2022, Vol.27, No.11

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

pISSN : 1598-849X / eISSN : 2383-9945
https://journal.kci.go.kr/jksci

In this paper, we propose an NFT(Non-Fungible Token)-based access control method for safely sharing data between users in blockchain environment. Since all data stored in the blockchain can be accessed by anyone due to the nature of the technology, it is necessary to control access except for authorized users when sharing sensitive data. For that, we generate each data as NFT and controls access to the data through the smart contract. In addition, in order to overcome the limitations of single ownership of the existing NFT, we separated the NFT into ownership and use rights, so that data can be safely shared between users. Ownership is represented as an original NFT, use rights is represented as a copied NFT, and all data generated as NFT is encrypted and uploaded, so data can be shared only through the smart contract with access control. To verify this approach, we set up a hypothetical scenario called Building Information Modeling (BIM) data trade, and deployed a smart contract that satisfies 32 function call scenarios that require access control. Also, we evaluated the stability in consideration of the possibility of decryption through brute-force attack. Through our approach, we confirmed that the data can be safely shared between users in blockchain environment.

In this paper, we proposed VW-FedAVG(Validation based Weighted FedAVG) which updates the global model by weighting according to performance verification from the models of each device participating in the training. The first method is designed to validate each local client model through validation dataset before updating the global model with a server side validation structure. The second is a client-side validation structure, which is designed in such a way that the validation data set is evenly distributed to each client and the global model is after validation. MNIST, CIFAR-10 is used, and the IID, Non-IID distribution for image classification obtained higher accuracy than previous studies.

In this study, we propose a language and platform to describe and manage the MLOps(Machine Learning Operations) workflow for time series data anomaly detection. Time series data is collected in many fields, such as IoT sensors, system performance indicators, and user access. In addition, it is used in many applications such as system monitoring and anomaly detection. In order to perform prediction and anomaly detection of time series data, the MLOps platform that can quickly and flexibly apply the analyzed model to the production environment is required. Thus, we developed Python-based AI/ML Modeling Language (AMML) to easily configure and execute MLOps workflows. Python is widely used in data analysis. The proposed MLOps platform can extract and preprocess time series data from various data sources (R-DB, NoSql DB, Log File, etc.) using AMML and predict it through a deep learning model. To verify the applicability of AMML, the workflow for generating a transformer oil temperature prediction deep learning model was configured with AMML and it was confirmed that the training was performed normally.