Query expansion is a method for alleviating the vocabulary mismatch problem present in information retrieval tasks. Previous works have shown that terms selected for query expansion by traditional methods such as pseudo-relevance feedback are not always helpful to the retrieval process. In this paper, we show that this is also true for more recently proposed embedding-based query expansion methods. We then introduce an artificial neural network classifier to predict the usefulness of query expansion terms. This classifier uses term word embeddings as inputs. We perform experiments on four TREC newswire and web collections show that using terms selected by the classifier for expansion significantly improves retrieval performance when compared to competitive baselines. The results are also shown to be more robust than the baselines.
The core of an earth dam is normally made of clayey base soils and protected by a sandy filter. Designing the filter can be very challenging considering the various criteria established today. This study commences with a review of three prominent criteria referred to in designing filters by the indirect method. No-erosion filter tests are then carried out involving 11 fine-grained, base soil samples with different properties; these are a highly dispersive soil, a soil with a medium to high degree of dispersion, a soil with a low-plasticity silt (ML) classification and a soil with a broad gradation curve. These tests represent the direct method. A comparison of results arising from the two methods shows that designing filters the indirect way, by following established criteria alone, is not always safe. Hence, it is proposed that such designs should be experimentally verified using no-erosion filter tests. Furthermore, instead of following the prevalent criteria in the indirect design method, the perfect filtering method for highly dispersive soils and the reduced particle size distribution method for broadly graded soils are proposed to be used, with merits given for each method. Finally, the paper recommends certain filter criteria from the literature for base soils with a moderate degree of dispersion and an ML classification respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.