{"title":"A Method Mining Correlation High Utility Itemsets with Negative Profits","authors":"Tung Anh Cao","doi":"10.32913/mic-ict-research.v2023.n2.1228","DOIUrl":"https://doi.org/10.32913/mic-ict-research.v2023.n2.1228","url":null,"abstract":"High-utility itemset (HUI) mining is an important task in the data mining process. Recently, many algorithms have been proposed to mine sets of HUIs. Most algorithms only work with data sets with positive return values. However, in the real world, the set of business items often includes both positive (+) and negative (-) profit values. To solve this problem, an algorithm is needed to find HUIs with negative returns in the database.The issue is how to mine efficiently correlation high utility itemsets with negative profits. Given the transactional database D and the profit table of products, how to find the best correlation high utility itemsets on the database with items having negative profits. To address these issues, the paper proposes a COHUIs_CoHUN method to mine correlation high utility itemsets with negative profits","PeriodicalId":496983,"journal":{"name":"Journal on Information Technologies & Communications","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135304884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Proposals and Implementations of MDS Diffusion Layer Dynamic Algorithms for AES Block Cipher","authors":"Trần Thị Lượng","doi":"10.32913/mic-ict-research.v2023.n2.1194","DOIUrl":"https://doi.org/10.32913/mic-ict-research.v2023.n2.1194","url":null,"abstract":"Block ciphers are an interesting cryptographictopic that is widely used today. Therefore, the issue of improving the security of block ciphers is increasingly concerned today. There have been many research directions to animate block ciphers to improve their security against well-known strong attacks. In [1], we proposed two diffusion layer dynamic algorithms for SPN block ciphers. In this paper, we will execute these two algorithms with recursive MDS matrices of size 4, 8, and 16, then evaluate the cryptographic properties of the obtained dynamic MDS matrices including branch number, number of fixed points, coefficient of fixed points, number of XOR and Xtime operations. Then, we apply these two dynamic algorithms to dynamically modify the AES block cipher and evaluate the execution speed of thedynamic AES block ciphers. We propose two new diffusionlayer dynamic algorithms, implement these two algorithmsand modify dynamically AES block cipher according to thesetwo algorithms, and evaluate the speed of dynamic AES blockciphers. We also evaluate the security and implementationresources for dynamically modified AES block ciphers. Theproposed diffusion layer dynamic algorithms contribute toimproving the security of the AES block cipher against manycurrent strong attacks.","PeriodicalId":496983,"journal":{"name":"Journal on Information Technologies & Communications","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135752209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ngoc-Hung Nguyen, Kinh-Tai Le, Tran-Duc Dang, Duc-Diep Can, Minh-Thai Pham
{"title":"An Improvemant of the IGEPVO-k Reversible Data Hiding Method","authors":"Ngoc-Hung Nguyen, Kinh-Tai Le, Tran-Duc Dang, Duc-Diep Can, Minh-Thai Pham","doi":"10.32913/mic-ict-research.v2023.n2.1207","DOIUrl":"https://doi.org/10.32913/mic-ict-research.v2023.n2.1207","url":null,"abstract":"PVO-K method is an expansion of PVO method in which all ???? largest (???? smallest pixels) are used as a single unitto embed a bit. In PVOK-method, each pixel of an original image is modified by one at most. GePVO-K method is again a development of the PVOK-method by embedding k bits in k largest pixels and ???? bits in ???? smallest pixels. However, this method has two drawbacks. First, each pixel of the original image is changed by two at most, so marked image quality is not good. Second, the location map consisting of two bits for each pixel is so large that in some cases compression code length of the map is greater method’s embedding capacity, and then data embedding capacity is zero. Both drawbacks are overcome in the iGePVO-K method. Specifically, in this method, each pixel only is changed by one at most and the two-bit location map as in iGePVO-K method is replaced bya one-bit map and a sequence of binary flags. This paper is a development of the iGePVO-K method by discarding the sequence of binary flags. That means it is not necessary to embed flags with the data in the original image. Therefore, the proposed method not only has a larger data embedding capacity but also has a better image quality than the methods mentioned