While Peer-to-Peer (P2P) lending is rapidly growing, it is also accompanied by high credit risk due to information asymmetry. Besides conventional hard information, soft information also enters into the lending decision process. The descriptive loan texts submitted by borrowers have great potential for exploiting useful soft factors, but also pose great challenges due to the semantic sensitivity to context and the complexity of content representation. We propose a novel text mining method for automatically extracting semantic soft factors from descriptive loan texts. The method maps terms to an embedding space, assembles semantically related terms together into semantic cliques, and then defines semantic soft factors corresponding to the semantic cliques. Empirical evaluation shows that the extracted semantic soft factors contributed to significant improvement on credit risk evaluation in terms of both discrimination performance and granting performance. This work advances our knowledge of soft information indicative of a borrower’s credit risk.

Owing to information asymmetry, evaluating the credit risk of small- and medium-sized enterprises (SMEs) is difficult. While previous studies evaluating the credit risk of SMEs have mostly focused on intrinsic risk generated by SMEs, our study considers both intrinsic and relational risks generated by neighbor firms’ publicly available risk events. We propose a framework for quantifying relational risk based on publicly available risk events for SMEs’ credit risk evaluation. Our proposed framework quantifies relational risk by weighting the impact of publicly available risk events of each firm in an interfirm network—considering the impact of interfirm network type, risk event type, and time dependence of risk events—and combines the relational risk score with financial and demographic features to evaluate SMEs credit risk. Our results reveal that relational risk score significantly improves both discrimination and granting performances of credit risk evaluation of SMEs, providing valuable managerial and practical implications for financial institutions.

Soil nutrient prediction has been eliciting increasing attention in agricultural production. Backpropagation (BP) neural networks have demonstrated remarkable ability in many prediction scenarios. However, directly utilizing BP neural networks in soil nutrient prediction may not yield promising results due to the random assignment of initial weights and thresholds and the tendency to fall into local extreme points. In this study, a BP neural network model optimized by an improved genetic algorithm (IGA) was proposed to predict soil nutrient time series with high accuracy. First, the crossover and mutation operations of the genetic algorithm (GA) were improved. Next, the IGA was used to optimize the BP model. The symmetric nature of the model lies in its feedforward and feedback connections, i.e., the same weights must be used for the forward and backward passes. An empirical evaluation was performed using annual soil nutrient data from China. Soil pH, total nitrogen, organic matter, fast-acting potassium, and effective phosphorus were selected as evaluation indicators. The prediction results of the IGA–BP, GA–BP, and BP neural network models were compared and analyzed. For the IGA–BP prediction model, the coefficient of determination for soil pH was 0.8, while those for total nitrogen, organic matter, fast-acting potassium, and effective phosphorus were all greater than 0.98, exhibiting a strong generalization ability. The root-mean-square errors of the IGA–BP prediction models were reduced to 50% of the BP models. The results indicated that the IGA–BP method can accurately predict soil nutrient content for future time series.

Practice- and policy-oriented abstract for “Research Spotlights” Although enjoying rapid development, online lending also endures some unusual risk, that is, platform risk. We address a new problem at the macro platform level, platform risk evaluation, and explore types of information and methods that are effective in predicting platform risk. We identify four types of information, that is, platform characteristic, risk management, commercial competition, and online word of mouth, and examine their utilities, separately and jointly, in predicting platform risk. We also propose the use of survival analysis, especially the mixture survival model, in predicting whether and when a platform will default. We carry out a cross-stage analysis using data crawled from two leading web portals for online lending in China with the two stages separated by the recent dramatic policy intervention. The results reveal the differences among the four identified factors in terms of predictive utility, the heterogeneity between the two types of default platforms, and differences between the start-up and stable periods of platform development. Based on the results, we derive some insights and examine the cross-stage changes and commonalities. We provide both lessons learned from the past and practical implications for market managers and lenders in the current online lending market.