Credit Risk Assessment in P2P Lending Using LightGBM and Particle Swarm Optimization

Dasril, Yosza; Muslim, Much Aziz; Hakim, M. Faris Al; Jumanto, Jumanto; Prasetiyo, Budi

doi:10.26594/register.v9i1.3060

Cited by 3 publications

(1 citation statement)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the most popular classification algorithm models is the Support Vector Machine (SVM) which separates two classes of data with a hyperplane. SVM has been widely used in various fields due to its superior capabilities in fault diagnosis [44], disease detection [45], [46], credit fraud detection [47], [48], and financial prediction [49]. Certain investigations applied PCA feature extraction method for model optimization [50] by reducing data dimensionality and computational burden, as well as expediting the classification process.…”

Section: Introductionmentioning

confidence: 99%

Optimizing Support Vector Machine Performance for Parkinson's Disease Diagnosis Using GridSearchCV and PCA-Based Feature Extraction

Jumanto,

Rofik,

Sugiharti

et al. 2024

J. Inf. Syst. Eng. Bus. Intell.

View full text Add to dashboard Cite

Background: Parkinson's disease (PD) is a critical neurodegenerative disorder affecting the central nervous system and often causing impaired movement and cognitive function in patients. In addition, its diagnosis in the early stages requires a complex and time-consuming process because all existing tests such as electroencephalography or blood examinations lack effectiveness and accuracy. Several studies explored PD prediction using sound, with a specific focus on the development of classification models to enhance accuracy. The majority of these neglected crucial aspects including feature extraction and proper parameter tuning, leading to low accuracy. Objective: This study aims to optimize performance of voice-based PD prediction through feature extraction, with the goal of reducing data dimensions and improving model computational efficiency. Additionally, appropriate parameters will be selected for enhancement of the ability of the model to identify both PD cases and healthy individuals. Methods: The proposed new model applied an OpenML dataset comprising voice recordings from 31 individuals, namely 23 PD patients and 8 healthy participants. The experimental process included the initial use of the SVM algorithm, followed by implementing PCA for feature extraction to enhance machine learning accuracy. Subsequently, data balancing with SMOTE was conducted, and GridSearchCV was used to identify the best parameter combination based on the predicted model characteristics. Result: Evaluation of the proposed model showed an impressive accuracy of 97.44%, sensitivity of 100%, and specificity of 85.71%. This excellent result was achieved with a limited dataset and a 10-fold cross-validation tuning, rendering the model sensitive to the training data. Conclusion: This study successfully enhanced the prediction model accuracy through the SVM+PCA+GridSearchCV+CV method. However, future investigations should consider an appropriate number of folds for a small dataset, explore alternative cross-validation methods, and expand the dataset to enhance model generalizability. Keywords: GridSearchCV, Parkinson Disaese, SVM, PCA, SMOTE, Voice/Speech

show abstract

Section: Introductionmentioning

confidence: 99%

Optimizing Support Vector Machine Performance for Parkinson's Disease Diagnosis Using GridSearchCV and PCA-Based Feature Extraction

Jumanto,

Rofik,

Sugiharti

et al. 2024

J. Inf. Syst. Eng. Bus. Intell.

View full text Add to dashboard Cite

show abstract

Estado del arte del financiamiento colectivo de préstamos: un enfoque de probabilidad de default, 2008-2023

González,

García Salgado,

Morales Castro

2024

Revista de Economía

View full text Add to dashboard Cite

El préstamo entre particulares o mejor conocido como préstamos de persona a persona (P2P) es un modelo de negocio esencial en las finanzas tecnológicas y una alternativa orientada al fomento de obtención de fondos para prestatarios de forma novedosa. Sin embargo, la posibilidad de que solicitantes que acceden a préstamos mediante este modelo de negocio no cumplan con sus obligaciones de pago es una preocupación importante para prestamistas e inversores. De este modo, el presente estudio tiene como objetivo examinar los enfoques bajo los que se ha analizado este modelo de negocio e identificar patrones y tendencias de esta temática a través de un análisis bibliométrico, con el apoyo de la herramienta Bibliometrix de R y VOSviewer. Se identificaron 7 enfoques de estudio y una parte esencial de los factores que pueden aumentar la probabilidad de incumplimiento. Los hallazgos de la bibliometría sugieren que las publicaciones principales provienen de Estados Unidos y China, de la University of Maryland y National Kaohsiung University of Science and Technology, como revista destacada se tiene a SSRN Electronic Journal, las plataformas mayormente analizadas son Lendingclub.com y Prosper.com, las técnicas estadísticas utilizadas principalmente son probit y regresión logística. Bin Gu se destaca por su producción y colaboración con otros autores. La coocurrencia de palabras clave muestra que peer to peer lending, default risk y fintech aparecen con mayor frecuencia. Finalmente, el mapeo temático indica que los temas screening, market inference, P2P lending, misreporting, China e information asimetry son tópicos principales.

show abstract

Exploring the Machine Learning-Based Decision Tree Technique in Assessing Credit Risk Likelihood of P2P Fintech Lending for Effective Marketing

Kaur,

Bhatnagar,

Singh

2024

Advances in Business Strategy and Competitive Advantage

View full text Add to dashboard Cite

Peer-to-Peer lending as a significant alternative finance source in reducing the credit gap worldwide. It is an innovative Fintech business model that integrates lenders and borrowers through the internet without the interference of banks. These lending platforms as Non-banking financial institutions (NBFCs) perform the credit assessment and determine borrower's eligibility to safeguard investors and reduce default risk. The study aims to provide insight into their functionality introducing a novel concept of linking borrowers and lenders by exploring the decision tree technique algorithm based on machine learning for forecasting default risk. Additionally, it highlights the parameters for judging the reliability of borrowers raising funds through such platforms. Additionally, the credibility of these techniques has a psychological effect on investors in making effective investment decisions, and it has a significant impact on corporate professionals in the effective marketing of such platforms.

show abstract

Credit Risk Assessment in P2P Lending Using LightGBM and Particle Swarm Optimization

Cited by 3 publications

References 45 publications

Optimizing Support Vector Machine Performance for Parkinson's Disease Diagnosis Using GridSearchCV and PCA-Based Feature Extraction

Optimizing Support Vector Machine Performance for Parkinson's Disease Diagnosis Using GridSearchCV and PCA-Based Feature Extraction

Estado del arte del financiamiento colectivo de préstamos: un enfoque de probabilidad de default, 2008-2023

Exploring the Machine Learning-Based Decision Tree Technique in Assessing Credit Risk Likelihood of P2P Fintech Lending for Effective Marketing

Contact Info

Product

Resources

About