• 한국IT서비스학회지
• /
• 제18권1호
• /
• pp.219-234
• /
• 2019

Fake news threaten democratic elections and causes social conflicts, resulting in major damage. However, the concept of fake news is hard to define, as there is a saying, "News is not fake, fake is not news." Fake news, however, has irreversible characteristics that can not be recovered or reversed completely through post-punishment of economic and political benefits. It is also rapidly spreading in the early days. Therefore, it is very important to preemptively detect these types of articles and prevent their blind proliferation. The existing countermeasures are focused on reporting fake news, raising the level of punishment, and the media & academia to determine the authenticity of the news. Researchers are also trying to determine the authenticity by analyzing its contents. Apart from the contents of fake news, determining the behavioral characteristics of the promoters and its qualities can help identify the possibility of having fake news in advance. The online community has a fake news interception and response tradition through its long-standing community-based activities. As a result, I attempted to model the fake news by analyzing the affirmation-denial analysis and posting behavior by securing the web board crawl of the 'M community' bulletin board during the 2017 Korean presidential election period. Random forest algorithm deemed significant. The results of this research will help counteract fake news and focus on preemptive blocking through behavioral analysis rather than post-judgment after semantic analysis.

• 한국멀티미디어학회논문지
• /
• 제24권5호
• /
• pp.659-666
• /
• 2021

With the continuous development of deep learning, human behavior recognition algorithms have achieved good results. However, in a multi-person recognition environment, the complex behavior environment poses a great challenge to the efficiency of recognition. To this end, this paper proposes a multi-person pose estimation model. First of all, the human detectors in the top-down framework mostly use the two-stage target detection model, which runs slow down. The single-stage YOLOv3 target detection model is used to effectively improve the running speed and the generalization of the model. Depth separable convolution, which further improves the speed of target detection and improves the model's ability to extract target proposed regions; Secondly, based on the feature pyramid network combined with context semantic information in the pose estimation model, the OHEM algorithm is used to solve difficult key point detection problems, and the accuracy of multi-person pose estimation is improved; Finally, the Euclidean distance is used to calculate the spatial distance between key points, to determine the similarity of postures in the frame, and to eliminate redundant postures.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권2호
• /
• pp.206-211
• /
• 2022

As the level of consumption is improved and cultural life is pursued, the consumer's consciousness structure is rapidly changing, and the demand for product selection level, variety, and quality is becoming more diverse. The restaurant economy is falling due to the prolonged COVID-19, the economic recession, income decline, and changes in population structure and lifestyle, but the Meal- kit market is growing rapidly. This study aims to identify the consumer perception of Meal-kit, which is rapidly growing as an alternative to existing meals in the fields of dining out, food, and distribution due to the development of technology and social environment using big data. As a result of the analysis, the keywords with the highest frequency of appearance were in the order of Meal-kit, Cooking, Product, Launching, and Market and were divided into 8 groups through the CONCOR analysis. We want to identify consumer trends related to the key keywords of Meal-kit, present effective data related to Meal-kit demand for Meal-kit specialized companies, and provide implications for establishing marketing strategies for differentiated competitive advantage.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권3호
• /
• pp.771-791
• /
• 2022

Protein-protein interaction (PPI) extraction from original text is important for revealing the molecular mechanism of biological processes. With the rapid growth of biomedical literature, manually extracting PPI has become more time-consuming and laborious. Therefore, the automatic PPI extraction from the raw literature through natural language processing technology has attracted the attention of the majority of researchers. We propose a PPI extraction model based on the large pre-trained language model and adversarial training. It enhances the learning of semantic and syntactic features using BioBERT pre-trained weights, which are built on large-scale domain corpora, and adversarial perturbations are applied to the embedding layer to improve the robustness of the model. Experimental results showed that the proposed model achieved the highest F1 scores (83.93% and 90.31%) on two corpora with large sample sizes, namely, AIMed and BioInfer, respectively, compared with the previous method. It also achieved comparable performance on three corpora with small sample sizes, namely, HPRD50, IEPA, and LLL.

• ETRI Journal
• /
• 제44권3호
• /
• pp.450-461
• /
• 2022

The Gen-Z protocol is a memory semantic protocol between the memory and CPU used in computer architectures with large memory pools. This study presents the implementation of the Gen-Z hardware system configured using Gen-Z specification 1.0 and reports its performance. A hardware prototype of a DDR4 Gen-Z memory pool with an optimized character, a block device driver, and a file system for the Gen-Z hardware was designed. The Gen-Z IP was targeted to the FPGA, and a 512 GB Gen-Z memory pool was configured on an ×86 server. In the experiments, the latency and throughput of the Gen-Z memory were measured and compared with those of the local memory, SATA SSD, and NVMe using character or block device interfaces. The Gen-Z hardware exhibited superior throughput and latency performance compared with SATA SSD and NVMe at block sizes under 4 kB. The MySQL and File IO benchmark of Gen-Z showed good write performance in all block sizes and threads. Besides, it showed low latency in RocksDB's fillseq dbbench using the ext4 direct access filesystem.

