• Journal of Korea Multimedia Society
• /
• v.13 no.9
• /
• pp.1322-1328
• /
• 2010

Recently, keyword spotting system is greatly in the limelight as UI(User Interface) technology of ubiquitous home network system. Keyword spotting system is vulnerable to non-stationary noises such as TV, radio, dialogue. Especially, speech recognition rate goes down drastically under the embedded DSP(Digital Signal Processor) environments because it is relatively low in the computational capability to process input speech in real-time. In this paper, we propose a new keyword spotting system using the call-command method, which is consisted of small number of recognition networks. We select the call-command such as 'narae', 'home manager' and compose the small network as a token which is consisted of silence with the noise and call commands to carry the real-time recognition continuously for input speeches.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.3
• /
• pp.1442-1453
• /
• 2016

Now, there are prodigiously expanding worldwide economic networks in the information society, which require their social structural changes through technology innovations. This paper so tries to formally define a process-centered knowledge model to be used to analyze policy-making procedures on technology innovations. The eventual goal of the proposed knowledge model is to apply itself to analyze a topic network based upon composite keywords from a document written in a natural language format during the technology innovation procedures. Knowledge model is created to topic network that compositing driven keyword through text mining from natural language in document. And we show that the way of analyzing knowledge model and automatically generating feature keyword and relation properties into topic networks.

• Journal of KIISE
• /
• v.44 no.11
• /
• pp.1236-1243
• /
• 2017

In this paper, we propose a novel way of producing keyword networks, named LSI-based ClusterTextRank, which extracts significant key words from a set of clusters with a mutual information metric, and constructs an association network using latent semantic indexing (LSI). The proposed method reduces the dimension of documents through LSI, decomposes documents into multiple clusters through k-means clustering, and expresses the words within each cluster as a maximal spanning tree graph. The significant key words are identified by evaluating their mutual information within clusters. Then, the method calculates the similarities between the extracted key words using the term-concept matrix, and the results are represented as a keyword association network. To evaluate the performance of the proposed method, we used travel-related blog data and showed that the proposed method outperforms the existing TextRank algorithm by about 14% in terms of accuracy.

• Health Policy and Management
• /
• v.18 no.2
• /
• pp.86-106
• /
• 2008

This research analyzed. knowledge structure and its effect factor by analysis of co-author and keyword network in Korea's health policy and administration sector. The data was extracted from 339 articles listed in the Korean Journal of Health Policy and Administration, and was transformed into a co-author and keyword matrix. In this matrix the existence of a link was defined by impact factors which were calculated by the weight value of what the role was and the rate of how many authors contributed. We demonstrated that the research achievement was dependent on the author's status and network index. Analysis methods were neighborhood degree, correspondence analysis, multiple regression and the difference of weight distribution by research fields. Co-author networks were developed as closeness centrality as well as degree centrality by a few high productivity researchers. In particular, power law distribution was discovered in impact factor and research productivity. The effect of the author's role was significant in both the impact factor calculated by the participatory rate and the number of listed articles. Especially, this journal shared its major researchers who had a licensed physician with the Journal of Preventive Medicine and Public Health. Therefore, social scientists were likely to be small co-author network differently from natural scientists. It was so called 'two cultures' phenomenon. This study showed how can we verified academic research structure existed in the unit of journal like as citation networks. The co-author networks in the field of health policy and administration had more differentiated and clustered than preventive medicine and epidemiology fields.

• Asia-Pacific Journal of Business
• /
• v.15 no.1
• /
• pp.305-318
• /
• 2024

Purpose - The purpose of this study was to conduct a preliminary study to identify key trends on research articles indexed in KCI in relation to tourism in Jeju and sports tourism. Design/methodology/approach - Information regarding research articles focused on Jeju tourism and sports tourism indexed in KCI (145 and 120 articles respectively) were collected and finally abstract written in Korean of 100 and 91 articles on sports tourism and Jeju tourism respectively were chosen for the further analysis after removing redundant articles. R program was used to analyze keyword frequencies, co-occurring terms, and degree/betweeness centrality measures and visualize the keyword network results. Findings - Event, marketing, content, program, implication, service, stadium, and tourism destination have been identified as keywords with highest frequencies among research on sport tourism, whereas tourism destination, image, brand, content, data, Chinese, satisfaction, eco-tourism service, place of arrival were highly appearing terms among research on Jeju tourism. Research implications or Originality - This study highlighted that Jeju has been interlinked with a range of terms such as programs influencing Jeju tourism, natural environment, tourism-related resources (e.g., museums, dramas, etc.), whereas sports has been closely related to sports event and vaiours types of sports (e.g., bicycle, staking, and scuber), but not to Jeju-do.

