• Journal of Korea Society of Digital Industry and Information Management
• /
• v.20 no.2
• /
• pp.37-47
• /
• 2024

The food tech market, which uses artificial intelligence robots for the restaurant industry, is gradually expanding. Among them, the robot barista, a representative food tech case for the restaurant industry, is characterized by increasing the efficiency of operators and providing things for visitors to see and enjoy through a 24-hour unmanned operation. This research was conducted through text mining analysis to examine trends related to robot baristas in the restaurant industry. The research results are as follows. First, keywords such as coffee, cafe, certification, ordering, taste, interest, people, robot cafe, coffee barista expert, free, course, unmanned, and wine sommelier were highly frequent. Second, time, variety, possibility, people, process, operation, service, and thought showed high closeness centrality. Third, as a result of CONCOR analysis, a total of 5 keyword clusters with high relevance to the restaurant industry were formed. In order to activate robot barista in the future, it is necessary to pay more attention to functional development that can strengthen its functions and features, as well as online promotion through various events and SNS in the robot barista cafe.

• Phonetics and Speech Sciences
• /
• v.12 no.4
• /
• pp.27-36
• /
• 2020

The aim of this study is to investigate the perceptive aspects of the polysemic sentence final ending "-(eu)lgeol" produced by Chinese Korean learners. "-(Eu)lgeol" has two different meanings, that is, a guess and a regret, and these different meanings are expressed by the different prosodic features of the last syllable of "-(eu)lgeol". To examine how Korean native speakers perceive "-(eu)lgeol" sentences produced by Chinese Korean learners and the most saliant prosodic variable for the semantic discrimination of "-(eu)lgeol" at the perceptive level, we performed a perceptual experiment. The analysed material constituted four Korean sentences containing "-(eu)lgeol" in which two sentences expressed guesses and the other two expressed regret. Twenty-five Korean native speakers participated in the perceptual experiment. Participants were asked to mark whether "-(eu)lgeol" sentences they listened to were (1) definitely regrets, (2) probably regrets, (3) ambiguous, (4) probably guesses, or (5) definitely guesses based on the prosodic features of the last syllable of "-(eu)lgeol". The analysed prosodic variables were sentence boundary tones, slopes of boundary tones, pitch difference between sentence-final and penultimate syllables, and pitch levels of boundary tones. The results show that all the analysed prosodic variables are significantly correlated with the semantic discrimination of "-(eu)lgeol" and among these prosodic variables, the most salient role in the semantic discrimination of "-(eu)lgeol" is pitch difference between sentence-final syllable and penultimate syllable.

• The Korean Journal of Archival Studies
• /
• no.44
• /
• pp.51-99
• /
• 2015

The purpose of the archival appraisal has gradually changed from the selection of records to the documentation of the society. In particular, the qualitative and quantitative developments of the current digital technology and web have become the driving force that enables semantic acquisition, rather than physical one. Under these circumstances, the concept of 'documentary heritage' has been re-established internationally, led by UNESCO. Library and Archives Canada (LAC) reflects this trend. LAC has been trying to develop a new appraisal model and an acquisition model at the same time to revive the spirit of total archives, which is the 'Whole-of-society approach'. Features of this approach can be summarized in three main points. First, it is for documentary heritage and the acquisition refers to semantic acquisition, not the physical one. And because the object of management is documentary heritage, the cooperation between documentary heritage institutions has to be a prerequisite condition. Lastly, it cannot only documenting what already happened, it can documenting what is happening in the current society. 'Whole-of-society approach', as an appraisal method, is a way to identify social components based on social theories. The approach, as an acquisition method, is targeting digital recording, which includes 'digitized' heritage and 'born-digital' heritage. And it makes possible to the semantic acquisition of documentary heritage based on the data linking by mapping identified social components as metadata component and establishing them into linked open data. This study pointed out that it is hard to realize documentation of the society based on domestic appraisal system since the purpose is limited to selection. To overcome this limitation, we suggest a guideline applied with 'Whole-of-society approach'.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_3
• /
• pp.939-951
• /
• 2022

