Digital Library[ Search Result ]
A Pretrained Model-Based Approach to Improve Generalization Performance for ADMET Prediction of Drug Candidates
http://doi.org/10.5626/JOK.2025.52.7.601
Accurate prediction of Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) properties plays an important role in reducing clinical trial failure rates and lowering drug development costs. In this study, we propose a novel method to improve ADMET prediction performance for drug candidate compounds by integrating molecular embeddings from a graph transformer model with pretrained embeddings from a UniMol model. The proposed model can capture bond type information from molecular graph structures, generating chemically refined representations, while leveraging UniMol’s pretrained 3D embeddings to effectively learn spatial molecular characteristics. Through this, the model is designed to address the problem of data scarcity and enhance the generalization performance. In this study, we conducted prediction experiments on 10 ADMET properties. The experiment results demonstrated that our proposed model outperformed existing methods and that the prediction accuracy for ADMET properties could be improved by effectively integrating atomic bond information and 3D structures.
An Efficient Continuous Subgraph Matching Technique for Graph Stream Processing in a Memory-constrained Environment
Somin Lee, Sanghyeuk Kim, Hyeonbyeong Lee, Dojin Choi, Jongtae Lim, Kyoungsoo Bok, Jaesoo Yoo
http://doi.org/10.5626/JOK.2022.49.12.1154
Recently, with the proliferation of social network services, the size of graph data has been becoming increasingly vast and graph data are changed in real-time. Therefore, it is necessary to perform continuous query processing on real-time graph streams. Moreover, it is difficult keep the entire large graph data in the main memory since its size is constrained in real-world application environments. Consequently, continuous subgraph matching techniques are required by considering memory-constrained environments. In this paper, we propose a continuous subgraph matching technique for graph streams in a memory-constrained environment. The proposed technique consists of modules such as index manager, query processor, and cache manager for efficient continuous subgraph matching. We conduct performance evaluations to demonstrate the superiority of the proposed technique.
Solving Korean Math Word Problems Using the Graph and Tree Structure
Kwang Ho Bae, Sang Yeop Yeo, Yu Chul Jung
http://doi.org/10.5626/JOK.2022.49.11.972
In previous studies, there have been various efforts to solve math word problems in the English sentence system. In many studies, improved performance was achieved by introducing structures such as trees and graphs, beyond the Sequence-to-Sequence approaches. However, in the study of solving math problems in Korean sentence systems, there are no model cases, using structures such as trees or graphs. Thus, in this paper, we examine the possibility of solving math problems in Korean sentence systems for models using the tree structure, graph structure, and Korean pre-training language models together. Our experimental results showed that accuracy improved by approximately 20%, compared to the model of the Seq2seq structure, by introducing the graph and tree structure. Additionally, the use of the Korean pre-training language model showed an accuracy improvement of 4.66%-5.96%.
Explainable Graph Neural Network for Medical Science Research
Yewon Shin, Kisung Moon, Youngsuk Jung, Sunyoung Kwon
http://doi.org/10.5626/JOK.2022.49.11.999
Explainable AI (XAI) is a technology that provides explainability for our end-users to comprehend prediction results of ML algorithms. In particular, the reliability of the decision-making process of an AI algorithm through XAI technology is the most critical in the medical field in terms of real applications. However, complex interaction-based medical data restrict the application of existing XAI technologies developed mostly for image or text data. Graph Neural Network (GNN)-based XAI research has been highlighted in recent years because GNN is technically specialized to capture complex relationships in data. In this paper, we proposed a taxonomy according to the application method and algorithm of GNN-based XAI technology with current XAI research trends and its use-cases in four detailed areas of the medical field. We also expounded on the technical limitations and future works of XAI research specialized in the biomedical area.
A Graph2Tree Model for Solving Korean Math Word Problems
Donggeun Kim, Nayeon Lee, Hyunwoo Sim, Myoung-Wan Koo
http://doi.org/10.5626/JOK.2022.49.10.807
This paper builds its own data set of eight types of Korean math word problems and presents a Ko-Graph2Tree model, an automatic solution model for Korean math word problems based on Graph2Tree model not previously presented. The recently released Graph2Tree model is a graph-to-tree learning based model that shows better performance than existing natural language processing models for automatic solving English math word problems. Using two types of graphs reflecting the relationship and order between numbers in the problem text, that is, mathematical relations, in solution generation, the model showed improved performance compared to existing tree-based models. As a result of measuring performance after learning with a self-produced Korean math word problem dataset, the transformer model with sequence-to-sequence structure showed an accuracy of 42.3%, whereas the Ko-Graph2Tree model showed an accuracy of 68.3%, resulting in 26.0%p higher performance.
