• 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.

• International Journal of Concrete Structures and Materials
• /
• v.8 no.2
• /
• pp.141-155
• /
• 2014

The use of fiber reinforced polymer (FRP) composites in strengthening reinforced concrete beam-column subassemblies has been scrutinised both experimentally and numerically in recent years. While a multitude of numerical models are available, and many match the experimental results reasonably well, there are not many studies that have looked at the efficiency of different finite elements in a comparative way in order to clearly identify the best practice when it comes to modelling FRP for strengthening. The present study aims at investigating this within the context of FRP retrofitted reinforced concrete beam-column subassemblies. Two programs are used side by side; ANSYS and VecTor2. Results of the finite element modeling using these two programs are compared with a recent experimental study. Different failure and yield criteria along with different element types are implemented and a useful technique, which can reduce the number of elements considerably, is successfully employed for modeling planar structures subjected to in-plane loading in ANSYS. Comparison of the results shows that there is good agreement between ANSYS and VecTor2 results in monotonic loading. However, unlike VecTor2 program, implicit version of ANSYS program is not able to properly model the cyclic behavior of the modeled subassemblies. The paper will be useful to those who wish to study FRP strengthening applications numerically as it provides an insight into the choice of the elements and the methods of modeling to achieve desired accuracy and numerical stability, a matter not so clearly explored in the past in any of the published literature.

• Journal of Information Technology Services
• /
• v.20 no.2
• /
• pp.77-90
• /
• 2021

Online and mobile games represent digital entertainment. Not only the game grows fast, but also it has been noted for unique business models such as a subscription revenue model and free-to-play with partial payment. But, a recent revenue mechanism, called a loot-box system, has been criticized due to overspending, weak protection to teenagers, and more over gambling-like features. Policy makers and research communities have counted on expert opinions, review boards, and temporal survey studies to build countermeasures to minimize negative effects of online and mobile games. In this process, speed was not seriously considered. In this study, we attempt to use a big data source to find a way of observing a trend for policy makers and researchers. Specifically, we tried to apply the Word2Vec data mining algorithm to news repositories. From the findings, we acknowledged that the suggested design would be effective in lightening issues timely and precisely. This study contributes to digital entertainment service communities by providing a practical method to follow up trends; thus, helping practitioners have concrete grounds for balancing public concerns and business purposes.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.79-80
• /
• 2021

인터넷은 우리 경제를 디지털 경제로 변화시키며 전자상거래도 증가하고 있다. 따라서 구매자가 전자상거래에서 남기는 긍정적인, 부정적인 상품평은 상품기획의 주요 정보가 될 수 있다. 본 논문에서는 버티컬 무소음 마우스 10,000개에 대한 정형화된 데이터셋을 Word2Vec을 이용하여 유사도 분석, 온라인 상품평 빈도분석 상위 50개 단어를 제시하여 실제 상품을 사용한 후 설문조사 시행을 하였다. 온라인 상품평 유사도 분석결과 클릭 키워드에 대한 장점으로 통증(.986), 디자인(.982)가 분석되었으며 단점은 적응(.866), 불편(.854)이었다. 오프라인 상품평에서는 장점으로 디자인(17명), 단점으로 불편(11명)이었다. 또한 온라인과 오프라인의 상품평을 비교함으로써 구매자의 긍정, 부정의 의미를 교차 확인하여 유의미한 정보를 제시 하였다고 볼수 있다. 따라서 본 연구에서 제시하는 상품기획 프로세스를 신상품 개발 및 기존 상품의 개선 전략으로 적용할 수 있겠다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.376-379
• /
• 2018

본 논문의 목적은 오픈 소스로 공개된 3가지 한국어 형태소 분석기 (kkma, twitter 및 mecab-ko)를 비교해서 한국어 자연어 처리에 가장 적합한 분석기를 선정하는 것이다. 이를 위해, 자연어 처리 분야에서 중요한 단어 임베딩 방법론 중 하나인 word2vec 모델의 성능 검증 방법을 사용해서 각 형태소 분석기의 성능을 정량적으로 비교했다. 그 결과 mecab-ko 형태소 분석기가 최적임이 확인되었다. 단 성능 검증에 사용된 어휘가 오직 명사뿐이라는 한계가 있으므로, 향후 연구에서는 좀 더 다양한 품사에 대한 성능검증이 필요할 것으로 보인다.