The most important factor in designing autonomous driving systems is to recognize the exact location of the vehicle within the surrounding environment. To date, various sensors and navigation systems have been used for autonomous driving systems; however, all have limitations. Therefore, the need for high-definition (HD) maps that provide high-precision infrastructure information for safe and convenient autonomous driving is increasing. HD maps are drawn using three-dimensional point cloud data acquired through a mobile mapping system (MMS). However, this process requires manual work due to the large numbers of points and drawing layers, increasing the cost and effort associated with HD mapping. The objective of this study was to improve the efficiency of HD mapping by segmenting semantic information in an MMS point cloud into six classes: roads, curbs, sidewalks, medians, lanes, and other elements. Segmentation was performed using various machine learning techniques including random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN), and gradient-boosting machine (GBM), and 11 variables including geometry, color, intensity, and other road design features. MMS point cloud data for a 130-m section of a five-lane road near Minam Station in Busan, were used to evaluate the segmentation models; the average F1 scores of the models were 95.43% for RF, 92.1% for SVM, 91.05% for GBM, and 82.63% for KNN. The RF model showed the best segmentation performance, with F1 scores of 99.3%, 95.5%, 94.5%, 93.5%, and 90.1% for roads, sidewalks, curbs, medians, and lanes, respectively. The variable importance results of the RF model showed high mean decrease accuracy and mean decrease gini for XY dist. and Z dist. variables related to road design, respectively. Thus, variables related to road design contributed significantly to the segmentation of semantic information. The results of this study demonstrate the applicability of segmentation of MMS point cloud data based on machine learning, and will help to reduce the cost and effort associated with HD mapping.

• Journal of KIISE:Software and Applications
• /
• v.36 no.10
• /
• pp.777-785
• /
• 2009

Collaborative filtering is a technique used to predict whether a particular user will like a particular item. User-based or item-based collaborative techniques have been used extensively in many commercial recommender systems. In this paper, a hybrid collaborative filtering method that combines user-based and item-based methods using a low-dimensional linear model is proposed. The proposed method solves the problems of sparsity and a large database by using NMF among the low-dimensional linear models. In collaborative filtering systems the methods using the NMF are useful in expressing users as semantic relations. However, they are model-based methods and the process of computation is complex, so they can not recommend items dynamically. In order to complement the shortcomings, the proposed method clusters users into groups by using NMF and selects features of groups by using TF-IDF. Mutual information is then used to compute similarities between items. The proposed method clusters users into groups and extracts features of groups on offline and determines the most suitable group for an active user using the features of groups on online. Finally, the proposed method reduces the time required to classify an active user into a group and outperforms previous methods by combining user-based and item-based collaborative filtering methods.

• Journal of the Korean Society for information Management
• /
• v.32 no.4
• /
• pp.137-165
• /
• 2015

This study aimed to investigate the period-specific changes (Library 1.0, Library 2.0, Library 3.0 Period) of job factors and performance features in academic libraries. For this, the study categorized an academic library's job into five dimensions: 1) library administration 2) collection development and management 3) information organization 4) information services and 5) information system development and management, After the categorized library's job was defined in detail, the Delphi survey was conducted twice on librarians and professors of library and information science. The result showed that there were many changes in job factors and performance features in academic libraries towards the period of library 2.0 characterized by user participation, sharing and openness and into library 3.0 characterized by social network and semantic web. Library 3.0 is likely to bring about a significant change in user services with ever changing technological advances stemming from library 2.0, such as mobile services, RFID and NFC etc. The finding of the study suggest that library systems need to be continually upgraded in the period of library 3.0.

• Journal of the Korean earth science society
• /
• v.36 no.4
• /
• pp.404-416
• /
• 2015

We investigated the cognitive frame of high school students and inservice high school science teachers about effective teaching method, and we also explored how they understood about the teaching methods suggested by the 2009 revised Science Curriculum. Data were collected from 275 high school science teachers and 275 high school students. We analyzed data in terms of the words and the cognitive frame using the Semantic Network Analysis. The results were as follows. First, the teachers perceived that an activity oriented class was the effective science class that helped improve students' problem-solving abilities and their inquiry skills. The students had the cognitive frame that their teacher had to present relevant and enough teaching materials to students, and that they should also receive assistance from teachers in science class to better prepare for college entrance exam. Second, both students and teachers retained the cognitive frame about the efficient science class that was not reflected 2009 revised Science Curriculum exactly. Especially, neither groups connected the elements of 'convergence' as well as 'integration' embedded across science subject areas to their cognitive frame nor cognized the fact that many science learning contents were closed related to one another. Therefore, various professional development opportunities should be offered so that teachers succinctly comprehend the essential features and the intents of the 2009 revised Science Curriculum and thereby implement it in their science lessons effectively.

