• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.11 no.4
• /
• pp.238-246
• /
• 2011

This paper proposes a new semantic representation and its associated similarity measure. The representation expresses textual context observed in a context of a certain term as a network where nodes are terms and edges are the number of cooccurrences between connected terms. To compare terms represented in networks, a graph kernel is adopted as a similarity measure. The proposed representation has two notable merits compared with previous semantic representations. First, it can process polysemous words in a better way than a vector representation. A network of a polysemous term is regarded as a combination of sub-networks that represent senses and the appropriate sub-network is identified by context before compared by the kernel. Second, the representation permits not only words but also senses or contexts to be represented directly from corresponding set of terms. The validity of the representation and its similarity measure is evaluated with two tasks: synonym test and unsupervised word sense disambiguation. The method performed well and could compete with the state-of-the-art unsupervised methods.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.295-300
• /
• 2024

This paper discusses the incompleteness of the SNOMED CT and proposes a noble metric which evaluates similarity among sibling nodes as a method to address this incompleteness. SNOMED CT encompasses an extensive range of medical terms, but it faces issues of ontology incompleteness, such as missing concepts in the hierarchy. We propose a noble metric for evaluating similarity among nodes within a node group, composed of multiple sibling nodes, to identify missing concepts, and identify groups with low similarity. Analyzing the similarity of sibling node groups in the March 2023 international release of SNOMED CT, the average similarity of 29,199 sibling node groups, which are sub-concepts of the clinical finding concept and are consist of two or more sibling nodes, was found to be 0.81. The group with the lowest similarity was associated with child concepts of poisoning, with a similarity of 0.0036.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.3
• /
• pp.1330-1336
• /
• 2012

The Image retrieval has been study different approaches which are text-based, contents-based, area-based method and sub-image finding. The sub-image retrieval is to find a query image in the target one. In this paper, we propose a novel sub-image retrieval algorithm by Dot-Matrix method to be used in the bioinformatics. Dot-Matrix is a method to evaluate similarity between two sequences and we redefine the problem for retrieval of sub-image to the finding similarity of two images. For the approach, the 2 dimensional array of image converts a the vector which has gray-scale value. The 2 converted images align by dot-matrix and the result shows candidate sub-images. We used 10 images as target and 5 queries: duplicated, small scaled, and large scaled images included x-axes and y-axes scaled one for experiment.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.34 no.2
• /
• pp.174-184
• /
• 1998

In order to accumulate fundamental. data for control of fishes’ behavior at the real fishing ground, the fitness of the numerical model for describing the behavior of fishes was examined by the marine fish. Mullet, Mugil cephalus were used as experimental fishes. The numerical model of fishes’ behavior presented in our earlier paper was modified on the vertical movement of fish school. For the comparision of parameters of the modified numerical model between mullet and rainbow trout, the estimated values of parameters were identified with dimension. The fitness of the modified numerical model was examined by the comparision between experiment and simulation on the several indexes represented by fishes’ swimming characteristics. The obtained result are summarized a follows : 1. The non-dimensional parameter a’ of propulsive force and kb’ of interactive force by the experiment without model net showed a similarity, but the non-dimensional parameter k sub(c’) of schooling force for rainbow trout was lager than one for mullet and the non-dimensional parameter k sub(w’) of repulsive force for mullet was lager than one for rainbow trout. 2. The non-dimensional parameter a’ and k sub(b’) for rainbow trout by the experiment with model net were a little lager than ones for mullet, but non-dimensional parameter k sub(c’) and k sub(w’) for mullet were lager than ones for rainbow trout. 3. The non-dimensional parameter k sub(c’) and k sub(b’) showed the largest and the smallest value among the non-dimensional parameters for rainbow trout and mullet, respectively. 4. The fitness of the modified numerical model was confirmed by means of the compulsion between experiment and simulation on the swimming trajectory of fishes, the mean distance of individual from wall, the mean swimming speed, the mean swimming depth and the mean distance between the nearest individuals. Especially, the similarity of mean swimming depth was improved by using the modified numerical model.

