• 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 information and communication convergence engineering
• /
• v.7 no.4
• /
• 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.

• Journal of the Korean Mathematical Society
• /
• v.38 no.3
• /
• pp.623-631
• /
• 2001

Let $\kappa$ be a real abelian field of conductor f and $\kappa$(sub)$\infty$ = ∪(sub)n$\geq$0$\kappa$(sub)n be its Z(sub)p-extension for an odd prime p such that płf$\phi$(f). he aim of this paper is ot compute the cohomology groups of circular units. For m>n$\geq$0, let G(sub)m,n be the Galois group Gal($\kappa$(sub)m/$\kappa$(sub)n) and C(sub)m be the group of circular units of $\kappa$(sub)m. Let l be the number of prime ideals of $\kappa$ above p. Then, for mm>n$\geq$0, we have (1) C(sub)m(sup)G(sub)m,n = C(sub)n, (2) H(sup)i(G(sub)m,n, C(sub)m) = (Z/p(sup)m-n Z)(sup)l-1 if i is even, (3) H(sup)i(G(sub)m,n, C(sub)m) = (Z/P(sup)m-n Z)(sup l) if i is odd (※Equations, See Full-text).

• Journal of KIISE:Software and Applications
• /
• v.31 no.2
• /
• pp.203-212
• /
• 2004

The watermarking is a method to achieve the copyright protection of multimedia contents. Among several media, the left documents show very peculiar properties: block/line/word patterning, clear separation between foreground and background areas. So algorithms specific to the text documents are required that meet those properties. This paper proposes a novel watermarking algorithm for the grayscale text document images. The algorithm inserts the watermark signals through the edge direction histograms. A concept of sub-image consistency is developed that the sub-images have similar shapes in terms of edge direction histograms. Using Korean, Chinese, and English document images, the concept is evaluated and proven to be valid over a wide range of document images. To insert watermark signals, the edge direction histogram is modified slightly. The experiments were performed on various document images and the algorithm was evaluated in terms of imperceptibility and robustness.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.285-289
• /
• 2009

Tag is important to retrieve and classify contents. However, someone uses so many unrelated tags with contents for the high ranking In this work, we propose tag refinement algorithm using TextRank. We calculate the importance of keywords occurred a title, description, tag, and comments. We refine tags removing unrelated keywords from user generated tags. From the results of experiments, we can see that proposed method is useful for refining tags.

• Human Ecology Research
• /
• v.59 no.2
• /
• pp.245-259
• /
• 2021

The purpose of this study was to analyze big data to identify the sub-dimensions of ethical consumption, as well as the consumption value associated with ethical consumption that changes over time. For this study, data were collected from Naver and Daum using the keyword 'ethical consumption' and frequency and matrix data were extracted through Textom, for the period January 1, 2016, to December 31, 2018. In addition, a two-way mode network analysis was conducted using the UCINET 6.0 program and visualized using the NetDraw function. The results of text mining show increasing keyword frequency year-on-year, indicating that interest in ethical consumption has grown. The sub-dimensions derived for 2014 and 2015 are fair trade, ethical consumption, eco-friendly products, and cooperatives and for 2016 are fair trade, ethical consumption, eco-friendly products and animal welfare. The results of deriving consumption value keywords were classified as emotional value, social value, functional value and conditional value. The influence of functional value was found to be growing over time. Through network analysis, the relationship between the sub-dimensions of ethical consumption and consumption values derived each year from 2014 to 2018 showed a significantly strong correlation between eco-friendly product consumption and emotional value, social value, functional value and conditional value.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.9
• /
• pp.2312-2325
• /
• 2013

In this paper, we propose a preprocessing method that it is to correct the distortion of text area in Korean signboard images as a preprocessing step to improve character recognition. Distorted perspective in recognizing of Korean signboard text may cause of the low recognition rate. The proposed method consists of four main steps and eight sub-steps: main step consists of potential vertical components detection, vertical components detection, text-boundary estimation and distortion correction. First, potential vertical line components detection consists of four steps, including edge detection for each connected component, pixel distance normalization in the edge, dominant-point detection in the edge and removal of horizontal components. Second, vertical line components detection is composed of removal of diagonal components and extraction of vertical line components. Third, the outline estimation step is composed of the left and right boundary line detection. Finally, distortion of the text image is corrected by bilinear transformation based on the estimated outline. We compared the changes in recognition rates of OCR before and after applying the proposed algorithm. The recognition rate of the distortion corrected signboard images is 29.63% and 21.9% higher at the character and the text unit than those of the original images.

• Bulletin of the Korean Mathematical Society
• /
• v.58 no.5
• /
• pp.1279-1300
• /
• 2021

Let P be a finite point set in ℝ² with the set of distance n-chains defined as ∆_n(P) = {(|p₁ - p₂|, |p₂ - p₃|, …, |p_n - p_n+1|) : p_i ∈ P}. We show that for 2 ⩽ n = O_|P|(1) we have ${\mid}{\Delta}_n(P){\mid}{\gtrsim}{\frac{{\mid}P{\mid}^n}{{\log}^{\frac{13}{2}(n-1)}{\mid}P{\mid}}}$. Our argument uses the energy construction of Elekes and a general version of Rudnev's rich-line bound implicit in [28], which allows one to iterate efficiently on intersecting nested subsets of Guth-Katz lines. Let G is a simple connected graph on m = O(1) vertices with m ⩾ 2. Define the graph-distance set ∆_G(P) as ∆_G(P) = {(|p_i - p_j|)_{i,j}∈E(G) : p_i, p_j ∈ P}. Combining with results of Guth and Katz [17] and Rudnev [28] with the above, if G has a Hamiltonian path we have ${\mid}{\Delta}_G(P){\mid}{\gtrsim}{\frac{{\mid}P{\mid}^{m-1}}{\text{polylog}{\mid}P{\mid}}}$.

• Communications of the Korean Mathematical Society
• /
• v.16 no.2
• /
• pp.291-296
• /
• 2001

For randomly indexed sums of the form (Equation. See Full-text), where {X(sub)ni, i$\geq$1, n$\geq$1} are random variables, {N(sub)n, n$\geq$1} are suitable conditional expectations and {b(sub)n, n$\geq$1} are positive constants, we establish a general weak law of large numbers. Our result improves that of Hong [3].

• Journal of the Korean Mathematical Society
• /
• v.57 no.4
• /
• pp.1019-1030
• /
• 2020

Efficient computation of r-th root in 𝔽_q has many applications in computational number theory and many other related areas. We present a new r-th root formula which generalizes Müller's result on square root, and which provides a possible improvement of the Cipolla-Lehmer type algorithms for general case. More precisely, for given r-th power c ∈ 𝔽_q, we show that there exists α ∈ 𝔽_q^r such that $$Tr{\left(\begin{array}{cccc}{{\alpha}^{{\frac{({\sum}_{i=0}^{r-1}\;q^i)-r}{r^2}}}\atop{\text{ }}}\end{array}\right)}^r=c,$$ where $Tr({\alpha})={\alpha}+{\alpha}^q+{\alpha}^{q^2}+{\cdots}+{\alpha}^{q^{r-1}}$ and α is a root of certain irreducible polynomial of degree r over 𝔽_q.