• Journal of Broadcast Engineering
• /
• v.11 no.1 s.30
• /
• pp.28-41
• /
• 2006

Due to the rapid development of multimedia appliances, the increasing amount of multimedia data enforces the development of automatic video analysis techniques. In this paper, a method of ToC generation is proposed for educational video contents. The proposed method consists of two parts: scene segmentation followed by scene annotation. First, video sequence is divided into scenes by the proposed scene segmentation algorithm utilizing the characteristics of educational video. Then each shot in the scene is annotated in terms of scene type, existence of enclosed caption and main speaker of the shot. The ToC generated by the proposed method represents the structure of a video by the hierarchy of scenes and shots and gives description of each scene and shot by extracted features. Hence the generated ToC can help users to perceive the content of a video at a glance and. to access a desired position of a video easily. Also, the generated ToC automatically by the system can be further edited manually for the refinement to effectively reduce the required time achieving more detailed description of the video content. The experimental result showed that the proposed method can generate ToC for educational video with high accuracy.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1045-1055
• /
• 2007

Latent topic models are statistical models which automatically captures salient patterns or correlation among features underlying a data collection in a probabilistic way. They are gaining an increased popularity as an effective tool in the application of automatic semantic feature extraction from text corpus, multimedia data analysis including image data, and bioinformatics. Among the important issues for the effectiveness in the application of latent topic models to the massive data set is the efficient learning of the model. The paper proposes an accelerated learning technique for PLSA model, one of the popular latent topic models, by an incremental EM algorithm instead of conventional EM algorithm. The incremental EM algorithm can be characterized by the employment of a series of partial E-steps that are performed on the corresponding subsets of the entire data collection, unlike in the conventional EM algorithm where one batch E-step is done for the whole data set. By the replacement of a single batch E-M step with a series of partial E-steps and M-steps, the inference result for the previous data subset can be directly reflected to the next inference process, which can enhance the learning speed for the entire data set. The algorithm is advantageous also in that it is guaranteed to converge to a local maximum solution and can be easily implemented just with slight modification of the existing algorithm based on the conventional EM. We present the basic application of the incremental EM algorithm to the learning of PLSA and empirically evaluate the acceleration performance with several possible data partitioning methods for the practical application. The experimental results on a real-world news data set show that the proposed approach can accomplish a meaningful enhancement of the convergence rate in the learning of latent topic model. Additionally, we present an interesting result which supports a possible synergistic effect of the combination of incremental EM algorithm with parallel computing.

• Land and Housing Review
• /
• v.2 no.3
• /
• pp.249-258
• /
• 2011

In order to achieve an efficient urban regeneration of the nation, it is required to determine the extent of decline nation-wide and the declined areas for each district and also to evaluate the potentials of the concerned areas. For this task to be accomplished, a construction of a comprehensive diagnostic system based on spatial information considering diversity and complexity is required. In this study, a total information system architecture for urban regeneration is designed as part of the construction of such a diagnostic system. In order to develop the system, a city and district level unit decline diagnostic indicators has been constructed and a decline diagnostic system has been developed. Also, a scheme to promote the advancement of the system is proposed. The DB construction is based on the city and district level nation-wide and metadata for the concerned level is constructed as well. The system is based on the Open API and designed to be flexible for extension. Also, an RIA-based intuitive UI has been implemented. Main features of the system consist of the management of the indicators, diagnostic analysis (city and district level decline diagnosis), related information, etc. As for methods for the advancement, an information model in consideration of the spation relations of the urban regeneration DB has been designed and application methods of semantic webs. Also, for improvement methods for district unit analytical model, district level analysis models, GIS based spatial analysis platforms and linked utiliation of KOPSS analysis modules are suggested. A use of a total information system for urban regeneration is anticipated to facilitate concerned policy making through the identification of the status of city declines to identify and the understanding of the demands for regeneration.