• Journal of information and communication convergence engineering
• /
• 제7권4호
• /
• pp.419-430
• /
• 2009

Reading comprehension (RC) systems are to understand a given text and return answers in response to questions about the text. Many previous studies extract sentences that are the most similar to questions as answers. However, texts for RC tests are generally short and facts about an event or entity are often expressed in multiple sentences. The answers for some questions might be indirectly presented in the sentences having few overlapping words with the questions. This paper proposes a conceptual graph matching method towards RC tests to extract answer strings. The method first represents the text and questions as conceptual graphs, and then extracts subgraphs for every candidate answer concept from the text graph. All candidate answer concepts will be scored and ranked according to the matching similarity between their sub-graphs and question graph. The top one will be returned as answer seed to form a concise answer string. Since the sub-graphs for candidate answer concepts are not restricted to only covering a single sentence, our approach improved the performance of answer extraction on the Remedia test data.

• KSBB Journal
• /
• 제27권2호
• /
• pp.121-126
• /
• 2012

In order to explore structural features of minoxidil analogs with a view of enhancing lysyl hydroxylase (LH) inhibitory activity, molecular holographic QSAR (HQSAR) and CoMSIA (comparative molecular similarity indices analysis) were performed. The results from the atomic contributions with optimized the HQSAR 6-2 model indicated that, in case of pyrimidine-1-N-oxide substituent, C2 atom of pyrimidine ring and C'3-C'4 bond of 4-piperidinol group showed the highest impact on the inhibitory activity towards LH enzyme. It was also evident from the information of the optimized CoMSIA F5 model that the inhibitory activity mainly depended on the hydrophobic field contribution (36%) and the hydrogen bond (H-bond) field contribution (49.2%) of substrate molecule. Particularly, it is predicted that the functional groups which disfavor H-bond acceptors in large space around the piperidinol group and also the functional groups which favor the H-bond acceptors at C'4 (& C'5) atom in $R_5$ group play a role for increased inhibitory activity. With this in mind, it is likely that a novel candidate having more improved inhibitory activity on hair growth could be designed in the future.

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권4호
• /
• pp.262-268
• /
• 2017

Intelligent video surveillance systems have been developed to monitor global areas and find specific target objects using a large-scale database. However, person re-identification presents some challenges, such as pose change and occlusions. To solve the problems, this paper presents an improved person re-identification method using sparse representation and saliency-based dictionary construction. The proposed method consists of three parts: i) feature description based on salient colors and textures for dictionary elements, ii) orthogonal atom selection using cosine similarity to deal with pose and viewpoint change, and iii) measurement of reconstruction error to rank the gallery corresponding a probe object. The proposed method provides good performance, since robust descriptors used as a dictionary atom are generated by weighting some salient features, and dictionary atoms are selected by reducing excessive redundancy causing low accuracy. Therefore, the proposed method can be applied in a large scale-database surveillance system to search for a specific object.

• Journal of Advanced Marine Engineering and Technology
• /
• 제40권8호
• /
• pp.726-732
• /
• 2016

Clustering analysis is widely used in data mining to classify data into categories on the basis of their similarity. Through the decades, many clustering techniques have been developed, including hierarchical and non-hierarchical algorithms. In gene profiling problems, because of the large number of genes and the complexity of biological networks, dimensionality reduction techniques are critical exploratory tools for clustering analysis of gene expression data. Recently, clustering analysis of applying dimensionality reduction techniques was also proposed. PCA (principal component analysis) is a popular methd of dimensionality reduction techniques for clustering problems. However, previous studies analyzed the performance of PCA for only full data sets. In this paper, to specifically and robustly evaluate the performance of PCA for clustering analysis, we exploit an improved FCBF (fast correlation-based filter) of feature selection methods for supervised clustering data sets, and employ two well-known clustering algorithms: k-means and k-medoids. Computational results from supervised data sets show that the performance of PCA is very poor for large-scale features.

• 한국인터넷방송통신학회논문지
• /
• 제11권6호
• /
• pp.283-288
• /
• 2011

본 연구에서는 객체 코너의 분산치에 기반한 코렐로그램 형태검출 기법을 제안한다. 제안된 알고리즘은 다음 단계로 진행된다. 먼저 영상 내 객체의 코너 점을 추출한 후 이들의 분산치를 구한다. 그리고 각각의 코너영역들의 분산치 중 최대/최소값을 추출한다. 그리고 이 최대/최소값을 이용하여 코렐로그램 매핑을 한 후 유사도를 측정하게 된다. 제안된 기법은 영상 내에서 형태 구조가 분명한 객체의 실험에서 성능이 우수하였으며 객체의 이동이나 회전에도 강인하였으며 코너 패치 히스토그램을 이용한 형태 검색에 비해 약 0.03%의 향상된 recall을 나타내었다.

