Taking the Human Out of the Loop: A Review of Bayesian OptimizationBobak Shahriari, Kevin Swersky, Ziyu Wang et al.|Proceedings of the IEEE|2015 Big Data applications are typically associated with systems involving large numbers of users, massive complex software systems, and large-scale heterogeneous computing and storage architectures. The construction of such systems involves many distributed design choices. The end products (e.g., recommendation systems, medical analysis tools, real-time game engines, speech recognizers) thus involve many tunable configuration parameters. These parameters are often specified and hard-coded into the software by various developers or teams. If optimized jointly, these parameters can result in significant improvements. Bayesian optimization is a powerful tool for the joint optimization of design choices that is gaining great popularity in recent years. It promises greater automation so as to increase both product quality and human productivity. This review paper introduces Bayesian optimization, highlights some of its methodological aspects, and showcases a wide range of applications.
Object-oriented classification and application in land use classification using SPOT-5 PAN imageryHigh-resolution remotely sensed data have been actively employed in urban land use/cover. Object-oriented classification techniques based on image segmentation are being actively studied in the high-resolution image process and interpretation to extract a variety of thematic information. Different from the pixel-based image analysis, the processing of the object-oriented method is based on image object or segment, not single pixel. The object-oriented classification includes two consecutive processes. An image is subdivided into separated regions according to the spectral and spatial heterogeneity in the image segmentation process. Then the objects are assigned to a specific class according to the class's detailed description in the image classification process. As a case study, the study area is a pail of the planning Beijing Olympic Games Cottage, which has changed greatly with the advent of the year of 2008. The panchromatic SPOT-5 image in August of 2002 is segmented and these segments then are classified to hierarchically linked objects by the eCognition software. The overall classification accuracy is up to 87%.