire set of training data.

Abstract: The present invention relates, in general, to technology applied to entertainment information devices, such as video game machines or computer game programs, information devices for representing geo-spatial information, or information devices for performing three-dimensional rendering using computer graphics and, more particularly, to a protruded displacement mapping method, which represents an object as protruding from a surface. In the image-based protruded displacement mapping method of the present invention, an intersection point search step of searching for an intersection point of a surface of a protruding object, defined by the displacement information, and an observers sight line, and displaying the shape of the protruding object on the polygonal surface, is performed.

Abstract: There is provided a method and a device of monitoring oil oxidation in real time. The method of the present invention comprises the steps of: irradiating ultraviolet light into oil to be monitored; measuring fluorescence emission intensity of the oil in red, green and blue wavelength bands; determining one value measured in a relatively long wavelength band and the other value measured in a relatively short wavelength band among the fluorescence emission intensity measured in the red, green and blue wavelength bands; calculating a fluorescence emission ratio which is defined as a ratio of the value measured in the relatively long wavelength band to the value measured in the relatively short wavelength band; and monitoring a change in the fluorescence emission ratio. It is then determined whether the fluorescence emission ratio reaches a predetermined critical magnitude. When the fluorescence emission ratio reaches the critical magnitude, the necessity of replacing the oil with new one is indicated.

Abstract: A patch-type extrinsic Fabry-Perot interferometric fiber optic sensor and a real-time structural vibration monitoring method using the same are disclosed. The patch-type extrinsic Fabry-Perot interferometric fiber optic sensor is provided by combining the existing EFPI (Extrinsic Fabry-Perot Interferometer) fiber optic sensor with a direction-detecting sensor which can acquire direction information of a strain of a structure, which can solve a signal distortion problem occurring in the existing EFPI fiber optic sensor through a simple signal process. The patch-type extrinsic Fabry-Perot interferometric fiber optic sensor includes a piezoelectric material which can apply a control force to the existing EFPI fiber optic sensor, and a self-sensing bridge circuit for extracting the direction information when the piezoelectric material is used as an actuator, so that the sensible range of strain can be extended and the piezoelectric material can directly be used as the actuator based on the sensed signal.

Abstract: A method for using machine learning to solve problems having either a “positive” result (the event occurred) or a “negative” result (the event did not occur), in which the probability of a positive result is very low and the consequences of the positive result are significant. Training data is obtained and a subset of that data is distilled for application to a machine learning system. The training data includes some records corresponding to the positive result, some nearest neighbors from the records corresponding to the negative result, and some other records corresponding to the negative result. The machine learning system uses a co-evolution approach to obtain a rule set for predicting results after a number of cycles. The machine system uses a fitness function derived for use with the type of problem, such as a fitness function based on the sensitivity and positive predictive value of the rules. The rules are validated using the entire set of training data.

Abstract: A method for using machine learning to solve problems having either a “positive” result (the event occurred) or a “negative” result (the event did not occur), in which the probability of a positive result is very low and the consequences of the positive result are significant. Training data is obtained and a subset of that data is distilled for application to a machine learning system. The training data includes some records corresponding to the positive result, some nearest neighbors from the records corresponding to the negative result, and some other records corresponding to the negative result. The machine learning system uses a co-evolution approach to obtain a rule set for predicting results after a number of cycles. The machine system uses a fitness function derived for use with the type of problem, such as a fitness function based on the sensitivity and positive predictive value of the rules. The rules are validated using the entire set of training data.

Abstract: A method for generating a three-dimensional stereographic image on the basis of a two-dimensional planar image for one eye and depth information of the image stored in the Z-buffer. The first two-dimensional image for one eye of a viewer is created by rendering a three-dimensional model. Color information and depth information of the first two-dimensional image are stored in the frame buffer and the Z-buffer, respectively. The second two-dimensional image for the other eye of the viewer is created on the basis of the first two-dimensional image, the color information stored in the frame bufer, and the depth information stored in the Z-buffer. A stereographic image is generated on the basis of the first two-dimensional image and the second two-dimensional image.

Abstract: A method for using machine learning to solve problems having either a “positive” result (the event occurred) or a “negative” result (the event did not occur), in which the probability of a positive result is very low and the consequences of the positive result are significant. Training data is obtained and a subset of that data is distilled for application to a machine learning system. The training data includes some records corresponding to the positive result, some nearest neighbors from the records corresponding to the negative result, and some other records corresponding to the negative result. The machine learning system uses a co-evolution approach to obtain a rule set for predicting results after a number of cycles. The machine system uses a fitness function derived for use with the type of problem, such as a fitness function based on the sensitivity and positive predictive value of the rules. The rules are validated using the entire set of training data.

Abstract: A process for preparing a textile fiber of sulfonated poly(p-phenylene terephthalamide) that is rapidly dyeable to deep shades.