Abstract: A method for clustering based on unsupervised learning according to an embodiment of the invention enables clustering for newly generated patterns and is robust against noise, and does not require tagging for training data. According to one or more embodiments, noise is accurately removed using three-dimensional stacked spatial auto-correlation, and multivariate spatial probability distribution values and polar coordinate system spatial probability distribution values are used as learning features for clustering model generation, making them robust to noise, rotation, and fine unusual shapes. In addition, clusters resulting from clustering are classified into multi-level clusters, and stochastic automatic evaluation of normal/defect clusters is possible only with measurement data without a label.

Abstract: A method for clustering based on unsupervised learning according to an embodiment of the invention enables clustering for newly generated patterns and is robust against noise, and does not require tagging for training data. According to one or more embodiments, noise is accurately removed using three-dimensional stacked spatial auto-correlation, and multivariate spatial probability distribution values and polar coordinate system spatial probability distribution values are used as learning features for clustering model generation, making them robust to noise, rotation, and fine unusual shapes. In addition, clusters resulting from clustering are classified into multi-level clusters, and stochastic automatic evaluation of normal/defect clusters is possible only with measurement data without a label.

Abstract: A method for clustering based on unsupervised learning according to an embodiment of the invention enables clustering for newly generated patterns and is robust against noise, and does not require tagging for training data. According to one or more embodiments of the invention, noise is accurately removed using three-dimensional stacked spatial auto-correlation, and multivariate spatial probability distribution values and polar coordinate system spatial probability distribution values are used as learning features for clustering model generation, making them robust to noise, rotation, and fine unusual shapes. In addition, clusters resulting from clustering are classified into multi-level clusters, and stochastic automatic evaluation of normal/defect clusters is possible only with measurement data without a label.

Abstract: A polyamide resin composition of the present invention comprises: approximately 100 parts by weight of a polyamide resin comprising approximately 5 to 40 weight % of an aromatic polyamide resin and approximately 60 to 95 weight % of an aliphatic polyamide resin; approximately 30 to 70 parts by weight of an inorganic filler; and approximately 0.5 to 5 parts by weight of modified polyalkylene glycol. The modified polyalkylene glycol comprises an isocyanate group, a succinimide group or an epoxy group at one terminal of polyalkylene glycol. The polyamide resin composition exhibits excellent adhesiveness with other materials, excellent antifouling properties, impact resistance, stiffness, injection moldability and the like.

Abstract: A thermoplastic resin composition of the present invention comprises: about 100 parts by weight of polycarbonate resin; about 20 to about 60 parts by weight of mica which has been surface-treated with a silane compound; about 0.1 to about 20 parts by weight of a modified polyolefin comprising a repeating unit represented by chemical formula 1 and a repeating unit represented by chemical formula 2; and about 3 to about 20 parts by weight of a phosphorus flame retardant. The thermoplastic resin composition has excellent impact resistance, strength, flame retardancy, fluidity, and the like.

Abstract: A thermoplastic resin composition of the present invention comprises: about 100 parts by weight of polycarbonate resin; about 20 to about 60 parts by weight of mica which has been surface-treated with a silane compound; about 0.1 to about 20 parts by weight of a modified polyolefin comprising a repeating unit represented by chemical formula 1 and a repeating unit represented by chemical formula 2; and about 3 to about 20 parts by weight of a phosphorus flame retardant. The thermoplastic resin composition has excellent impact resistance, strength, flame retardancy, fluidity, and the like.

Abstract: A method for clustering based on unsupervised learning according to an embodiment of the invention enables clustering for newly generated patterns and is robust against noise, and does not require tagging for training data. According to one or more embodiments of the invention, noise is accurately removed using three-dimensional stacked spatial auto-correlation, and multivariate spatial probability distribution values and polar coordinate system spatial probability distribution values are used as learning features for clustering model generation, making them robust to noise, rotation, and fine unusual shapes. In addition, clusters resulting from clustering are classified into multi-level clusters, and stochastic automatic evaluation of normal/defect clusters is possible only with measurement data without a label.

Abstract: A polyamide resin composition includes: about 5% by weight (wt %) to about 50 wt % of an aromatic polyamide resin; about 10 wt % to about 60 wt % of an aliphatic polyamide resin; about 20 wt % to about 60 wt % of inorganic fillers; and about 0.5 wt % to about 5 wt % of a polyetheresteramide block copolymer, wherein the polyetheresteramide block copolymer is a polymer of a reaction mixture including about 5 wt % to about 95 wt % of an amino carboxylic acid, lactam or diamine-dicarboxylic acid salt and about 5 wt % to about 95 wt % of polyalkylene glycol. The polyamide resin composition can exhibit good adhesion with respect to other materials and anti-contamination (antifouling) properties.