Feng Wu

The bond behavior between the textile and the earth-based matrix determined the reinforcement effectiveness of the composite systems. This paper presented a pull-out experimental study on the glass textile mesh reinforced earth-based matrix. The bond behavior was studied using different development length, mesh spacing size and matrix thickness, with a total of 32 experimental specimens. The test results showed that the peak pull-out force had increased by 31.7% and 40.5% with 200 mm and 300 mm versus 100 mm development length, respectively. The 16 mm compared to 10 mm matrix thickness specimens had a high strength improvement (9.73%) because the elevated thickness had increased the matrix strength. However, the 20 mm versus 10 mm mesh spacing size specimens had achieved a slight reduction (5.72%) due to the reduction in the number of textiles along the weft direction. The failure mode shifted from pulling out, compound modes (both pulling out and textile rupture) to textile rupture mainly accompanied by elevated development length. In addition, we discussed the applicability of the trilinear bond-slip model on the earth-based matrix and proposed a method based on the fracture energy concept for estimating the effective development length, which could provide a reference for future research.

Recently, sentiment analysis has seen remarkable advance with the help of pre-training approaches. However, sentiment knowledge, such as sentiment words and aspect-sentiment pairs, is ignored in the process of pre-training, despite the fact that they are widely used in traditional sentiment analysis approaches. In this paper, we introduce Sentiment Knowledge Enhanced Pre-training (SKEP) in order to learn a unified sentiment representation for multiple sentiment analysis tasks. With the help of automatically-mined knowledge, SKEP conducts sentiment masking and constructs three sentiment knowledge prediction objectives, so as to embed sentiment information at the word, polarity and aspect level into pre-trained sentiment representation. In particular, the prediction of aspect-sentiment pairs is converted into multi-label classification, aiming to capture the dependency between words in a pair. Experiments on three kinds of sentiment tasks show that SKEP significantly outperforms strong pre-training baseline, and achieves new state-of-the-art results on most of the test datasets. We release our code at https://github.com/baidu/Senta.