Thư viện số Bộ sưu tập:http://thuvienso.thanglong.edu.vn//handle/TLU/132572026-04-08T15:56:25Z2026-04-08T15:56:25ZBalancing Accuracy and Interpretability in Credit Risk Modeling: Evidence from Peer-to-Peer LendingThuy Tien Dinhhttps://ikr.inceif.org/retrieve/bd8db5c8-a35c-4be3-a153-fb50b18db576/TCTL.0000110_Cân bằng giữa độ chính xác và khả năng diễn giải trong mô hình rủi ro tín dụng Bằng chứng từ bài toán cho vay ngang hàng (P2P Lending).pdf.jpg2025-08-14T07:43:32Z2025-01-01T00:00:00ZBài báo/NewspaperNhan đề : Balancing Accuracy and Interpretability in Credit Risk Modeling: Evidence from Peer-to-Peer Lending
Tác giả : Thuy Tien Dinh
Tóm tắt : Accurate credit risk assessment is crucial for the stability and growth of peer-to-peer (P2P) lending platforms. This study investigates the effectiveness of machine learning models in predicting loan defaults using historical Lending Club data. We evaluate logistic regression, decision tree, and random forest, employing feature engineering techniques like one-hot and weight of evidence encoding. Model performance is assessed using K-fold cross-validation and metrics such as accuracy and AUC. To enhance model interpretability, we utilize explainable AI techniques like LIME and SHAP, enabling lenders and borrowers to understand the factors driving loan defaults. Our findings demonstrate that while complex models offer higher predictive accuracy, simpler models like logistic regression with WoE encoding provide a suitable balance between accuracy and interpretability, fostering trust and responsible lending within the P2P lending ecosystem.2025-01-01T00:00:00ZImages of the Singer transfers and their possibility to be injectiveNguyen Huu Viet Hunghttps://ikr.inceif.org/retrieve/c9082feb-24b6-4580-98fe-e979e72e4dfc/TCTL.0000109_Images of the Singer transfers and their possibility to be injective.pdf.jpg2025-08-14T07:43:08Z2025-01-01T00:00:00ZBài báo/NewspaperNhan đề : Images of the Singer transfers and their possibility to be injective
Tác giả : Nguyen Huu Viet Hung
Tóm tắt : This article is an attempt to investigate the possibility to be injective of the Singer transfer TrM s : F2 ⊗GLs P(H∗Vs ⊗ M∗) → Exts A (Σ-sM, F2) for M being the A -modules F2 = Hå∗S0 or Hå∗RP∞. The existence of a positive stem critical element of Exts,t A (Hå∗RP∞, F2) in the image of the transfer TrRP∞ s is equivalent to the existence of a positive stem critical element of Exts+1,t+1 A (F2, F2) in the image of the transfer Trs+1. If the existences happen, then TrRP∞ s and Trs+1 are not injective. We show that the critical element P h ä2 is not in the image of the fourth transfer, TrRP∞ 4 : F2 ⊗GL4 P(H∗V4 ⊗ Hå∗RP∞)t-4 → Ext4 A,t(Hå∗RP∞, F2). Singer’s conjecture is still open, as we have not known any critical element, which is in the image of the transfer.2025-01-01T00:00:00ZRandom dynamical system generated by nonautonomous stochastic differential equations driven by fractional Brownian motionsHong Phan Thanhhttps://ikr.inceif.org/retrieve/08fa5c17-697b-4a56-a3ee-841700b8bede/TCTL.0000108_Random dynamical systems generated by nonautonomous stochastic differential equations driven by fractional Brownian motions.pdf.jpg2025-08-14T07:42:43Z2025-01-01T00:00:00ZBài báo/NewspaperNhan đề : Random dynamical system generated by nonautonomous stochastic differential equations driven by fractional Brownian motions
Tác giả : Hong Phan Thanh
Tóm tắt : In this paper, we prove that a non-autonomous stochastic differential equation generates a continuous random dynamical system. The flow then possesses a random pullback attractor under the dissipativity condition(s) of the drift and smallness of diffusion part2025-01-01T00:00:00ZAn algorithm for solving the variational inequality problem over the solution set of the split variational inequality and fixed point problemViệt Anh Trầnhttps://ikr.inceif.org/retrieve/48092d0a-f64e-4410-98d7-7a4f46082956/TCTL.0000107_Thuật toán giải bài toán bất đẳng thức biến phân trên tập nghiệm của bài toán bất đẳng thức biến phân và điểm bất động tách.pdf.jpg2025-08-14T07:42:02Z2025-01-01T00:00:00ZBài báo/NewspaperNhan đề : An algorithm for solving the variational inequality problem over the solution set of the split variational inequality and fixed point problem
Tác giả : Việt Anh Trần
Tóm tắt : In this paper, we introduce a new algorithm for solving strongly monotone variational inequality problem, where the constraint set is the solution set of the split variational inequality and fixed point problem. Our method uses dynamic step sizes selected based on information of the previous step, which gives strong convergence result without the prior knowledge of the given bounded linear operator’s norm. In addition, using our method, we do not require any information of the Lipschitz and strongly monotone constants of the mappings. Several corollaries of our main result are also presented. Finally, a numerical example has been given to illustrate the effectiveness of our proposed algorithm.2025-01-01T00:00:00Z