No abstract

No abstract

No abstract

Text is an important form of information transfer. In a P2P (Peer-to-Peer) loan market with severe information asymmetry, loan descriptive information voluntarily written by a borrower may generate significant influence on investors’ decision-making. As an important language feature, readability determines whether the audience can accurately identify textual contents. To discuss the influence of readability of loan description on borrowing behaviors, the data of a Chinese P2P lending platform Renrendai during 2013–2017 were used in this study. Readability index of loan description was constructed through the textual analysis method, followed by an empirical verification of the influences of readability of loan description on loan success rate and loan cost. Results indicate that the readability of loan description in China’s P2P loan market can generate incremental information, and the readability of loan description has evident nonlinear relations with loan success rate and loan interest rate. Moreover, the readability of loan description presents a “reverse U-shaped” relation with loan success rate and a “U-shaped” relation with loan interest rate. Demographic information (gender and educational background) of borrowers will not influence investors’ feedback behaviors toward the readability of loan description. Conclusions provide empirical evidence for standardizing loan descriptive information on P2P lending platforms.

No abstract

No abstract

We investigated the occurrence sequence of the deformation-induced e-martensite (DIEM) and the mechanical twinning (MT) during different tensile deformation at 300 K and 223 K in an Fe-17Mn-3Si-0.6C high manganese steel, respectively. The results showed that the MT occurred first when deformed at 300 K, while the DIEM occurred first when deformed at 223 K. The occurrence sequence of the DIEM and the MT in this steel could be predicted when the critical stress for the DIEM was determined by the 0.01% proof stress, and that for the MT was calculated by the Byun model only if the stacking fault energy was thermodynamically calculated using mainly the parameters of Scientific Group Thermodata Europe.

In the process of power operation, the correct identification of tools can lay a foundation for the detection of violations in power operation. In order to realize the recognition of power instruments, based on the current Yolo V5 algorithm, a detection algorithm for power instruments is proposed by improving Yolo V5 algorithm. Firstly, the model of Yolo V5 algorithm is introduced. Then the establishment of the power tool database and the process of model training are analysed. Finally, the test results are analysed and evaluated. The models generated after training were accelerated by TensorRT and then deployed on Jetson Xavier NX.