• 대한수의학회지
• /
• 제35권2호
• /
• pp.307-315
• /
• 1995

The characterization of the MHC of domestic animals may constitute a first step towards increasing the efficiency of food production through improved disease resistance. In order to study the role of the MHC in regulating immune response it is first necessary to identify the different MHC alleles. In this research we try to investigate the possible associations between BoLA of Korean native cattles and infectious cattle disease. For this purpose we used one approach, serology. The results were summarized as follows : 1. Korean native cattle's lymphocyte reacted with alloantisera which recognized seven official BoLA allele. Korean native cattle's lymphocytes were reacted same as European breeds(especially with 673/3(W20)). 2. Korean native cattle's lymphocytes reacted with alloantisera 773/2, 673/3, 638/3, 773/3, 602/2, 639/2 and 639/3 at high reaction frequency. But alloantisera 642/1 was not expressed on Korean native cattle. If this allele, recognized by alloantisera(642/1), officially certificate In BoLA workshop it will be characterization factor of Korean native cattle. 3. According to cellular similarity index, we can presume on genetic relativity which has no family relationship.

• 한국측량학회지
• /
• 제6권1호
• /
• pp.35-41
• /
• 1988

지상사진에 의한 삼차원변형해석을 하는데 있어서 변위양계산의 정확도를 향상시키기 위해 반복경증률 상사변환법이 사용되었으며, 변위점검출에서는 Bayesian Inference가 적용되었고, 변위형태해석을 위해 변위방정식을 이용하는 방법을 제시하였다. 그 결과 변위양계산에서는 최소절대법($\Sigma$$\mid$d$\mid$⇒min)에 의한 경중률조건이 정확도를 향상시켰으며, 또한 Bayesian Inference을 적용하므로써 정확한 변위점검출을 할 수 있었다. 변위형태해석에서는 최적변위방정식을 이용하여 대상들의 전체 또는 부분적인 움직임을 해석할 수 있었다.

• 한국컴퓨터정보학회논문지
• /
• 제12권2호
• /
• pp.71-76
• /
• 2007

주어진 영상 내에서 의미 있는 객체의 존재여부를 판단하는 객체검출과정은 컴퓨터 비전 시스템을 이용한 영상인식의 핵심요소 가운데 하나이다. 템플릿 정합 방식의 객체 검출방식에서 사용되어 얼굴 인식 등에서 좋은 성능을 보이는 Hausdorff 거리는 점의 집합으로 표현된 두 객체 사이의 다른 정도를 정의함으로써 주어진 모델과 가장 차이가 적은 객체를 검출하도록 한다. 본 논문에서는 일반적으로 사용되는 바와 같이 점의 집합이 영상의 에지를 검출하여 얻어진 경우, 에지가 가지고 있는 방향정보를 이용함으로써 검출의 정확성을 높일 수 있는 방법을 제안하고 있다. 일반적으로 사용되는 Euclidean 거리 대신에 방향을 가진 두 점을 일치시키는데 필요한 탄성에너지를 정의함으로써 검출의 정확성을 높이는 방법을 제시하고 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권9호
• /
• pp.4567-4587
• /
• 2017

With the growth of the Internet and the extensive applications of image editing software, it has become easier to manipulate digital images without leaving obvious traces. Copy-move is one of the most common techniques for image forgery. Image blind forensics is an effective technique for detecting tampered images. This paper proposes an improved copy-move forgery detection method based on the discrete cosine transform (DCT). The quantized DCT coefficients, which are feature representations of image blocks, are truncated using a truncation factor to reduce the feature dimensions. A method for judging whether two image blocks are similar is proposed to improve the accuracy of similarity judgments. The main transfer vectors whose frequencies exceed a threshold are found to locate the copied and pasted regions in forged images. Several experiments are conducted to test the practicability of the proposed algorithm using images from copy-move databases and to evaluate its robustness against post-processing methods such as additive white Gaussian noise (AWGN), Gaussian blurring, and JPEG compression. The results of experiments show that the proposed scheme effectively detects both copied region and pasted region of forged images and that it is robust to the post-processing methods mentioned above.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권7호
• /
• pp.2785-2799
• /
• 2020

We propose a novel speaker adaptation method using acoustic model clustering. The similarity of different speakers is defined by the cosine distance between their i-vectors (intermediate vectors), and various efficient clustering algorithms are applied to obtain a number of speaker subsets with different characteristics. The speaker-independent model is then retrained with the training data of the individual speaker subsets grouped by the clustering results, and an unknown speech is recognized by the retrained model of the closest cluster. The proposed method is applied to a large-scale speech recognition system implemented by a hybrid hidden Markov model and deep neural network framework. An experiment was conducted to evaluate the word error rates using Resource Management database. When the proposed speaker adaptation method using i-vector based clustering was applied, the performance, as compared to that of the conventional speaker-independent speech recognition model, was improved relatively by as much as 12.2% for the conventional fully neural network, and by as much as 10.5% for the bidirectional long short-term memory.