Credit assessment method and system development for small-businesses' based on daily transaction data

由于目前已公布的信用评估模型和方法几乎都是基于财务数据或股价和调查数据，因此很难将其应用于不披露财务信息的中小企业的信用评估。本研究提出了一些新的方法和系统，只根据日常交易数据，如销售、客户付款、逾期付款金额等来评估客户的信用，并将所提出的方法和系统应用于实际的信用评估问题，验证了所提出的方法和系统的性能和有效性。所提出的方法和系统可以非常有效地识别不健康的客户，并且具有很高的识别健康客户的能力。

As almost all of models and methods for credit assessment published so far are based on financial data or stock price and investigation data, It is difficult to apply them to assess credit of minor and small businesses that do not disclose financial information. This study proposed some new approaches and systems to assess the customers' credit only based on daily transaction data such as sales, payments by customers, amount of overdue payment, etc. The performance and usefulness of the proposed approaches and systems has been confirmed by applying them to practical credit assessment problems. The proposed approaches and systems can indentify unsound customers very effectively and has also high ability to indentify healthy customers.