Automatic identification of coding best practices can scale the development of code and application analyzers. We present Doc2BP, a deep learning tool to identify coding best practices in software documentation. Natural language descriptions are mapped to an informative embedding space, optimized under the dual objectives of binary and few shot classification. The binary objective powers general classification into known best practice categories using a deep learning classifier. The few shot objective facilitates example-based classification into novel categories by matching embeddings with user-provided examples at run-time, without having to retrain the underlying model. We analyze the effects of manually and synthetically labeled examples, context, and cross-domain information.We have applied Doc2BP to Java, Python, AWS Java SDK, and AWS CloudFormation documentations. With respect to prior works that primarily leverage keyword heuristics and our own parts of speech pattern baselines, we obtain 3-5% F1 score improvement for Java and Python, and 15-20% for AWS Java SDK and AWS CloudFormation. Experiments with four few shot use-cases show promising results (5-shot accuracy of 99%+ for Java NullPointerException and AWS Java metrics, 65% for AWS CloudFormation numerics, and 35% for Python best practices).Doc2BP has contributed new rules and improved specifications in Amazon's code and application analyzers: (a) 500+ new checks in cfn-lint, an open-source AWS CloudFormation linter, (b) over 97% automated coverage of metrics APIs and related practices in Amazon DevOps Guru, (c) support for nullable AWS APIs in Amazon CodeGuru's Java NullPointerException (NPE) detector, (d) 200+ new best practices for Java, Python, and respective AWS SDKs in Amazon CodeGuru, and (e) 2% reduction in false positives in Amazon CodeGuru's Java resource leak detector.

Tagged Web images provide an abundance of labeled training examples for visual concept learning. However, the performance of automatic training data selection is susceptible to highly inaccurate tags and atypical images. Consequently, manually curated training data sets are still a preferred choice for many image annotation systems. This paper introduces ARTEMIS - a scheme to enhance automatic selection of training images using an instance-weighted mixture modeling framework. An optimization algorithm is derived to learn instance-weights in addition to mixture parameter estimation, essentially adapting to the noise associated with each example. The mechanism of hypothetical local mapping is evoked so that data in diverse mathematical forms or modalities can be cohesively treated as the system maintains tractability in optimization. Finally, training examples are selected from top-ranked images of a likelihood-based image ranking. Experiments indicate that ARTEMIS exhibits higher resilience to noise than several baselines for large training data collection. The performance of ARTEMIS-trained image annotation system is comparable with usage of manually curated data sets.

Religious iconography is an integral component of the cultural heritage of Cyprus, which was once a part of the great Byzantine empire. On one hand, icons exhibit strict adherence to conventional symbols, poses and apparel. On the other hand, there is a great variety in the style of depiction that can be attributed to different schools and periods. This paper proposes an active shape model (ASM) based technique for icon face representation that can be used for style comparison and attribution. For centuries-old icons suffering from loss of paint, cracks and added noise from digitization artifacts, we apply an independent component analysis (ICA) technique to enhance the paintings' original work. The experimental results show that our method can effectively characterize Cypriot icons.

A semantic concept can be a physical object (e.g., 'car', 'zebra-fish'), an activity (e.g., 'demonstration', 'running') or an obscure category (e.g., 'historical', 'autumn'). In automatic image annotation, machines are taught to infer such semantic concepts by training with hundreds of manually selected images that contain those concepts. So far, remarkable progress has been made in the areas of visual feature extraction and machine learning algorithms that relate features to concepts. Now the new challenge is to scale the inference and annotation capacity to thousands of semantic concepts in the real world.

The traditional near-duplicate detection systems developed for digital photo management and copyright protection are not applicable for the de-duplication of large-scale web image corpus. In this paper, we present a fast, accurate and highly scalable image fingerprinting technique suited for near-duplicate detection at the web-scale. The image fingerprint is a compact 130 bit representation computed using Fourier-Mellin transform. Near-duplicate images are detected in O(1) time using fingerprint equality and is faster than fast approximate near-neighbor searches like LSH.

vADeo is a next-generation video ad system that analyzes the video to find appropriate scene changes where ads can be inserted. The context of each ad insertion point is determined through high-level analysis of the surrounding video segment thereby making ads contextual. Further vADeo implements two novelties on the player side - ad book-marking and delayed interaction - which encourage ad clicks without disrupting the video viewing experience.