• Journal of Multimedia Information System
• /
• 제5권3호
• /
• pp.147-154
• /
• 2018

Natural language processing (NLP) is an emerging research area in which we study how machines can be used to perceive and alter the text written in natural languages. We can perform different tasks on natural languages by analyzing them through various annotational tasks like parsing, chunking, part-of-speech tagging and lexical analysis etc. These annotational tasks depend on morphological structure of a particular natural language. The focus of this work is part-of-speech tagging (POS tagging) on Hindi language. Part-of-speech tagging also known as grammatical tagging is a process of assigning different grammatical categories to each word of a given text. These grammatical categories can be noun, verb, time, date, number etc. Hindi is the most widely used and official language of India. It is also among the top five most spoken languages of the world. For English and other languages, a diverse range of POS taggers are available, but these POS taggers can not be applied on the Hindi language as Hindi is one of the most morphologically rich language. Furthermore there is a significant difference between the morphological structures of these languages. Thus in this work, a POS tagger system is presented for the Hindi language. For Hindi POS tagging a hybrid approach is presented in this paper which combines "Probability-based and Rule-based" approaches. For known word tagging a Unigram model of probability class is used, whereas for tagging unknown words various lexical and contextual features are used. Various finite state machine automata are constructed for demonstrating different rules and then regular expressions are used to implement these rules. A tagset is also prepared for this task, which contains 29 standard part-of-speech tags. The tagset also includes two unique tags, i.e., date tag and time tag. These date and time tags support all possible formats. Regular expressions are used to implement all pattern based tags like time, date, number and special symbols. The aim of the presented approach is to increase the correctness of an automatic Hindi POS tagging while bounding the requirement of a large human-made corpus. This hybrid approach uses a probability-based model to increase automatic tagging and a rule-based model to bound the requirement of an already trained corpus. This approach is based on very small labeled training set (around 9,000 words) and yields 96.54% of best precision and 95.08% of average precision. The approach also yields best accuracy of 91.39% and an average accuracy of 88.15%.

• 인터넷정보학회논문지
• /
• 제22권2호
• /
• pp.29-39
• /
• 2021

새롭게 발생되는 사이버 공격으로 인해 개인, 민간 및 기업의 피해가 증가함에 따라, 이에 기반이 되는 네트워크 보안 문제는 컴퓨터 시스템의 주요 문제로 부각되었다. 이에 기존에 사용되는 네트워크 침입 탐지 시스템(Network Intrusion Detection System: NIDS)에서 발생되는 한계점을 개선하고자 기계 학습과 딥러닝을 활용한 연구 이뤄지고 있다. 이에 본 연구에서는 CNN(Convolution Neural Network) 알고리즘을 이용한 NIDS 모델 연구를 진행한다. 이미지 분류 기반의 CNN 알고리즘 학습을 위해 기존 사용되는 전처리 단계에서 연속성 변수 이산화(Discretization of Continuous) 알고리즘을 추가하여 예측 변수에 대해 선형 관계로 표현하여 해석에 용이한 데이터로 변환 후, 정사각형 행렬(Square Matrix) 구조에 매칭된 픽셀(Pixel) 이미지 구조를 모델에 학습한다. 모델의 성능 평가를 위해 네트워크 패킷 데이터인 NSL-KDD를 사용하였으며, 정확도(Accuracy), 정밀도(Precision), 재현율(Recall) 및 조화평균(F1-score)을 성능지표로 사용하였다. 실험 결과 제안된 모델에서 85%의 정확도로 가장 높은 성능을 보였으며, 학습 표본이 적은 R2L 클래스의 조화평균이 71% 성능으로 다른 모델에 비해서 매우 좋은 성능을 보였다.

• 지능정보연구
• /
• 제24권4호
• /
• pp.111-136
• /
• 2018

지식베이스를 구축하는 작업은 도메인 전문가가 온톨로지 스키마를 이해한 뒤, 직접 지식을 정제하는 수작업이 요구되는 만큼 비용이 많이 드는 활동이다. 이에, 도메인 전문가 없이 다양한 웹 환경으로부터 질의에 대한 답변 정보를 추출하기 위한 자동화된 시스템의 연구개발의 필요성이 제기되고 있다. 기존의 정보 추출 관련 연구들은 웹에 존재하는 다양한 형태의 문서 중 학습데이터와 상이한 형태의 문서에서는 정보를 효과적으로 추출하기 어렵다는 한계점이 존재한다. 또한, 기계 독해와 관련된 연구들은 문서에 정답이 있는 경우를 가정하고 질의에 대한 답변정보를 추출하는 경우로서, 문서의 정답포함 여부를 보장할 수 없는 실제 웹의 비정형 문서로부터의 정보추출에서는 낮은 성능을 보인다는 한계점이 존재한다. 본 연구에서는 지식베이스 확장을 위하여 웹에 존재하는 멀티소스 비정형 문서로부터 질의에 대한 정보를 추출하기 위한 시스템의 개발 방법론을 제안하고자 한다. 본 연구에서 제안한 방법론은 "주어(Subject)-서술어(Predicate)"로 구분된 질의에 대하여 위키피디아, 네이버 백과사전, 네이버 뉴스 3개 웹 소스로부터 수집된 비정형 문서로부터 관련 정보를 추출하며, 제안된 방법론을 적용한 시스템의 성능평가를 위하여, Wu and Weld(2007)의 모델을 베이스라인 모델로 선정하여 성능을 비교분석 하였다. 연구결과 제안된 모델이 베이스라인 모델에 비해, 위키피디아, 네이버 백과사전, 네이버 뉴스 등 다양한 형태의 문서에서 정보를 효과적으로 추출하는 강건한 모델임을 입증하였다. 본 연구의 결과는 현업 지식베이스 관리자에게 지식베이스 확장을 위한 웹에서 질의에 대한 답변정보를 추출하기 위한 시스템 개발의 지침서로서 실무적인 시사점을 제공함과 동시에, 추후 다양한 형태의 질의응답 시스템 및 정보추출 연구로의 확장에 기여할 수 있을 것으로 기대한다.

