Search

• Bond strength between non-uniformly corroded bars and concrete is discussed. • Cross-section area along the corroded bar length follows the lognormal distribution. • Uncertainties of the cross-section area of rebars and concrete strength are considered. • Probabilistic bond strength is predicted by using the Monte-Carlo method. The bond strength considering concrete strength and non-uniformly corrosion was tested by twelve beam-type specimens, and the corrosion was characterized by the 3D scanner. According to the K-S test, the cross-sectional area of the corroded bar follows the lognormal distribution at the 5% significance level, where its mean and standard deviation linearly correlate with the mass loss rate. The probabilistic bond strength is predicted by the Monte-Carlo method, incorporating the uncertainties of the cross-sectional area of rebars and the concrete strength. It is indicated that the failure probability can be decreased by decreasing mass loss rate and increasing concrete strength. [ABSTRACT FROM AUTHOR]

Aggregate question answering essentially returns answers for given questions by obtaining query graphs with unique dependencies between values and corresponding objects. Word order dependency , as the key to uniquely identify dependency of the query graph, reflects the dependencies between the words in the question. However, due to the semantic gap caused by the expression difference between questions encoded with word vectors and query graphs represented with logical formal elements, it is not trivial to match the correct query graph for the question. Most existing approaches design more expressive query graphs for complex questions and rank them just by directly calculating their similarities, ignoring the semantic gap between them. In this paper, we propose a novel Structure-sensitive Semantic Matching(SSM) approach that learns aligned representations of dependencies in questions and query graphs to eliminate their gap. First, we propose a cross-structure matching module to bridge the gap between two modalities(i.e., textual question and query graph). Then, we propose an entropy-based gated AQG filter to remove the structural noise caused by the uncertainty of dependencies. Finally, we present a two-channel query graph representation that fuses the semantics of abstract structure and grounding content of the query graph explicitly. Experimental results show that SSM could learn aligned representations of questions and query graphs to eliminate the gaps between their dependencies, and improves up to 12% (F1 score) on aggregation questions of two benchmark datasets. [ABSTRACT FROM AUTHOR]

In this paper, we present a forgetting-based approach to handling inconsistency in DL-Lite. Our proposed approach cannot only characterize distance-based reasoning, which is proven to rationally draw meaningful conclusions even from inconsistent DL-Lite knowledge bases but also recovery the consistency of DL-Lite knowledge bases. We first present vectors forgetting for DL-Lite by extending predicates forgetting in DL-Lite and show the predicates to be forgotten which are obtained by computing minimal hitting sets. Moreover, we develop algorithms to compute those predicates by employing Reiter’s HS-tree method and then analyze the computational complexity of those proposed algorithms. Finally, we implement our proposed algorithms and evaluate them on both consistent and inconsistent ontologies. Besides, we discuss some applications of vectors forgetting in privacy protection. [ABSTRACT FROM AUTHOR]

In this study, we proposed a reliable method to quickly and simultaneously determine the content of sodium benzoate and potassium sorbate simultaneously by ultraviolet (UV) spectrophotometry with BP-neural network algorithm(BP-ANN) and partial least squares regression(PLS). The pure reagents were used to prepare a series of sodium benzoate and potassium sorbate solutions and used deionized water as a reference solution. The calibration models were constructed by using 36 reference samples according to the orthogonal design and the prediction model set consisted of 9 sets of randomly configured mixed solutions. The results used BP-ANN and PLS showed that the root mean square error prediction (RMSEP) of sodium benzoate and potassium sorbate were 0.129, 0.09 and 0.155, 0.089, and the correlation coefficient (R2) were 0.9997, 0.9994 and 0.9998, 0.9995, respectively. The recovery of the actual samples by both methods was over 97%. [ABSTRACT FROM AUTHOR]