A variety of approaches to knowledge base refinement [3, 81 and rule acquistion [4] have appeared recently. This paper is concerned with the means by which alternative refinement systems themselves may be specified, developed, and studied. The anticipated virtues of taking a metalinguistic approach to these tasks are described, and shown to be confirmed by experience with an actual refinement m...

Transaction frauds impose serious threats onto e-commerce. We present CLUE, a novel deep-learning-based transaction fraud detection system we design and deploy at JD.com, one of the largest ecommerce platforms in China with over 220 million active users. CLUE captures detailed information on users’ click actions using neural-network based embedding, and models sequences of such clicks using the...

We analyze recycling decisions for bioplastics using a natural field experiment. Bioplastics have environmental benefits – such as reduced energy use in production and enhanced biodegredation compared to conventional plastics. Recycling that are not consistent with government guidelines, however, may cause rebound effect. For instance when biobased plastics contaminate organic waste streams, or...

What is the logic of scope? By “scope”, I mean scopetaking in natural languages such as English, as illustrated by the sentence Ann saw everyone. In this example, the quantifier denoted by everyone takes scope over the rest of the sentence, that is, it takes the denotation of the rest of the sentence as its semantic argument: everyone(λx.saw(x)(ann)). The answer I will give here will be to prov...

”Semantic gap” is an open challenging problem in content-based image retrieval. It reflects the discrepancy between low-level imagery features used by the retrieval algorithm and high-level concepts required by system users. This paper introduces a novel image retrieval scheme, CLUster-based rEtrieval of images by unsupervised learning (CLUE), to tackle the semantic gap problem. CLUE is built o...