above.","PeriodicalId":496983,"journal":{"name":"Journal on Information Technologies & Communications","volume":"280 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135842902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Proposing of Imaging Graph Neural Network with Defined Security Pattern for Improving Smart Contract Vulnerability Detection","authors":"None Pham Trong Linh, Tạ Minh Thanh","doi":"10.32913/mic-ict-research.v2023.n2.1198","DOIUrl":"https://doi.org/10.32913/mic-ict-research.v2023.n2.1198","url":null,"abstract":"A smart contract is a special set of protocols based on blockchain technology to implement the terms oragreements between the parties in the contract. Lots of smart contracts are built and deployed everyday. However, to ensure the safety of smart contracts is still a big challenge. Smart contracts are built to carry out transactions directly related to cryptocurrencies, therefore, the loss of security of smart contracts leads to huge financial losses. Common methods being used to check and to verify smart contract security are heavily dependent on hard rules defined by experts, leading to low detection accuracy and non-scalable, which can be bypassed by experienced attackers. In this paper, we propose to use the combination of Imaging Graph Neural Network With Defined Pattern to detect vulnerabilities in smart contracts. We construct a contract graph that shows the relationship between the main components in a smart contract. Then we extract graph features from normalized graphs, and combine graph features with defined security patterns to create combined features. Finally, we implemented normalization to gray scale image and feed it to the Convolutional Neural Network (CNN) to learn for vulnerability detection. Results show significantly improved accuracy compared to previous methods or other models. Specifically, 96,42%, 90,12%, 79% for reentrancy, timestamp dependence and infinite loop","PeriodicalId":496983,"journal":{"name":"Journal on Information Technologies & Communications","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135842901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Evaluation of Autoencoder-based Communications with Reconfigurable Intelligent Surfaces","authors":"Thi-Nga Dao, Chi-Hieu Ta","doi":"10.32913/mic-ict-research.v2023.n2.1204","DOIUrl":"https://doi.org/10.32913/mic-ict-research.v2023.n2.1204","url":null,"abstract":"Reconfigurable intelligent surfaces (RIS) have emerged as a promising technique for wireless communication in 5G and beyond. In an environment with lots of physical obstacles, RIS provides an alternative solution to the coverage problem by beam-forming signal towards the desired direction of the receiver. Due to the fixed constellation diagram, traditional modulation schemes cannot be used for the RIS-aided communication system with the receiver movement. In this work, we aim to design and evaluate an autoencoder-based communication model with RIS support, which can adapt to changes in the environment and the receiver’s position. Specifically, we use neural networks to present the encoder and decoder of the system and the parameters of these networks are trained to minimize the reconstruction error at the receiver. The simulation setup is based on characteristics of the environment and real RIS. Performance results show the benefits of RIS when a physical obstruction is present between the transmitter and receiver. Moreover, we prove that the selection of RIS codeword plays an important role in system performance","PeriodicalId":496983,"journal":{"name":"Journal on Information Technologies & Communications","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135910490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research on Deep Learning and Its Application in Stock Price Prediction","authors":"Van Hai Hoang","doi":"10.32913/mic-ict-research.v2023.n2.1136","DOIUrl":"https://doi.org/10.32913/mic-ict-research.v2023.n2.1136","url":null,"abstract":"The article studies the problem of forecasting the closing price of a stock based on historical data of a previousday. The paper uses and compares algorithms based on deep learning such as LSTM, BiLSTM, and CNN. The dataset includes data on price, trading volume and some technical indicators related to VCB, MSN, and HPG shares. The results show that CNN performs better for predicting the next day’s closing price than the other architectures.","PeriodicalId":496983,"journal":{"name":"Journal on Information Technologies & Communications","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135239278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}