• International Journal of Computer Science & Network Security
• /
• 제21권12spc호
• /
• pp.521-525
• /
• 2021

An essential area is the creation of a single digital market between the EU and Ukraine through information technology. Purpose: to investigate and analyze civil law relations in the field of adaptation of Ukrainian civil law to civil law regulations of the EU. The object of research: Ukrainian civil law and civil law of the EU. The subject of the study is civil law in the context of adaptation of civil law to the legislation of the EU. The following methods of scientific cognition were used during the research: semantic, historical, comparison, analysis and synthesis, generalization. The results of the study show that the harmonization of the legal system of Ukraine with EU law is caused by several goals: successful integration of Ukraine into the EU, legal reforms based on the positive example of EU countries, promoting access of Ukrainian enterprises to the EU market; attracting foreign investment, increasing the welfare of Ukrainian citizens. The adaptation includes three stages, the final of which is the preparation of an expanded program of harmonization of Ukrainian legislation with EU legislation. In the process of adaptation, it is important to take into account the legal history, tradition, features and mentality of Ukraine and before borrowing legal structures to analyze the feasibility of their application in the Ukrainian legal field.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권11호
• /
• pp.3723-3737
• /
• 2022

Google Play is one of the largest Android phone app markets and it contains both free and paid apps. It provides a variety of categories for every target user who has different needs and purposes. The customer's rate every product based on their experience of apps and based on the average rating the position of an app in these arch varies. Fraudulent behaviors emerge in those apps which incorporate search rank maltreatment and malware proliferation. To distinguish the fraudulent behavior, a novel framework is structured that finds and uses follows left behind by fraudsters, to identify both malware and applications exposed to the search rank fraud method. This strategy correlates survey exercises and remarkably joins identified review relations with semantic and behavioral signals produced from Google Play application information, to distinguish dubious applications. The proposed model accomplishes 90% precision in grouping gathered informational indexes of malware, fakes, and authentic apps. It finds many fraudulent applications that right now avoid Google Bouncers recognition technology. It also helped the discovery of fake reviews using the reviewer relationship amount of reviews which are forced as positive reviews for each reviewed Google play the android app.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.267-268
• /
• 2022

최근, 빅데이터 분석은 기계학습의 발전에 따른 다양한 기법들을 이용할 수 있다. 현실에서 수집된 빅데이터는 단어 간의 관계성에 대한 의미적 분석을 바탕으로 같거나 유사한 용어에 대한 자동화된 정제기법이 부족하다. 빅데이터는 보통 문장의 형태로 구성되어 있고, 이에 대한 형태소 분석이나 문장의 이해가 필요하다. 이에 자연어를 분석하기 위한 기법인 NLP는 단어의 관계성과 문장을 이해할 수 있다. 본 논문에서는 빅데이터를 시계열 접근법인 RNN의 단점을 보완한 기법인 트랜스포머와 리포머의 장단점에 대해 연구한다.

• 국제학술발표논문집
• /
• The 7th International Conference on Construction Engineering and Project Management Summit Forum on Sustainable Construction and Management
• /
• pp.52-61
• /
• 2017

Considering its significant impact on the cost and schedule of construction projects, formwork as one part of temporary facility categories in construction should be arranged precisely. Current practice in the formwork planning is often conducted manually and repetitively, causing low efficiency and time waste. This study proposes an automated framework to generate more accurate and detailed formwork plans by utilizing information from building information modeling (BIM) considering the adequate geometric and semantic information provided by the BIM model. The dimensions and quantities information of elements in a building can be extracted automatically. Then, a rule is prepared for calculating the required forms erected around elements based on the contact areas. Finally, an algorithm of integrating first fit decreasing (FFD) with coordinated bottom left (CBL) is applied to automatically generate the formwork plan. The BIM-based automated planning framework is demonstrated by an illustrative example. The results show that the proposed framework can generate the formwork plan accurately and automatically, and significantly improve the efficiency in the formwork plan and reuse.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2023년도 제67차 동계학술대회논문집 31권1호
• /
• pp.437-438
• /
• 2023

This paper takes the Pingjiang historical and cultural district of Suzhou city as an example, collects 1439 visitor review data from Ctrip.com with the help of Python technology, and uses web text analysis to conduct research on high-frequency words, semantic networks and emotional tendencies to comprehensively assess the tourist perception of the Grand Canal heritage. The study found that: natural and humanistic landscape, historical and cultural accumulation, and the style of Jiangnan Canal are fully reflected in the tourists' perception of Pingjiang historical and cultural district; tourists hold strong positive emotion towards Pingjiang Road, however, there is still more room for renovation and improvement of the historical and cultural district. Finally, countermeasure suggestions for improving the tourist perception of the Grand Canal heritage are given in terms of protection first, cultural integration and innovative utilization.