This paper explores protein-protein interaction extraction from biomedical literature using Support Vector Machines (SVM). Besides common lexical features, various overlap features and base phrase chunking information are used to improve the performance. Evaluation on the AIMed corpus shows that our feature-based method achieves very encouraging performances of 68.6 and 51.0 in F-measure with 10-fold pairwise cross-validation and 10-fold document-wise cross-validation respectively, which are comparable with other state-of-the-art feature-based methods.
Experiences manifest the importance of comovement and communicable characters among the risks of financial assets. Therefore, the portfolio view considering dependence relationship among credit entities is at the heart of risk measurement. This paper introduces a mixed Poisson model assuming default probabilities of obligors depending on a set of common economic factors to construct the dependence structure of obligors. Further, we apply mixed Poisson model into an empirical study with data of four industry portfolios in the financial market of China. In the process of model construction, the classical structural approach and option pricing formula contribute to estimate dynamic default probabilities of single obligor, which helps to obtain the dynamic Poisson intensities under the model assumption. Finally, given the values of coefficients in this model calculated by a nonlinear estimation, Monte Carlo technique simulates the progress of loss occurrence. Relationship between default probability and loss level reflected through the MC simulation has practical features. This study illustrates the practical value and effectiveness of mixed Poisson model in risk measurement for credit portfolio.
Hindawi Publishing Corporation
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.