• Journal of Navigation and Port Research
• /
• v.38 no.6
• /
• pp.607-614
• /
• 2014

This article aims to explore and discuss research trends in global cruise industry using keyword network analysis. We visualize keyword networks in each of four groups of 1982-1999, 2000-2004, 2005-2009, 2010-2014 based on the top 20 keyword nodes' degree centrality and betweenness centrality which are selected among four centrality measurements, comparing them with frequency order. The article shows that keyword frequency collected from 240 articles published in international journals is subject to Zipf's law and nodes degree distribution also exhibits power law. We try to find out research trends in global cruise industry to change some important keywords diachronically, visualizing several networks focusing on the top two keywords, cruise and tourism, belonging to all the four year groups, with high degree and betweenness centrality values. Interestingly enough, a new node, China, connecting the top most keywords, appears in the most recent period of 2010-2014 when China has emerged as one of the rapid development countries in global cruise industry. Therefore keyword network analysis used in this article will be useful to understand research trends in global cruise industry because of increase and decrease of numbers of network types in different year groups and the visual connection between important nodes in giant components.

• The Journal of the Korea Contents Association
• /
• v.16 no.6
• /
• pp.205-214
• /
• 2016

As a large amount of graph-structured data is widely used in various applications such as social networks, semantic web, and bio-informatics, keyword-based search over graph data has been getting a lot of attention. In this paper, we propose an efficient method for keyword search over graph data to find a set of top-k answers that are relevant as well as non-redundant in structure. We define a non-redundant answer structure for a keyword query and a relevance measure for the answer. We suggest a new indexing scheme on the relevant paths between nodes and keyword terms in the graph, and also propose a query processing algorithm to find top-k non-redundant answers efficiently by exploiting the pre-calculated indexes. We present effectiveness and efficiency of the proposed approach compared to the previous method by conducting an experiment using a real dataset.

• IE interfaces
• /
• v.25 no.4
• /
• pp.458-471
• /
• 2012

In this study, we investigate the cooperative relationships between researchers who have co-authorship in the logistics-related journals in Korea by using social network analysis (SNA). We analyzed the co-authorship data of 781 articles published from 2005 to 2011 in four journals of 'Logistics Study', 'Journal of Korean Society of SCM', 'Korea Logistics Review' and 'Journal of Shipping and Logistics.' We examined the trend of cooperative research in the field of logistics with basic data of the co-authorship network. Then, we analyzed structural properties of the network and the sub-networks of research groups having co-authorship. We could verify the authors who play important roles within the network by using SNA indicators. In addition, we constructed the keyword networks based on the keyword data of all articles by research groups in order to understand the research topics of each group, and thereby we could draw several implications on the cooperative researches in the field of logistics.

• The Journal of the Korea Contents Association
• /
• v.18 no.7
• /
• pp.449-458
• /
• 2018

As the usage of social network services increases, event information occurring offline is spreading more rapidly. Therefore, studies have been conducted to detect events by analyzing social data. In this paper, we propose a graph based event detection scheme considering user interest in social networks. The proposed scheme constructs a keyword graph by analyzing tweets posted by users. We calculates the interest measure from users' social activities and uses it to identify events by considering changes in interest. Therefore, it is possible to eliminate events that are repeatedly posted without meaning and improve the reliability of the results. We conduct various performance evaluations to demonstrate the superiority of the proposed event detection scheme.

• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.6
• /
• pp.11-24
• /
• 2019

The tendency of buyers of web information is changing from the cost-effectiveness to the cost-satisfaction. There is such tendency in the recommendation of multimedia contents, some of which are folksonomy-based recommendation services using mood. However, there is a problem that they does not consider synonyms. In order to solve this problem, some studies have solved the problem by defining 12 moods of Thayer model as AV values (Arousal and Valence), but the recommendation performance is lower than that of a keyword-based method at the recall level 0.1. In this paper, we propose a method based on using mood vector of multimedia contents. The method can solve the synonym problem while maintaining the same performance as the keyword-based method even at the recall level 0.1. Also, for performance analysis, we compare the proposed method with an existing method based on AV value and a keyword-based method. The result shows that the proposed method outperform the existing methods.