• Journal of the Chungcheong Mathematical Society
• /
• v.29 no.4
• /
• pp.585-598
• /
• 2016

This paper shows that the solutions to the nonlinear perturbed differential system $$y^{\prime}=f(t,y)+{\int_{t_0}^{t}}g(s,y(s),T_1y(s))ds+h(t,y(t),T_2y(t))$$, have bounded properties. To show these properties, we impose conditions on the perturbed part ${\int_{t_0}^{t}}g(s,y(s),T_1y(s))ds,h(t,y(t),T_2y(t))$, and on the fundamental matrix of the unperturbed system y' = f(t, y) using the notion of h-stability.

• Journal of the Chungcheong Mathematical Society
• /
• v.30 no.3
• /
• pp.291-304
• /
• 2017

This paper shows that the solutions to the perturbed differential system $$y^{\prime}=f(t,y)+{{\displaystyle\smashmargin{2}{\int\nolimits_{t_0}}^{t}}g(s,y(s),T_1y(s))ds+h(t,y(t),T_2y(t))$$, have bounded properties by imposing conditions on the perturbed part ${\int}_{t_0}^{t}g(s,y(s),T_1y(s))ds,h(t,y(t),T_2y(t))$, and on the fundamental matrix of the unperturbed system y' = f(t, y) using the notion of h-stability.

• The KIPS Transactions:PartB
• /
• v.12B no.2 s.98
• /
• pp.115-122
• /
• 2005

The sub-mage matching problem in which one image contains some part of the other image, has been mostly investigated on natural images. In this paper, we propose two sub-image matching techniques: mesh-based method and correlation-based method, that are efficiently used to match text images. Mesh-based method consists of two stages, box alignment and similarity measurement by extracting the mesh feature from the two images. Correlation-based method determines the similarity using the correlation of the two images based on FFT function. We have applied the two methods to the text verification in a postal automation system and observed that the accuracy of correlation-based method is $92.7\%$ while that of mesh-based method is $90.1\%$.

• Journal of Digital Convergence
• /
• v.11 no.2
• /
• pp.299-307
• /
• 2013

The aim of this paper is the improvement of identification performance and the reduction of computational quantities in the face recognition system based on SIFT-Grid. Firstly, we propose a composition method of integrated template by removing similar SIFT keypoints and blending different keypoints in variety training images of one face class. The integrated template is made up of computation of similarity matrix and threshold-based histogram from keypoints in a same sub-region which divided by applying SIFT-Grid of training images. Secondly, we propose a computation method of similarity for identify of test image from composed integrated templates efficiently. The computation of similarity is performed that a test image to compare one-on-one with the integrated template of each face class. Then, a similarity score and a threshold-voting score calculates according to each sub-region. In the experimental results of face recognition tasks, the proposed methods is founded to be more accurate than both two other methods based on SIFT-Grid, also the computational quantities are reduce.

• The Korean Journal of Pesticide Science
• /
• v.7 no.2
• /
• pp.100-107
• /
• 2003

To improve the growth inhibitory activity against the shoot and root of rice plant (Oryza sativa L) and barnyard grass (Echinochloa crus-galli), a series of N-substituted phenyl-3,4-dimethylmaleimdes derivatives as substrates were synthesized and then their the inhibitory activities of protoporphyrinogen oxidase (1.3.3.4), protox were measured. The quantitative structure-activity relationships (QSAR) between structures and the inhibitory activities were studied quantitatively using the 2D-QSAR method. And also, molecular sharp similarity between the substrate derivatives and protogen, substrare of protox enzyme were studied. The activities of the two plants indicated that barnyard grass had a higher activity than the rice plant and their correlation relationships have shown in proportion for each. Accordingly, the results of SARs suggest that the electron donating groups as $R_2=Sub.X$ group will bind to phenyl ring because the bigger surface area of negative charged atoms in the substrate molecule derivatives may increase to the higher the activity against barnyard grass. Based on the molecular shape similarity, when the derivatives and protogen, subsbrate of protox enzyme were superimposed by atom fitting, the similarity indices (S) were above 0.8 level but the correlation coefficients (r) between S values and the activities showed not good.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.432-436
• /
• 2010

Finding communities from network data including social networks can be done by clustering the nodes of the network as densely interconnected groups, where keeping interconnection between groups sparse. To exploit a clustering algorithm for community detection task, we need a well-defined similarity measure between network nodes. In this paper, we propose a new similarity measure named "Common Neighborhood Sub-graph density" and combine the similarity with affinity propagation, which is a recently devised clustering algorithm.