Research On Blockchain Association Traceability Method For Complex Financial Events

With the development of information technology,blockchain technology has been applied in more and more fields,such as finance,government affairs,digital currency,education and so on.Especially in the financial field,it is of extensive and far-reaching significance to classify and store massive financial data and find out the events with great potential value in promoting the integration of information technology and the healthy development of digital economy.Financial time series data has the characteristics of large scale,complex scenes,multiple types,easy to tamper with,and difficult to trace back to the source.Therefore,in recent years,efficient storage and traceability of financial data through blockchain technology has aroused wide discussion in the academic world,and also become the focus and difficulty of research in the field.As a shared database,the data or information stored in blockchain has the characteristics of non-forgery,whole-process traces,traceability,openness and transparency,and collective maintenance.Based on these characteristics,blockchain technology has laid the foundation of "trust" and "consensus".However,the existing financial blockchain mostly adopts the equal mining mode,in which all bookkeepers(entities)record the ledger in a single main chain.The data storage is random,and the main chain data is difficult to achieve correlation or regular storage in complex or classified financial scenarios,resulting in low efficiency of storage and traceability.At the same time,most of the event traceability in the existing blockchain systems can only be queried to the source block,and the implicit association between financial entities cannot be identified,which is not conducive to financial risk assessment.In conclusion,this paper conducts in-depth research on the storage and traceability of financial blockchain,and proposes a blockchain association traceability method for complex financial events in view of the shortcomings of existing systems and methods.The main research work and innovations of this paper are as follows:(1)In view of the existing financial blockchain system is mostly single chain structure,which cannot be applied to complex classification scenarios,the blockchain composite chain model of complex financial events is proposed.Firstly,combining the different application modes of blockchain and the characteristics of the data of financial entities that need to be stored,the alliance chain between different entities is built,and the private chain is built within the entity.Secondly,the block structure of private chain based on the Merkle tree and the block structure of alliance chain based on the Merkle Patricia tree are established respectively.Finally,the data encryption mechanism at the data layer,the block communication mode at the network layer and the block consensus algorithm at the consensus layer of the blockchain composite chain model are presented to realize the storage of financial entity data and prepare for the follow-up tracing work.(2)In view of the problem that the existing single chain mode cannot be associated with traceability,the associated entity traceability method of blockchain events is proposed.Firstly,using the traceable characteristics of blockchain itself,it can trace out the transaction data of the source entity block with problems in complex financial events.Secondly,it is proposed to introduce the Apriori association rule algorithm to trace the transaction data of other entities blocks associated with the source entity with problems.Finally,the source event association diagram is constructed for all the traced entity block data,and the association relationship between entities in complex financial events is expressed through the source event association diagram.(3)On this basis,in order to further reduce the adverse impact of complex financial events,the block data of all entities in the source event association diagram are represented as feature vectors in numerical form,and a risk assessment model based on reinforcement learning is constructed for the traceability entities to realize the risk assessment of the traceability entities.(4)Conduct a lot of experiments on multiple real data sets,and compare with several major blockchain storage and traceability methods.The experimental results show that the proposed method has obvious advantages in storage cost,storage efficiency,traceability accuracy and traceability efficiency,and can effectively store and trace financial data with high quality.