Fast Personalized PageRank Computation on Very Large Graphs
Sungchan Park, Youna Kim, Sang-goo Lee
http://doi.org/10.5626/JOK.2022.49.10.859
Computation of Personalized PageRank (PPR) in graphs is an important function that is widely utilized in myriad application domains such as search, recommendation, and knowledge discovery. As the computation of PPR is an expensive process, a good number of innovative and efficient algorithms for computing PPR have been developed. However, efficient computation of PPR within very large graphs with over millions of nodes is still an open problem. Moreover, previously proposed algorithms cannot handle updates efficiently, thereby severely limiting their capability of handling dynamic graphs. In this paper, we present a fast converging algorithm that guarantees high and controlled precision. We attempted to improve the convergence rate of the traditional Power Iteration approximation methods and fully exact methods. The results revealed that the proposed algorithm is at least 20 times faster than the Power Iteration and outperforms other state-of-the-art algorithms in terms of computation time.
Proposal of a Graph Based Chat Message Analysis Model for Messenger User Verification
http://doi.org/10.5626/JOK.2022.49.9.696
As crimes and accidents through messengers increase, the necessity of verifying messenger users is emerging. In this study, two graph-based messenger user verification models that apply the traditional author verification problem to chat text were proposed. First, the graph random walk model builds an n-gram transition graph with a previous chat message and verifies the user by learning the characteristic of traversing the transition graph with a message whose author is unknown. The results showed an accuracy of 86% in 10,000 chat conversations. Second, the graph volume model verified the user using the characteristic that the size of the transition graph increased over time and achieved an accuracy of 87% in 1,000 chat conversations. When the density of the chat messages was calculated based on the transmission time, both graph models could guarantee more than 80% accuracy when the chat density was 15 or more.
Knowledge Graph Embedding with Entity Type Constraints
Seunghwan Kong, Chanyoung Chung, Suheon Ju, Joyce Jiyoung Whang
http://doi.org/10.5626/JOK.2022.49.9.773
Knowledge graph embedding represents entities and relationships in the feature space by utilizing the structural properties of the graph. Most knowledge graph embedding models rely only on the structural information to generate embeddings. However, some real-world knowledge graphs include additional information such as entity types. In this paper, we propose a knowledge graph embedding model by designing a loss function that reflects not only the structure of a knowledge graph but also the entity-type information. In addition, from the observation that certain type constraints exist on triplets based on their relations, we present a negative sampling technique considering the type constraints. We create the SMC data set, a knowledge graph with entity-type restrictions to evaluate our model. Experimental results show that our model outperforms the other baseline models.
Efficient Compositional Translation Embedding for Visual Relationship Detection
Yu-Jung Heo, Eun-Sol Kim, Woo Suk Choi, Kyoung-Woon On, Byoung-Tak Zhang
http://doi.org/10.5626/JOK.2022.49.7.544
Scene graphs are widely used to express high-order visual relationships between objects present in an image. To generate the scene graph automatically, we propose an algorithm that detects visual relationships between objects and predicts the relationship as a predicate. Inspired by the well-known knowledge graph embedding method TransR, we present the CompTransR algorithm that i) defines latent relational subspaces considering the compositional perspective of visual relationships and ii) encodes predicate representations by applying transitive constraints between the object representations in each subspace. Our proposed model not only reduces computational complexity but also outperformed previous state-of-the-art performance in predicate detection tasks in three benchmark datasets: VRD, VG200, and VrR-VG. We also showed that a scene graph could be applied to the image-caption retrieval task, which is one of the high-level visual reasoning tasks, and the scene graph generated by our model increased retrieval performance.
Knowledge Graph Embedding for Link Prediction using Node-Link Interaction-based Graph Attention Networks
http://doi.org/10.5626/JOK.2022.49.7.555
Knowledge graphs are structures that express knowledge in the real world in the form of nodes and links-based triple form. These knowledge graphs are incomplete and many embedding techniques have been studied to effectively represent nodes and links in low-dimensional vector spaces to find other missing relationships. Recently, many neural network-based knowledge graph link prediction methods have been studied. However existing models consider nodes and links independently when determining the importance of a triple to a node which makes it difficult to reflect the interaction between nodes and links. In this paper, we propose an embedding method that will be used to analyze the importance of triple units by simultaneously considering nodes and links using composition operators, and at the same time prove that the model outperforms other methods in knowledge graph link prediction.
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