Scholarlink

<p>Credit risk assessment is critical for financial institutions’ stability and profitability. Traditional methods struggle to assess borrowers with limited credit history or non-traditional income due to insufficient data and outdated models. Big data analytics, characterized by the 5Vs (Volume, Variety, Velocity, Veracity, and Value), offers a transformative solution. This study reviews 50 papers published between 2019 and 2025, demonstrating that leveraging large volumes of diverse data sources (Volume and Variety) with real-time processing (Velocity) and accurate data (Veracity) significantly enhances prediction and financial inclusion (Value). We found that big data reduces biases, refines risk profiles, and provides actionable insights, addressing the limitations of traditional models. The study also analyzes the interconnectedness of big data components, including data sources, infrastructure, and the 5Vs, emphasizing the challenges and opportunities. The paper concludes with recommendations for organizations of different sizes on implementing big data for credit risk assessment, highlighting a phased approach for adoption.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The impact of big data characteristics on credit risk assessment

  • 期刊论文

  • 发表日期：

    2025年3月19日

查看全文
  • Amin Karami,
  • Chukwuemeka Igbokwe

摘要

Credit risk assessment is critical for financial institutions’ stability and profitability. Traditional methods struggle to assess borrowers with limited credit history or non-traditional income due to insufficient data and outdated models. Big data analytics, characterized by the 5Vs (Volume, Variety, Velocity, Veracity, and Value), offers a transformative solution. This study reviews 50 papers published between 2019 and 2025, demonstrating that leveraging large volumes of diverse data sources (Volume and Variety) with real-time processing (Velocity) and accurate data (Veracity) significantly enhances prediction and financial inclusion (Value). We found that big data reduces biases, refines risk profiles, and provides actionable insights, addressing the limitations of traditional models. The study also analyzes the interconnectedness of big data components, including data sources, infrastructure, and the 5Vs, emphasizing the challenges and opportunities. The paper concludes with recommendations for organizations of different sizes on implementing big data for credit risk assessment, highlighting a phased approach for adoption.