• Journal of KIISE:Software and Applications
• /
• v.30 no.1_2
• /
• pp.31-39
• /
• 2003

It is often assumed that text regions in images are characterized by some distinctive or characteristic spatial frequencies. This feature is highly intuitive, and thus appealing as much. We propose a method of detecting horizontal texts in natural scene images. It is based on the use of two features that can be employed separately or in succession: the frequency of edge pixels across vertical and horizontal scan lines, and the fundamental frequency in the Fourier domain. We confirmed that the frequency features are language independent. Also addressed is the detection of quadrilaterals or approximate rectangles using Hough transform. Since texts that is meaningful to many viewers usually appear within rectangles with colors in high contrast to the background. Hence it is natural to assume the detection rectangles may be helpful for locating desired texts correctly in natural outdoor scene images.

• Journal of the Korean Society for information Management
• /
• v.27 no.1
• /
• pp.61-77
• /
• 2010

Blogs are now one of the major information resources on the web. The purpose of this study is to enhance the performance of blog retrieval by means of user assigned tags and trackback information. To this end, retrieval experiments were performed with a dataset of 4,908 blog pages together with their associated trackback URLs. In the experiments, text terms, user tags, and network centrality values based on trackbacks were variously combined as retrieval features. The experimental results showed that employing user tags and network centrality values as retrieval features in addition to text words could improve the performance of blog retrieval.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.8 no.2
• /
• pp.82-95
• /
• 1998

In this paper, we propose a text-independent speaker identification system which has a classifier composed of two parts; to calculate the degree of likeness of each speech frame and to select the most probable speaker from the entire speech duration. The first part is realized using RBFN which is selforganized through learning and in the second part the speaker is determined using a con-tbination of MAXNET and delayed summings. And we use features from linear speech production model and features from fractal geometry. Closed-set speaker identification experiments on 13 male homogeneous speakers show that the proposed techniques can achieve the identification ratio of 100% as the number of delays increases.

• The KIPS Transactions:PartD
• /
• v.10D no.2
• /
• pp.343-350
• /
• 2003

This paper presents a new agent which convert HTML contents to VXML contents automatically for voice services via web. In this paper, I propose an interactive hybrid sequential contents selection method to select desired contents fast and robustly from known web pages. It uses real time structural features as well as embedded text and/or priori structural knowledge such as link symbol position. To verify its effectiveness, a full agent system is implemented and tested. The method reflects user intention more accurately than conventional selections using structural features and is more robust to variations of HTML programming techniques. The agent is fast and has less computational burden than methods use XML or XHTML conversion as intermediate stage.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.5
• /
• pp.395-399
• /
• 2009

In this work, we describe the syntactic tag prediction system for Korean using the decision tree and CRFs. Generally they select features by their intuition. It depends on their prior knowledge. In this works, we combine features systematically using the decision tree. We also analyze errors and optimize features for the best performance. From the result of experiments, we can see that the proposed method is effective for the syntactic tag estimation and will be helpful for the syntactic analysis.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.592-595
• /
• 2009

We propose an automatic spam mail classifier for e-mail data using Support Vector Machines (SVM). We use a lexical form of a word and its part of speech (POS) tags as features. We select useful features with ${\chi}^2$ statistics and represent each feature using text frequency (TF) and inversed document frequency (IDF) values for each feature. After training SVM with the features, SVM classifies each email as spam mail or not. In experiment, we acquired 82.7% of accuracy with e-mail data collected from a web mail system.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.17 no.2
• /
• pp.81-97
• /
• 2022

This paper's purpose is to get hold of the recent research trend by analyzing the variables uesd in startups related papers. The startups related papers in this paper are the papers which include 'startups' in the title of the registered papers from the year 2013 to the year 2020. This study's analysis methods are text-mining of all variables and text-network analysis of affected variables. Visualizing tool for network analysis is Gephi. The result of variables' analysis is as follows. First, independent variables consist mainly of variables about startups' internal factors and outside environment, but due to startups' features like early stage company's features, innovative features, most of variables are about enterprise internal competitiveness, marketing 4P strategy, entrepreneurship, coopreation method, transformational leadership, enterprise features, lean startup strategy, enterprise internal communication, value orientation, task conflict, relationship conflict, knowledge sharing, etc. Second, dependent variables are mainly about outcome, and are classified into financial performance and non-financial performance by overall concept. In other words, startups related papers have higher interest in non-financial performance, like management performance, team performance, SCM performance as well as financial performance like sales quantity owing to startups' immaturity in getting good financial performance. Through this study we can find out as follows. Although there are not many officially registered papers dealing with startups, those papers include various themes about stratups. For example, there are trendy themes like lean startups strategy, crowdfunding, influencer and accelerator, etc.

• Phonetics and Speech Sciences
• /
• v.7 no.1
• /
• pp.3-10
• /
• 2015

In this paper, different frequency scales in cepstral feature extraction are evaluated for the text-independent speaker recognition. To this end, mel-frequency cepstral coefficients (MFCCs), linear frequency cepstral coefficients (LFCCs), and bilinear warped frequency cepstral coefficients (BWFCCs) are applied to the speaker recognition experiment. In addition, the spectro-temporal features extracted by the cepstral-time matrix (CTM) are examined as an alternative to the delta and delta-delta features. Experiments on the NIST speaker recognition evaluation (SRE) 2004 task are carried out using the Gaussian mixture model-universal background model (GMM-UBM) method and the joint factor analysis (JFA) method, both based on the ALIZE 3.0 toolkit. Experimental results using both the methods show that BWFCC with appropriate warping factor yields better performance than MFCC and LFCC. It is also shown that the feature set including the spectro-temporal information based on the CTM outperforms the conventional feature set including the delta and delta-delta features.

• KIISE Transactions on Computing Practices
• /
• v.24 no.2
• /
• pp.93-98
• /
• 2018

Recommendation systems are used to provide items of interests for users to maximize a company's profit. Matrix factorization is frequently used by recommendation systems, based on an incomplete user-item rating matrix. However, as the number of items and users increase, it becomes difficult to make accurate recommendations due to the sparsity of data. To overcome this drawback, the use of text data related to items was recently suggested for matrix factorization algorithms. Furthermore, a word-level convolutional neural network was shown to be effective in the process of extracting the word-level features from the text data among these kinds of matrix factorization algorithms. However, it involves a large number of parameters to learn in the word-level convolutional neural network. Thus, we propose a matrix factorization algorithm which utilizes a character-level convolutional neural network with which to extract the character-level features from the text data. We also conducted a performance study with real-life datasets to show the effectiveness of the proposed matrix factorization algorithm.

• KIISE Transactions on Computing Practices
• /
• v.22 no.6
• /
• pp.290-295
• /
• 2016

Recently, folksonomy-based image-sharing sites where users cooperatively make and utilize tags of image annotation have been gaining popularity. Typically, these sites retrieve images for a user request using simple text-based matching and display retrieved images in the form of photo stream. However, these tags are personal and subjective and images are not categorized, which results in poor retrieval accuracy and low user satisfaction. In this paper, we propose a categorization scheme for folksonomy images which can improve the retrieval accuracy in the tag-based image retrieval systems. Consequently, images are classified by the semantic similarity using text-information and image-information generated on the folksonomy. To evaluate the performance of our proposed scheme, we collect folksonomy images and categorize them using text features and image features. And then, we compare its retrieval accuracy with that of existing systems.