Vol. 51, No. 10,
Oct. 2024
Digital Library
Overhead Analyses of Cache Replacement Policies and Region Mapping Replacement Policy
http://doi.org/10.5626/JOK.2024.51.10.849
Cache has been widely used to improve performance in systems with fast and slow devices. Various cache replacement policies have been studied, but these policies often come with computation and memory overheads. Unfortunately, many studies do not consider these overheads seriously and instead evaluate cache replacement policies based solely on cache hit rate. However, in modern computer systems, cache sizes are constantly increasing, making these overheads more significant. In order to provide a more comprehensive evaluation of cache replacement policies, we aim to consider both overheads and hit rates. In this study, we analyze the memory and computational overheads of popular cache replacement policies such as LRU, CLOCK, 2Q, ARC, and RAND. Additionally, we propose the Region Mapping (RM) policy, which has low memory and computational overheads. Furthermore, we introduce the RM2 policy, which improves hit rates by separating hot and cold data. Our experimental results show that the hit rates of the RM and RM2 policies are competitive with state-of-the-art policies. Moreover, policies with low memory overheads can reduce overall data access time by caching more data within a given cache size.
Technology and Applications of Character Recognition for Distance Information of Sewerage Surface Recording Videos
Jong-Rul Park, Min-Gyu Kong, Jungho Kim, Joung Woo Ryu
http://doi.org/10.5626/JOK.2024.51.10.862
Our research proposes a character recognition method for distance information overlaid with sewerage surface recorded clips. To implement the character recognition function on specified locations for each district in Seoul, Republic of Korea, our proposed training procedure of the character recognition model consists of captured image frames and types of fonts from sewerage surface recorded clips of each district in Seoul. This study investigated image processing application that minimizes interferences among character output areas, including overlaid areas between rows of character output areas and background images. The accuracy of character recognition regarding sewerage recorded clips from a total of 25 districts in Seoul was 99.2%, which was 38.9% higher than the accuracy evaluated by character recognition from an open-sourced library, and 31.5% higher than previous research held by city hall in Seoul.
Reliability Evaluation of Cross Domain Solutions based on Reliability and Security Metrics
Eunjeong Ju, Jeonghwa Lee, Duksan Ryu
http://doi.org/10.5626/JOK.2024.51.10.870
This research focuses on developing a comprehensive reliability assessment approach for software used in Industry Control Systems (ICS) environments, particularly targeting safety inspection devices for control protocol traffic. Given the vulnerability of bidirectional communication in these devices, a novel reliability evaluation method tailored to their characteristics is essential. This study identifies and analyzes software reliability and security metrics throughout the Software Development Life Cycle (SDLC) based on prior research, with a specific emphasis on analysis, design, and implementation stages. The proposed assessment approach aimed to effectively address security and reliability issues that might arise in environments with bidirectional communication, offering valuable contributions to the development of highly reliable software for systems utilizing control protocols.
Task-Oriented Dialogue System Using a Fusion Module between Knowledge Graphs
Jinyoung Kim, Hyunmook Cha, Youngjoong Ko
http://doi.org/10.5626/JOK.2024.51.10.882
The field of Task-Oriented Dialogue Systems focuses on using natural language processing to assist users in achieving specific tasks through conversation. Recently, transformer-based pre-trained language models have been employed to enhance performances of task-oriented dialogue systems. This paper proposes a response generation model based on Graph Attention Networks (GAT) to integrate external knowledge data into transformer-based language models for more specialized responses in dialogue systems. Additionally, we extend this research to incorporate information from multiple graphs, leveraging information from more than two graphs. We also collected and refined dialogue data based on music domain knowledge base to evaluate the proposed model. The collected dialogue dataset consisted of 2,076 dialogues and 226,823 triples. In experiments, the proposed model showed a performance improvement of 13.83%p in ROUGE-1, 8.26%p in ROUGE-2, and 13.5%p in ROUGE-L compared to the baseline KoBART model on the proposed dialogue dataset.
Evaluating Table QA with Generative Language Models
Kyungkoo Min, Jooyoung Choi, Myoseop Sim, Minjun Park, Stanley Jungkyu Choi
http://doi.org/10.5626/JOK.2024.51.10.892
Tables in documents can be described as collections of information that condense and aggregate important data. Research on table question answering techniques focusing on querying such tables is underway, with studies utilizing language models showing promising results. This study applied an emerging generative language model technology to table question answering, examining outcomes based on changes in the language model and prompts. It analyzed results using methods suitable for short-answer and generative outcomes. Applying our custom-developed EXAONE 1.7B model to the KorWiki dataset yielded an EM of 92.49 and an F1 score of 94.81. This demonstrates that fine-tuning smaller models can achieve better performance than larger models such as GPT-4. Additionally, the EXAONE 25B model exhibited the best performance among tested models on the KorQuAD2Table dataset.
Improving Retrieval Models through Reinforcement Learning with Feedback
Min-Taek Seo, Joon-Ho Lim, Tae-Hyeong Kim, Hwi-Jung Ryu, Du-Seong Chang, Seung-Hoon Na
http://doi.org/10.5626/JOK.2024.51.10.900
Open-domain question answering involves the process of retrieving clues through search to solve problems. In such tasks, it is crucial that the search model provides appropriate clues, as this directly impacts the final performance. Moreover, information retrieval is an important function frequently used in everyday life. This paper recognizes the significance of these challenges and aims to improve performances of search models. Just as the recent trend involves adjusting outputs in decoder models using Reinforcement Learning from Human Feedback (RLHF), this study seeks to enhance search models through the use of reinforcement learning. Specifically, we defined two rewards: the loss of the answer model and the similarity between the retrieved documents and the correct document. Based on these, we applied reinforcement learning to adjust the probability score of the top-ranked document in the search model's document probability distribution. Through this approach, we confirmed the generality of the reinforcement learning method and its potential for further performance improvements.