• International Journal of Vascular Biomedical Engineering
• /
• v.3 no.2
• /
• pp.1-9
• /
• 2005

Flow-induced dilation of blood vessel is the result of a series of bioreaction in vascular endothelial cells(VEC). Shear stress change by blood flow in human artery or vein is sensed by the mechanoreceptor and responsible for such a chain reaction. The inositol(1,4,5)-triphophate($IP_3$) is produced in the first stage to elevate permeability of the intercellular membrane to calcium ions by which the cytosolic calcium concentration is consequently increased. This intracellular calcium transient triggers synthesis of EDRF and prostacyclin. The mathematical model of this VEC calcium dynamics is reproduced from the literature. We then use the Computational Fluid Dynamics(CFD) technique to investigate the blood stream dictating the VEC calcium dynamics. The pulsatile blood flow in a stenosed blood vessel is considered here as a part of study on thrombogenesis. We calculate the pulsating shear stress (thus its temporal change) distributed over the stenosed artery that is implemented to the VEC calcium dynamics model. It has been found that the pulsatile shear stress induces larger intracellular $Ca^{2+}$ transient plus much higher amount of EDRF and prostacyclin release in comparison with the steady shear stress case. It is concluded that pulsatility of the physiological shear stress is important to keep the vasodilation function in the stenosed part of the blood vessel.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.129-136
• /
• 2019

There are a variety of entities in natural language such as people, organizations, places, and products. These entities can have many various meanings. The ambiguity of entity is a very challenging task in the field of natural language processing. Entity Linking(EL) is the task of linking the entity in the text to the appropriate entity in the knowledge base. Pairwise based approach, which is a representative method for solving the EL, is a method of solving the EL by using the association between two entities in a sentence. This method considers only the interdependence between entities appearing in the same sentence, and thus has a limitation of global interdependence. In this paper, we developed an Entity2vec model that uses Word2vec based on knowledge base of RDF type in order to solve the EL. And we applied the algorithms using the generated model and ranked each entity. In this paper, to overcome the limitations of a pairwise approach, we devised a pairwise approach based on comprehensive interdependency and compared it.

• Journal of Software Assessment and Valuation
• /
• v.16 no.2
• /
• pp.45-52
• /
• 2020

Traditional methods of detecting security vulnerabilities in source-code require a lot of time and effort. If there is good data, the issue could be solved by using the data with machine learning. Thus, this paper proposes a source-code vulnerability detection method based on machine learning. Our method employs the code2vec model that has been used to propose the names of methods, and uses as a data set, Juliet Test Suite that is a collection of common security vulnerabilities. The evaluation shows that our method has high precision of 97.3% and recall rates of 98.6%. And the result of detecting vulnerabilities in open source project shows hopeful potential. In addition, it is expected that further progress can be made through studies covering with vulnerabilities and languages not addressed here.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.10
• /
• pp.397-402
• /
• 2019

Accurately predicting cancer prognosis to provide appropriate treatment strategies for patients is one of the critical challenges in bioinformatics. Many researches have suggested machine learning models to predict patients' outcomes based on their gene expression data. Gene expression data is high-dimensional numerical data containing about 17,000 genes, so traditional researches used feature selection or dimensionality reduction approaches to elevate the performance of prognostic prediction models. These approaches, however, have an issue of making it difficult for the predictive models to grasp any biological interaction between the selected genes because feature selection and model training stages are performed independently. In this paper, we propose a novel two-dimensional image formatting approach for gene expression data to achieve feature selection and prognostic prediction effectively. Node2Vec is exploited to integrate biological interaction network and gene expression data and a convolutional neural network learns the integrated two-dimensional gene expression image data and predicts cancer prognosis. We evaluated our proposed model through double cross-validation and confirmed superior prognostic prediction accuracy to traditional machine learning models based on raw gene expression data. As our proposed approach is able to improve prediction models without loss of information caused by feature selection steps, we expect this will contribute to development of personalized medicine.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.3
• /
• pp.355-360
• /
• 2021

Recently Youtube has been more popular. As many studies show the comparative deprivation of the Social Medeia, this study looks into whether the comparative deprivation is expressed on the YouTube comments. It focuses on the Luxury Haul contents, videos about huge amounts of luxurious products, of which Youtubers'economic feature are demonstrative. The comments of the videos are analyzed with LDA TF-IDF and Word2Vec. Additionally, the comments were classified into positive and negative groups by the LSTM model as well. As a result of the study, even though many comments turned out positive, the negative keywords were indicated related to comparative deprivation. Also it was found that the viewers compared themselves with Youtubers. In particular, some YouTubers are more criticized if they are younger or does not seem to afford the luxurious products themselves. This study suggests that the users express the comparative deprivation on YouTube as well like on the other Social Media.