Today's world places more emphasis on smart device transactions for the financial sector, which configures a large number of programs that are capable of processing massive volumes of data effectively in response to the expanding requirement for service facilities from network edges. These enormous increases have led to addressing issues with the security of the system and the functionality of smart devices regarding critical transaction records for financial transactions. Service centers currently exchange transaction records across many platforms, recognition to blockchain technology. In this paper, research developed a Distributed Blockchain-learning cloud (DistB-Learning Cloud) architecture to address these problems for financial sectors. It combines blockchain technology, properties of cloud computing and machine learning (ML) to provide secure information transfers in peer-to peer systems and efficient data-sharing services. There are four design models in this methodology. Initially, an attack detection approach uses a Support Vector Machine (SVM), using a transaction network-based anomalous traffic detector to locate the attack in the financial sector. Second, to manage risks and confirm the process of identity verification for secure transactions, a novel model for a blockchain transaction network is created using a CryptoAware Elliptic Curve (CAEC) and encryption. The ML model is then trained for the output prediction by logistic regression (LR) of the large-scale transaction record. In conclusion, a cloud assessment model facilitates the management of transaction records that are maintained and accessible sharing of services that each service center has access to several cloud platforms. The results of the experiments show that the DistB-learningCloud performs higher than existing schemes in terms of achieving more transactions in each block for the financial sector.