• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.265-274
• /
• 2021

In this paper, we propose a novel recommendation model based on Doc2vec using search keywords and product details. Until now, a lot of prior studies on recommender systems have proposed collaborative filtering (CF) as the main algorithm for recommendation, which uses only structured input data such as customers' purchase history or ratings. However, the use of unstructured data like online customer review in CF may lead to better recommendation. Under this background, we propose to use search keyword data and product detail information, which are seldom used in previous studies, for product recommendation. The proposed model makes recommendation by using CF which simultaneously considers ratings, search keywords and detailed information of the products purchased by customers. To extract quantitative patterns from these unstructured data, Doc2vec is applied. As a result of the experiment, the proposed model was found to outperform the conventional recommendation model. In addition, it was confirmed that search keywords and product details had a significant effect on recommendation. This study has academic significance in that it tries to apply the customers' online behavior information to the recommendation system and that it mitigates the cold start problem, which is one of the critical limitations of CF.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.903-904
• /
• 2008

This paper suggests a simple scheme of block motion estimation in which the search pattern selection is based on the classification of motion content available in the spatio temporal neighboring blocks. The search area is divided into eight sectors and the search pattern selection is also based on the direction of predicted motion vector. Experimental results show that the proposed algorithm has achieved good predicted image quality measured in terms of PSNR and has very less computational complexity.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.31A no.7
• /
• pp.55-62
• /
• 1994

Interpattern association (IPA) model which the interconnection weight matrix(IWM) is constructed by the association between patterns is effective in similar pattern recognitions. But, if the number of reference patterns is increased, the ability of recognition is decreased. Using a hierarchical recognition algorithm which adopts the tree search strategy, we classified reference patterns into sub-groups by similarity. In IPA model, if input includes random noise we make it converge to reference pattern by means of input includes random noise we make it converge to reference pattern by means of increasing the number of pixels of prohibited state in IWM. In relation to reference patterns the pixel of prohibited state made partially prohibited state of no connected state using which is not included common and feature regions by each reference patterns.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.1 s.45
• /
• pp.57-65
• /
• 2007

In the fast block matching algorithm. search patterns of different shapes or sizes and the distribution of motion vectors have a large impact on both the searching speed and the image qualify. In this paper, we propose a new fast block matching algorithm using the flat-hexagon search pattern that ate solved disadvantages of the diamond pattern search algorithm(DS) and the hexagon-based search algorithm(HEXBS). Our proposed algorithm finds mainly the motion vectors that not close to the center of search window using the flat-hexagon search pattern. Through experiments, compared with the DS and HEXBS, the proposed f)at-hexagon search algorithm(FHS) improves about $0.4{\sim}21.3%$ in terms of average number of search point per motion vector estimation and improves about $0.009{\sim}0.531dB$ in terms of PSNR(Peak Signal to Noise Ratio).

• The Journal of Society for e-Business Studies
• /
• v.6 no.2
• /
• pp.87-99
• /
• 2001

Along with the rapid emergence of the Internet and e-commerce, online auctions are hitting the spotlights. The inconveniences found in off-line auctions, such as time and place restriction and limited number of items, are solved in the online auction. However, not so many auction sites have integrated auction information systems, which monitor the present status of auctions, resulting in greater inconvenience for the online auction users. Moreover, there is no auction site that suggests the appropriate starting or closing price that is useful for users when they make, their bids, What the online auction users need is an auction system that can solve such problems. This study is purported for solving the problems by designing and implementing a real time auction system that applies the comparison search functions and the agent functions. In other words, an integrated database system using a bidder-oriented agent for providing information is built so that the users can search and compare the information on the item they are interested in and make a faster and more accurate purchase. Also the appropriate starting and closing prices are offered to the sellers and bidders through the integrated system for a closer and more accurate comparison and analysis of the prices. For future work, the product recommendation service, which accurately reflects the bidding patterns, and the methods for studying the multi bidding pattern will be applied to the suggested system to realize a real time auction information system that supports CRM(Customer Relationship Management) .

• Journal of the Ergonomics Society of Korea
• /
• v.24 no.1
• /
• pp.19-26
• /
• 2005

Mobile internet access using such devices as PDAs and cellular phones becomes popular as mobile technologies grow. However, the characteristics of small screen devices such as small screen size and pen-based input cause many usability problems. In this study, a human factors experiment was conducted to identify the factors affecting the usability of information search on a PDA. Factors manipulated in the experiment included use of wheel equipment, scroll dimension, and screen orientation. Task completion time, error frequency, and subjective satisfaction level were measured. In addition, various users' behavioral patterns such as scanning routes and mainly used scrolling methods were analyzed. Scroll dimension has a significant effect on task completion time. Scroll equipment and screen orientation affect subjective satisfaction level. The results could be applied to designing information structure of web sites for mobile use by providing vertical scroll and using external scroll equipments.

• Health Policy and Management
• /
• v.28 no.2
• /
• pp.119-127
• /
• 2018

Background: This study aimed to examine (1) the patterns of online health information search with respect to seeking and scanning, and (2) how online search, along with eHealth literacy, predicts perceived information usefulness in the context of diet and weight control. Methods: Online survey was conducted with 299 adults from the consumer panel recruited for the purpose of quality assessment of the Korean National Health Information Portal in 2016. We conducted paired sample t-test and multiple logistic regression to address the research questions. Data analysis was performed using IBM SPSS Statistics ver. 24.0 (IBM Corp., Armonk, NY, USA) and SAS ver. 9.3 (SAS Institute Inc., Cary, NC, USA). Results: Of the respondents, 38.8% were 'high seek-high scanners,' 35.8% were 'low seek-low scanners,' 13.0% were 'high seek-low scanners,' and 12.4% were 'low seek-high scanners.' eHealth literacy was a significant, positive predictor of online information scanning (odds ratio [OR], 2.46; 95% confidence interval [CI], 1.41-4.29), but not for online information seeking (OR, 1.75; 95% CI, 1.00-3.05). With respect to perceived usefulness of online information seeking, online seeking (OR, 4.90; 95% CI, 2.19-11.00) and eHealth literacy (OR, 2.30; 95% CI, 1.11-4.75) were significant predictors. Perceived usefulness of online scanning had a significant association with online scanning (OR, 2.38; 95% CI, 1.08-5.22), but not with eHealth literacy. Conclusion: To increase the effectiveness of the health policy for online information search and related outcomes in the context of diet and weight control, it is important to develop education programs promoting eHealth literacy.

• The Journal of Korean Academic Society of Nursing Education
• /
• v.8 no.1
• /
• pp.38-50
• /
• 2002

The purpose of this study is to identify the relationship among injury prevention practices, health locus of control, and response patterns to HLOC of the elderly. Subjects were 121 healthy elderly. The data had been collected from November 5 to 18 in 2001 and analyzed with descriptive statistics, t-test, ANOVA, Pearson's correlation and stepwise multiple regression. Health locus of control and injury prevention practices were measured by using MHLC scale and an instrument created by the researcher on the basis of the results of literature review respectively. The results of this study were as follows: The mean score of injury prevention practices was 2.80 and the mean scores for the health locus of control were internal health locus of control : 17.25, external health locus of control : 16.09, and chance health locus of control : 14.26. The response patterns of the HLOC identified were six types; pure internal, pure powerful others, pure chance, double external, believers in control, and complex control. The 'pure internal' was the largest group(35.5%), and the 'believers in control' was the next(31.4%). The relationship between internal health locus of control and the injury prevention practices of the elderly revealed a significant correlation(r=.215, p=.018). The relationship between external health locus of control and the injury prevention practices of the elderly revealed a significant correlation (r= .208, p=.022). There was significant difference between response patterns of the health locus of control and injury prevention practices(F=2.393, p=.042). There were significant differences between injury prevention practices and general characteristic factors, which were education, family type, administration of medication, injury experience, ADL, and self-directed search for health information. Self-directed search for health information, injury experience, and education explained 16.7% of the variance for injury prevention practices. The above results may be used as the basic data for seeking more efficient way of improving safety of the elderly.

• Journal of information and communication convergence engineering
• /
• v.13 no.2
• /
• pp.123-131
• /
• 2015

Training neural networks is a complex task with great importance in the field of supervised learning. In the training process, a set of input-output patterns is repeated to an artificial neural network (ANN). From those patterns weights of all the interconnections between neurons are adjusted until the specified input yields the desired output. In this paper, a new hybrid algorithm is proposed for global optimization of connection weights in an ANN. Dynamic swarms are shown to converge rapidly during the initial stages of a global search, but around the global optimum, the search process becomes very slow. In contrast, the gradient descent method can achieve faster convergence speed around the global optimum, and at the same time, the convergence accuracy can be relatively high. Therefore, the proposed hybrid algorithm combines the dynamic particle swarm optimization (DPSO) algorithm with the backpropagation (BP) algorithm, also referred to as the DPSO-BP algorithm, to train the weights of an ANN. In this paper, we intend to show the superiority (time performance and quality of solution) of the proposed hybrid algorithm (DPSO-BP) over other more standard algorithms in neural network training. The algorithms are compared using two different datasets, and the results are simulated.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.8
• /
• pp.281-286
• /
• 2018

With recent increase in complexity and variety of information and massively available information, interest in and necessity of ontology has been on the rise as a method of extracting a meaningful search result from massive data. Although there have been proposed many methods of extracting the ontology from a given text of a natural language, the extraction based on most of the current methods is not consistent with the structure of the ontology. In this paper, we propose a method of automatically creating ontology by distinguishing a term needed for establishing the ontology from a text given in a specific domain and extracting various relationships between the terms based on the pattern-based method. To extract the relationship between the terms, there is proposed a method of reducing the size of a searching space by taking a matching set of patterns into account and connecting a join-set concept and a pattern array. The result is that this method reduces the size of the search space by 50-95% without removing any useful patterns from the search space.