Generating Relation Descriptions with Large Language Model for Link Prediction
http://doi.org/10.5626/JOK.2024.51.10.908
The Knowledge Graph is a network consisting of entities and the relations between them. It is used for various natural language processing tasks. One specific task related to the Knowledge Graph is Knowledge Graph Completion, which involves reasoning with known facts in the graph and automatically inferring missing links. In order to tackle this task, studies have been conducted on both link prediction and relation prediction. Recently, there has been significant interest in a dual-encoder architecture that utilizes textual information. However, the dataset for link prediction only provides descriptions for entities, not for relations. As a result, the model heavily relies on descriptions for entities. To address this issue, we utilized a large language model called GPT-3.5-turbo to generate relation descriptions. This allows the baseline model to be trained with more comprehensive relation information. Moreover, the relation descriptions generated by our proposed method are expected to improve the performance of other language model-based link prediction models. The evaluation results for link prediction demonstrate that our proposed method outperforms the baseline model on various datasets, including Korean ConceptNet, WN18RR, FB15k-237, and YAGO3-10. Specifically, we observed improvements of 0.34%p, 0.11%p, 0.12%p, and 0.41%p in terms of Mean Reciprocal Rank (MRR), respecitvely.
Expected Addressee and Target Utterance Prediction for Construction of Multi-Party Dialogue Systems
Yoonjin Jang, Keunha Kim, Youngjoong Ko
http://doi.org/10.5626/JOK.2024.51.10.918
As the number of communication channels between people has increased in recent years, there has been a rise in both multi-party conversations and one-to-one conversations. Research on analyzing multi-party conversations has also been active. In the past, models for analyzing such dialogues typically predicted the addressee of the final response based on the previous responses. However, this differs from the task of generating multi-party dialogue responses, which requires the speaker to select the addressee to whom they will respond. In this paper, we propose a new task for predicting the addressee of a multi-party dialogue that does not rely on response information. Our task aims to predict and match the expected target utterance with the expected addressee in a real multi-party dialogue. To accomplish this, we introduce a model that uses a transform encoder-based masked token prediction learning method. This model predicts the expected target utterance and the expected addressee of the current speaker based on the previous dialogue context, without considering the final response. The proposed model achieves an accuracy of 82% in predicting the expected recipient and 68% in predicting the expected target utterance accuracy on the Ubuntu IRC dataset. These results demonstrate the potential of our model for use in a multi-party dialogue system, as it can accurately predict the target utterance that should be used. Moving forward, we plan to expand our research by creating additional datasets for multi-party dialogues and applying them to real-world multilateral dialogue response generation systems.
Cardiovascular Disease Prediction using Single-Lead ECG Data
Chaeyoon Park, Gihun Joo, Suhwan Ji, Junbeom Park, Junho Baek, Hyeonseung Im
http://doi.org/10.5626/JOK.2024.51.10.928
The most representative approach to diagnosing cardiovascular disease is to analyze electrocardiogram (ECG), and most ECG data measured in hospitals consist of 12 leads. However, wearable healthcare devices usually measure only single-lead ECG, which has limitations in diagnosing cardiovascular disease. Therefore, in this paper, we conducted a study to predict common cardiovascular diseases such as atrial fibrillation (AF), left bundle branch block (LBBB), and right bundle branch block (RBBB) using a single lead that could be measured with a wearable healthcare device. For experiments, we used a convolutional neural network model and measured its performance using various leads in terms of AUC and F1-score. For AF, LBBB, and RBBB, average AUC values were 0.966, 0.971, and 0.965, respectively, and average F1-scores were 0.867, 0.816, and 0.848, respectively. These experimental results confirm the possibility of diagnosing cardiovascular disease using only a single lead ECG that can be obtained with wearable healthcare devices.
Analysis of Vulnerabilities in Autonomous Driving Environments through Physical Adversarial Attacks Incorporating Natural Elements
Kyuchan Cho, Woosang Im, Sooyong Jeong, Hyunil Kim, Changho Seo
http://doi.org/10.5626/JOK.2024.51.10.935
Advancements in artificial intelligence technology have significantly impacted the field of computer vision. Concurrently, numerous vulnerabilities related to adversarial attacks, which are techniques designed to force models into misclassification, have been discovered. In particular, adversarial attacks such as physical adversarial attacks in the real world, pose a serious threat to autonomous vehicle systems. These attacks include artificially created attacks such as adversarial patches and attacks that exploit natural elements to cause misclassification. A common scenario in autonomous driving environments involves obstruction of traffic signs by natural elements such as fallen leaves or snow. These elements do not remain stationary. They can cause misclassification even in fleeting moments, highlighting a critical vulnerability. Therefore, this study investigated adversarial patch attacks based on natural elements, proposing fallen leaves as a natural adversarial element. Specifically, it reviewed current trends in adversarial attack research, presented an experimental environment based on natural elements, and analyzed experimental results to assess vulnerabilities posed by fallen leaves in physical environments to autonomous vehicles.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr