• International Journal of Computer Science & Network Security
• /
• 제23권4호
• /
• pp.1-6
• /
• 2023

Morphological analysis is a branch of natural language processing, is now a rapidly growing field. The fundamental tenet of morphological analysis is that it can establish the roots or stems of words and enable comparison to the original term. Arabic is a highly inflected and derivational language and it has a strong structure. Each root or stem can have a large number of affixes attached to it due to the non-concatenative nature of Arabic morphology, increasing the number of possible inflected words that can be created. Accurate verb recognition and extraction are necessary nearly all issues in well-known study topics include Web Search, Information Retrieval, Machine Translation, Question Answering and so forth. in this work we have designed and implemented an algorithm to detect and recognize Arbic Verbs from Arabic text.The suggested technique was created with "Python" and the "pyqt5" visual package, allowing for quick modification and easy addition of new patterns. We employed 17 alternative patterns to represent all verbs in terms of singular, plural, masculine, and feminine pronouns as well as past, present, and imperative verb tenses. All of the verbs that matched these patterns were used when a verb has a root, and the outcomes were reliable. The approach is able to recognize all verbs with the same structure without requiring any alterations to the code or design. The verbs that are not recognized by our method have no antecedents in the Arabic roots. According to our work, the strategy can rapidly and precisely identify verbs with roots, but it cannot be used to identify verbs that are not in the Arabic language. We advise employing a hybrid approach that combines many principles as a result.

• Structural Engineering and Mechanics
• /
• 제86권4호
• /
• pp.473-486
• /
• 2023

Nowadays, urban centers are increasingly vertical, making architects and engineers look for more efficient tools to analyze the effects of wind on tall buildings. Topology optimization can be used as an efficient tool for the design of bracing systems. Therefore, this work obtained the wind loads that act in the CAARC building, following the Brazilian standard NBR 6123/1988 and using Computational Fluid Dynamics. Four loading situations were considered, using the SIMP and BESO methods to optimize two-dimensional structures. A comparison between the SIMP and BESO methods is presented, showing the differences in the geometry of the solution found by both methods, the percentage variation in the objective function values and the dimensionless processing time. The solutions obtained through the loads obtained by the Brazilian standard are also compared with the numerical solutions obtained by CFD. The results show that the BESO method presented more rigid structures compared to the SIMP method. The bracing structures obtained with the SIMP method always present similar patterns in the distribution and quantity of bars, in contrast to the BESO method where no characteristic topology pattern was observed. It was concluded that even though the structures obtained by the BESO method presented greater stiffness, the SIMP method was less susceptible to the methodology used for the determination of wind loads. Additionally, it was evident the great potential that the combination topology optimization and computational wind engineering have in the design of bracing systems of high functional and aesthetic standards.

• Imaging Science in Dentistry
• /
• 제53권2호
• /
• pp.109-115
• /
• 2023

Purpose: The aim of this study was to evaluate changes in the trabecular bone through texture analysis and compare the texture analysis characteristics of different areas in patients with medication-related osteonecrosis of the jaw (MRONJ). Materials and Methods: Cone-beam computed tomographic images of 16 patients diagnosed with MRONJ were used. In sagittal images, 3 regions were chosen: active osteonecrosis(AO); intermediate tissue (IT), which presented a zone of apparently healthy tissue adjacent to the AO area; and healthy bone tissue (HT) (control area). Texture analysis was performed evaluating 7 parameters: secondary angular momentum, contrast, correlation, sum of squares, inverse moment of difference, sum of entropies, and entropy. Data were analyzed using the Kruskal-Wallis test with a significance level of 5%. Results: Comparing the areas of AO, IT, and HT, significant differences (P<0.05) were observed. The IT and AO area images showed higher values for parameters such as contrast, entropy, and secondary angular momentum than the HT area, indicating greater disorder in these tissues. Conclusion: Through texture analysis, changes in the bone pattern could be observed in areas of osteonecrosis. The texture analysis demonstrated that areas visually identified and classified as IT still had necrotic tissue, thereby increasing the accuracy of delimiting the real extension of MRONJ.

• 문화기술의 융합
• /
• 제9권3호
• /
• pp.835-843
• /
• 2023

XML 데이터의 구조를 고려할 때 경로 및 트리 패턴 매칭 알고리즘은 XML 질의 처리에 중요한 역할을 하고 있다. 노드 간의 결정 또는 관계를 용이하게 하기 위해 XML 트리의 노드는 일반적으로 두 노드 간의 조상-후손 관계를 신속하게 설정할 수 있는 방식으로 레이블링된다. 그러나 이러한 기법은 순서에 따른 업데이트로 삽입이 발생할 경우 기존 노드에 레이블을 다시 지정하거나 특정 값을 다시 계산해야 하는 단점이 발생한다. 따라서 현재 레이블링 기법들에서는 레이블을 업데이트 하는 비용이 매우 높다. 본 논문에서는 재레이블링 또는 재계산 없이 순서에 민감한 XML 문서의 업데이트를 지원하는 Fast XML 인코딩 기법이라는 새로운 레이블링을 제안한다. 또한 XML 트리의 동일한 위치에서 삭제된 레이블을 재사용하여 레이블의 길이를 제어한다. 제안한 재사용 알고리즘은 삭제된 모든 레이블을 동일한 위치에 삽입할 때 레이블의 길이를 줄일 수 있다. 실험 결과에서 제안된 기법은 순서에 민감한 질의 및 업데이트를 효율적으로 처리할 수 있다.

• 복식문화연구
• /
• 제31권1호
• /
• pp.53-69
• /
• 2023

This study aims to analyze research trends regarding outdoor wear. For this purpose, the data-collection period was limited to January 2002-October 2022, and the collection consisted of titles of papers, academic names, abstracts, and publication years from the Research Information Sharing Service (RISS). Frequency analysis was conducted on 227 papers in total to check academic journals and annual trends, and LDA topic-modeling analysis was conducted using 20,964 tokens. Data pre-processing was performed prior to topic-modeling analysis; after that, topic-modeling analysis, core topic derivation, and visualization were performed using a Python algorithm. A total of eight topics were obtained from the comprehensive analysis: experiential marketing and lifestyle, property and evaluation of outdoor wear, design and patterns of outdoor wear, outdoor-wear purchase behavior, color, designs and materials of outdoor wear, promotional strategies for outdoor wear, purchase intention and satisfaction depending on the brand image of outdoor wear, differences in outdoor wear preferences by consumer group. The results of topic-modeling analysis revealed that the topic, which includes a study on the design and material of outdoor wear and the pattern of jackets related to the overall shape, was the highest at 30.9% of the total topics. The next highest topic was also the design and color of outdoor wear, indicating that design-related research was the main research topic in outdoor wear research. It is hoped that analyzing outdoor wear research will help comprehend the research conducted thus far and reveal future directions.

• 복식문화연구
• /
• 제31권1호
• /
• pp.124-140
• /
• 2023

This study analyzes traditional women's Hu costumes of the Tang dynasty, and deploys a creative fashion design to converge contemporary and traditional styles. In this costume, women wear a robe with striped or plain patterns in the lower part of the pants, and it appears frequently in red and yellow colors. Depending on the sleeve, it is either a round collar or a turn down collar robe. In the Hu hat, the huntuomao and juanyanxumao were leather and mili and weimao were used to prevent the sand from flowing. This study uses the CLO 3D program with the "moment" theme based on the Hu costume for women to deploy 4 pairs of fashion design and to produce works for 2 pairs. The 3D virtual clothing program demonstrates important effects in design deployment and pattern arrangement through its efficiency and convenience of clothing production. The CLO 3D program was closely combined with the 2D design and the 3D affect, and it heightened the efficiency in saving the processing time and energy of the sample clothes. Through facilitating the 3D digital fashion design, the production may reduce time needed and contribute to an effective economy, and it may compare digital fashion design to actual products as well as illustrate the potential of digital fashion design.

• Nuclear Engineering and Technology
• /
• 제55권10호
• /
• pp.3775-3786
• /
• 2023

Mixing Vane Grid (MVG) is one of the most important structures in fuel assembly due to its high performance in mixing the coolant and ultimately increasing Critical Heat Flux (CHF), which avoids the temperature rising suddenly of fuel rods. To evaluate the mixing performance of the MVG, a Total Diffusion Coefficient (TDC) mixing coefficient is defined in the subchannel analysis code. Conventionally, the TDC of the spacer grid is obtained from the combination of experiments and subchannel analysis. However, the processing of obtaining and determine a reasonable TDC is much challenging, it is affected by boundary conditions and MVG geometries. In is difficult to perform all the large and costing rod bundle tests. In this paper, the CFD method was applied in TDC analysis. A typical 5 × 5 MVG was simulated and validated to estimate the mixing performance of the MVG. The subchannel code was used to calculate the TDC. Firstly, the CFD method was validated from the aspect of pressure drop and lateral temperature distribution in the subchannels. Then the effect of boundary conditions including the inlet temperature, inlet velocities, heat flux ratio between hot and cold rods and the arrangement of hot and cold rods on MVG mixing and TDC were studied. The geometric effects on mixing are also carried out in this paper. The effect of vane pattern on mixing was investigated to determine which one is the best to represent the grid's mixing performance.

• Tribology and Lubricants
• /
• 제39권5호
• /
• pp.197-202
• /
• 2023

Recently, research on experimental and analytical techniques utilizing microfluidic devices has been pursued. For example, lab-on-a-chip devices that integrate micro-devices onto a single chip for processing small sample quantities have gained significant attention. However, during sample preparation, unnecessary gases can be introduced into the internal channels, thus, impeding device flow and compromising specific function efficiency, including that of analysis and separation. Several methods have been proposed to mitigate this issue, however, many involve cumbersome procedures or suffer from complexities owing to intricate structures. Recently, some approaches have been introduced that utilize hydrophobic device structures to remove gases within channels. In such cases, the permeability of gases passing through the structure becomes a crucial performance factor. In this study, a method involving the deposition and sintering of diluted Ag-ink onto a silicon wafer surface is presented. This is followed by unstructured nano-pattern creation using a Metal Assisted Chemical Etching (MACE) process, which yields a nanostructured surface with unstructured pillar shapes. Subsequently, gas permeability in the spaces formed by these surface structures is investigated. This is achieved by experiments conducted to incorporate a pressure chamber and measure gas permeability. Trends are subsequently analyzed by comparing the results with existing theories. Finally, it can be confirmed that the significance of this study primarily lies in its capability to effectively evaluate gas permeability through unstructured pillar-like nanostructures, thus, providing quantitative values for the appropriate driving pressure and expected gas removal time in practical device operation.

• 정보화정책
• /
• 제30권3호
• /
• pp.69-91
• /
• 2023

기술 발전이 고용에 미치는 영향은 자동화에 의한 대체 또는 새로운 업무 도입에 따른 고용 증가 등 여전한 논쟁의 대상이다. 특히 인공지능 기술 발전과 고용에 대한 실증 논의는 더욱 부족한 실정이다. 이에 본 연구는 자연어처리 기법(SBERT)과 특허를 이용하여 직업별 인공지능 노출 점수를 계산하고 평균 점수를 기준으로 상위 집단과 하위 집단으로 구분하여 집단별 고용 추세를 분석한다. 자연어처리 기법을 통해 한국 특허와 미국 직업의 업무 설명을 연계하는 인공지능 노출 점수 계산 방식과 한미 표준직업분류 연계 방식을 제시하고 이를 국내 고용 통계에 적용하여 추세를 분석한다. 2013년 이후 국내 인공지능 출원 특허와 통계청 지역별고용조사를 분석한 결과 한국의 고용은 시간이 지남에 따라 평균 이상의 인공지능 노출 집단에서 우상향하고, 평균 이하 집단에서는 우하향하는 K자형 양극화 양상을 보인다.

• 경영정보학연구
• /
• 제25권3호
• /
• pp.27-41
• /
• 2023

최근 많은 분야에서 기계학습에 대한 연구가 활발히 진행되고 있는데, 상당수의 연구들이 학습 모델의 성능을 개선하는 최신 방법론을 제시하고 있다. 본 연구에서는 방법론의 개발 못지않게 기계학습에 투입되는 훈련용 데이터의 '품질'을 개선하는 것 역시 중요하다는 점에 착안하여, 코퍼스 분석에서 자주 사용되는 '부분 문법' 처리 프로세스를 통해 훈련 데이터의 품질을 향상시키는 방법을 제시한다. 우리나라 100대 기업에 근무하는 재직자들이 채용플랫폼에 게시하는 방대한 양의 비정형 기업 리뷰 텍스트 데이터를 수집하고, 데이터 품질을 부분 문법 프로세스로 개선한 후, 부분 문법이 적용된 분류 모델이 적용되지 않은 모델보다 분류 성능이 우수함을 확인하였다. 분류 카테고리는 직원 몰입의 5가지 요인으로 상정하였는데, 국내 직장인들이 기업 리뷰가 각 유형별로 빈도에 차이가 있는지를 분석하였다. 추가로 리뷰 양상이 코로나 팬데믹 전후로 어떠한 변화가 있었는지도 분석하였다. 본 연구를 통해 국내 직장인들의 생생한 일터 경험들을 자동적으로 식별하고 분류하여, 이직을 포함한 주요한 조직문화 현상의 행태와 유발 원인 등을 유추해 볼 수 있는 근거를 제공한다.