• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.45.2023.tde-03042024-164514
Document
Author
Full name
Roberto Almeida Shimizu
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2023
Supervisor
Committee
Silva, Flavio Soares Correa da (President)
Corrêa, Pedro Luiz Pizzigatti
Sultana, Tunazzina
Title in English
Creditworthiness using social capital: a theoretical framework
Keywords in English
Financial inclusion
Graph algorithms
Peer-to-peer loans
ROSCAs
Social capital
Trust and reputation systems
Abstract in English
It is well known that in small communities, social capital plays a very important role in allowing economic transactions, especially in solidary lending arrangements, such as Rotating Savings and Credit Associations (ROSCAs), peer-to-peer loans among family and friends, and so on. More recently there were lots of academic papers analyzing the performance of peer-to-peer credit in fintechs such as Prosper, Lending Club, Kiva. These studies reveal an intriguing trend: loans funded within personal networks often exhibit higher repayment rates compared to conventional loans. Nevertheless, there remains a significant gap in research, particularly in understanding the quantification of social capital within social networks and its subsequent application in assessing an individuals creditworthiness. Trust and Reputation systems is an area of computer science which have attracted attention in various networking environments, including online social networks, wireless communication networks, multi-agent systems, and peer-to-peer networks. Trust and Reputation systems can use explicit or implicit information for decision making. The research, embarked on a bibliographic analysis of algorithms specifically designed for computing trust and reputation, which act as proxies for social capital in social networks. The algorithms selected for this study were eigentrust \cite{conf/www/KamvarSG03}, tidaltrust \cite and graph algorithms \cite. They were applied in real cases of solidary credit arrangements observed in the community, more especifically a case of a ROSCA (Consorcio entre amigos) and two cases of peer-to-peer loans (Emprestimos entre amigos e familiares) that were derived from the former ROSCA. The objective was to evaluate the effectiveness and limitations of these algorithms and to address the challenges inherent in gathering trust and reputation data, considering its sensitive nature. Furthermore, the research extends to examining how these algorithms can predict the likelihood of creditworthiness using real dataset of indian villages, database provided in the paper The Diffusion of Microfinance By Abhijit Banerjee, Arun G. Chandrasekhar, Esther Duflo and Matthew Jackson. The goal of this research is to contribute to the development of more inclusive credit scoring methods, particularly targeting less-privileged segments of the population. By integrating interdisciplinary approaches, the study provides valuable insights into the potential of leveraging social capital and trust mechanisms to enhance financial inclusion and support economic empowerment in small communities.
Title in Portuguese
Confiabilidade de crédito usando capital social
Keywords in Portuguese
Algoritmos em grafos
Capital social
Empréstimos peer-to-peer
Inclusão financeira
ROSCAs
Sistemas de confiança e reputação
Abstract in Portuguese
O capital social é crucial em comunidades pequenas, facilitando transações econômicas e arranjos de empréstimos solidários, como Associações Rotativas de Poupança e Crédito (ROSCAs) e empréstimos peer-to-peer. Estudos recentes, focalizando empréstimos peer-to-peer em fintechs como Prosper e Lending Club, indicam que empréstimos dentro de redes pessoais têm maiores taxas de adimplência do que os convencionais. Os Sistemas de Confiança e Reputação, essenciais na ciência da computação, são utilizados para avaliar a confiabilidade em diversas redes, tais como em redes de comércio eletrônico e mais recentemente incluindo redes sociais. Esses sistemas operam com dados explícitos e implícitos para estabelecer medidas de confiança. A presente pesquisa analisou algoritmos voltados para a computação de confiança e reputação, representando o capital social em redes. Foram escolhidos algoritmos como eigentrust e tidaltrust para aplicação em estudos de caso reais de crédito solidário, incluindo um ROSCA e dois empréstimos ponto a ponto. O estudo buscou entender a eficácia desses algoritmos e os desafios de coletar dados sobre confiança e reputação. Adicionalmente, investigou-se como esses algoritmos poderiam prever solidez de crédito em redes sociais de aldeias indianas, utilizando dados do estudo A Difusão da Microfinança. Este trabalho visa contribuir para o desenvolvimento de métodos de pontuação (escore) de crédito mais inclusivos, focados em populações menos privilegiadas. Através de uma abordagem interdisciplinar, o estudo destaca o potencial de utilizar capital social e mecanismos de confiança para fomentar a inclusão financeira e empoderamento econômico em comunidades pequenas
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
SocialCapital.pdf (2.53 Mbytes)
Publishing Date
2024-04-03
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.