House loans decisive factors (or features) comparison: helping customers get loans

Abstract

This article provides a method, system, and related equipment for comparing decisive factors of house loans to assist consumers in securing loans. We chose the data that includes many factors about personal such as marriage/single, income. The study was conducted applying decision tree, neural network, and logic regression. The method includes obtaining consumer financial data; segmenting this data into multiple overlapping financial facets; generating a preliminary financial profile from these facets; performing noise reduction on the preliminary profile to create a detailed financial model; slicing the detailed model into various financial segments; and analyzing these segments to derive a multidimensional understanding of key house loan factors. This approach minimizes both explicit and implicit risks, ensuring a precise alignment with consumer financial profiles. Additionally, converting three-dimensional financial data into two-dimensional segments reduces computational complexity and errors, ensuring overall accuracy in loan evaluation and decision-making. In the last, we use these models to compare which factor influence the risk of loan mostly, so we advise if customers do not have house or car, keep working in one field and carefully view different places’ loan policies still can be helpful for their loaning.