• Journal of the Korea Convergence Society
• /
• v.11 no.6
• /
• pp.23-28
• /
• 2020

A comment system is essential for communication on the Internet. However, there are also malicious comments such as inappropriate expression of others by exploiting anonymity online. In order to protect users from malicious comments, classification of malicious / normal comments is necessary, and this can be implemented as text classification. Text classification is one of the important topics in natural language processing, and studies using pre-trained models such as BERT and graph structures such as GCN and GAT have been actively conducted. In this study, we implemented a comment classification system using BERT, GCN, and GAT for actual published comments and compared the performance. In this study, the system using the graph-based model showed higher performance than the BERT.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3383-3397
• /
• 2023

Scene graphs serve as semantic abstractions of images and play a crucial role in enhancing visual comprehension and reasoning. However, the performance of Scene Graph Generation is often compromised when working with biased data in real-world situations. While many existing systems focus on a single stage of learning for both feature extraction and classification, some employ Class-Balancing strategies, such as Re-weighting, Data Resampling, and Transfer Learning from head to tail. In this paper, we propose a novel approach that decouples the feature extraction and classification phases of the scene graph generation process. For feature extraction, we leverage a transformer-based architecture and design an adaptive calibration function specifically for predicate classification. This function enables us to dynamically adjust the classification scores for each predicate category. Additionally, we introduce a Distribution Alignment technique that effectively balances the class distribution after the feature extraction phase reaches a stable state, thereby facilitating the retraining of the classification head. Importantly, our Distribution Alignment strategy is model-independent and does not require additional supervision, making it applicable to a wide range of SGG models. Using the scene graph diagnostic toolkit on Visual Genome and several popular models, we achieved significant improvements over the previous state-of-the-art methods with our model. Compared to the TDE model, our model improved mR@100 by 70.5% for PredCls, by 84.0% for SGCls, and by 97.6% for SGDet tasks.

• Tunnel and Underground Space
• /
• v.15 no.5 s.58
• /
• pp.369-377
• /
• 2005

The stability of underground openings is a major concern for the safety and productivity of mining operations. Rock mass classification methods provide the basis of many empirical design methods as well as a basis for numerical analysis. Of the many factors which influence the stability of openings, the span of the opening for a given rock mass condition provides an important parameter of design. In this paper, the critical span curves proposed by Lang, the Mathews stability graph method and the modified critical span curve suggested by the authors have been assessed. The modified critical span curve was proposed by using Mathews stability graph method. The modified critical span curve by the author have been used to assess the stability of underground openings in several limestone mines.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.615-621
• /
• 2020

In this paper, we propose a multi-site based earthquake event classification method using graph convolution networks. In the traditional earthquake event classification methods using deep learning, they used single-site observation to estimate seismic event class. However, to achieve robust and accurate earthquake event classification on the seismic observation network, the method using the information from the multi-site observations is needed, instead of using only single-site data. Firstly, our proposed model employs convolution neural networks to extract informative embedding features from the single-site observation. Secondly, graph convolution networks are used to integrate the features from several stations. To evaluate our model, we explore the model structure and the number of stations for ablation study. Finally, our multi-site based model outperforms up to 10 % accuracy and event recall rate compared to single-site based model.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.3
• /
• pp.461-470
• /
• 2010

Semi supervised classification which is a method using labeled and unlabeled data has considerable attention in recent years. Among various methods the graph based manifold regularization is proved to be an attractive method. Least squares support vector machine is gaining a lot of popularities in analyzing nonlinear data. We propose a semi supervised classification algorithm using the least squares support vector machines. The proposed algorithm is based on the manifold regularization. In this paper we show that the proposed method can use unlabeled data efficiently.

• Korean Journal of Remote Sensing
• /
• v.27 no.6
• /
• pp.717-728
• /
• 2011

An efficient method for the unsupervised classification of high resolution imagery is suggested in this paper. It employs pixel-linking and merging based on the adjacency graph. The proposed algorithm uses the neighbor lines of 8 directions to include information in spatial proximity. Two approaches are suggested to employ neighbor lines in the linking. One is to compute the dissimilarity measure for the pixel-linking using information from the best lines with the smallest non. The other is to select the best directions for the dissimilarity measure by comparing the non-homogeneity of each line in the same direction of two adjacent pixels. The resultant partition of pixel-linking is segmented and classified by the merging based on the regional and spectral adjacency graphs. This study performed extensive experiments using simulation data and a real high resolution data of IKONOS. The experimental results show that the new approach proposed in this study is quite effective to provide segments of high quality for object-based analysis and proper land-cover map for high resolution imagery of urban area.

• Science of Emotion and Sensibility
• /
• v.18 no.1
• /
• pp.103-110
• /
• 2015

Sentiment lexicon is an essential element for expressing sentiment on a text or recognizing sentiment from a text. We propose a graph-based semi-supervised learning method to construct a sentiment dictionary as sentiment lexicon set. In particular, we focus on the construction of domain-specific sentiment dictionary. The proposed method makes up a graph according to lexicons and proximity among lexicons, and sentiments of some lexicons which already know their sentiment values are propagated throughout all of the lexicons on the graph. There are two typical types of the sentiment lexicon, sentiment words and sentiment phrase, and we construct a sentiment dictionary by creating each graph of them and infer sentiment of all sentiment lexicons. In order to verify our proposed method, we constructed a sentiment dictionary specific to the movie domain, and conducted sentiment classification experiments with it. As a result, it have been shown that the classification performance using the sentiment dictionary is better than the other using typical general-purpose sentiment dictionary.

• Journal of Information Processing Systems
• /
• v.14 no.3
• /
• pp.709-726
• /
• 2018

Database classification is an important preprocessing step for the multi-database mining (MDM). In fact, when a multi-branch company needs to explore its distributed data for decision making, it is imperative to classify these multiple databases into similar clusters before analyzing the data. To search for the best classification of a set of n databases, existing algorithms generate from 1 to ($n^2-n$)/2 candidate classifications. Although each candidate classification is included in the next one (i.e., clusters in the current classification are subsets of clusters in the next classification), existing algorithms generate each classification independently, that is, without taking into account the use of clusters from the previous classification. Consequently, existing algorithms are time consuming, especially when the number of candidate classifications increases. To overcome the latter problem, we propose in this paper an efficient approach that represents the problem of classifying the multiple databases as a problem of identifying the connected components of an undirected weighted graph. Theoretical analysis and experiments on public databases confirm the efficiency of our algorithm against existing works and that it overcomes the problem of increase in the execution time.

• Journal of the Korean Mathematical Society
• /
• v.54 no.1
• /
• pp.227-248
• /
• 2017

In this paper, we reduce the classification problem of equivariant (topological complex) vector bundles over a simple graph to the classification problem of their isotropy representations at vertices and midpoints of edges. Then, we solve the reduced problem in the case when the simple graph is homeomorphic to a circle. So, the paper could be considered as a generalization of [3].

• Bulletin of the Korean Mathematical Society
• /
• v.59 no.3
• /
• pp.567-608
• /
• 2022

We show that certain extensions of classifiable C^*-algebras are strongly classified by the associated six-term exact sequence in K-theory together with the positive cone of K₀-groups of the ideal and quotient. We use our results to completely classify all unital graph C^*-algebras with exactly one non-trivial ideal.