• 산업경영시스템학회지
• /
• 제39권4호
• /
• pp.90-96
• /
• 2016

Incremental sheet forming (ISF) is a highly versatile and flexible process for rapid manufacturing of complex sheet metal parts. Compared to conventional sheet forming processes, ISF is of a clear advantage in manufacturing small batch or customized parts. ISF needs die-less machine alone, while conventional sheet forming requires highly expensive facilities like dies, molds, and presses. This equipment takes long time to get preparation for manufacturing. However, ISF does not need the full facilities nor much cost and time. Because of the facts, ISF is continuously being used for small batch or prototyping manufacturing in current industries. However, spring-back induced in the process of incremental forming becomes a critical drawback on precision manufacturing. Since sheet metal, being a raw material for ISF, has property to resilience, spring-back would come in the case. It is the research objective to investigate how geometrical shaping parameters make effect on shape dimensional errors. In order to analyze the spring-back occurred in the process, this study experimented on Al 1015 material in the ISF. The statistical tool employed experimental design with factors. The table of orthogonal arrays of $L_8(2^7)$ are used to design the experiments and ANOVA method are employed to statistically analyze the collected data. The results of the analysis from this study shows that the type of shape and the slope of bottom are the significant, whereas the shape size, the shape height, and the side angle are not significant factors on dimensional errors. More error incurred on the pyramid than on the circular type in the experiments. The sloped bottom showed higher errors than the flat one.

• 한국소음진동공학회논문집
• /
• 제14권12호
• /
• pp.1322-1329
• /
• 2004

End-milling has been used widely in industrial system because it is effective to a material manufacturing with various shapes. Recently the end-milling processing is needed the high-precise technique with good surface roughness and rapid time in precision machine part and electronic part. The optimum mechanical vibration of main spindle, surface roughness and cutting temperature have an effect on end-milling condition such as, cutting direction, revolution of spindle, feed rate and depth of cut, etc. Therefore, this study carried to decide the working condition for optimum mechanical vibration of main spindle, surface roughness and cutting temperature using design of experiments, ANOVA and characteristic function. From the results of experimentation, mechanical vibration has an effect on revolution of spindle, radial depth of cut, and axial depth of cut. The surface roughness has an effect on cutting direction, revolution of spindle and depth of cut. And then the optimum condition used design of experiments is upward cutting In cutting direction, 600 rpm in revolution of spindle, 240 mm/min in feed rate, 2 mm in axial depth of cut and 0.25 mm in radial depth of cut. By design of experiments and characteristic function, it is effectively represented shape characteristics of mechanical vibration, surface roughness and cutting temperature in end-milling.

• 한국기계가공학회지
• /
• 제14권3호
• /
• pp.36-42
• /
• 2015

The aspherical lens was designed to be able to array a focal point. For this reason, it has very curved surface. The aspherical lens is fabricated by injection molding or diamond turning machine. With the aspherical lens, tool marks and surface roughness affect the optical characteristics, such as transmissivity. However, it is difficult to polish free form surface shapes uniformly with conventional methods. Therefore, in this paper, the ultra-precision polishing method with MR fluid was used to polish an aspherical lens with 4-axis position control systems. A Tool path and polishing mechanism were developed to polish the aspherical lens shape. An MR polishing experiment was performed using a generated tool path with a PMMA aspherical lens after the turning process. As a result, surface roughness was improved from $R_a=40.99nm$, $R_{max}=357.1nm$ to $R_a=4.54nm$, $R_{max}=35.72nm$. Finally, the MR polishing system can be applied to the finishing process of fabrication of the aspherical lens.

• 한국언어정보학회:학술대회논문집
• /
• 한국언어정보학회 2007년도 정기학술대회
• /
• pp.489-497
• /
• 2007

Named entity translation plays an important role in many applications, such as information retrieval and machine translation. In this paper, we focus on translating person names, the most common type of name entity in Korean-Chinese cross language information retrieval (KCIR). Unlike other languages, Chinese uses characters (ideographs), which makes person name translation difficult because one syllable may map to several Chinese characters. We propose an effective hybrid person name translation method to improve the performance of KCIR. First, we use Wikipedia as a translation tool based on the inter-language links between the Korean edition and the Chinese or English editions. Second, we adopt the Naver people search engine to find the query name's Chinese or English translation. Third, we extract Korean-English transliteration pairs from Google snippets, and then search for the English-Chinese transliteration in the database of Taiwan's Central News Agency or in Google. The performance of KCIR using our method is over five times better than that of a dictionary-based system. The mean average precision is 0.3490 and the average recall is 0.7534. The method can deal with Chinese, Japanese, Korean, as well as non-CJK person name translation from Korean to Chinese. Hence, it substantially improves the performance of KCIR.

• 한국조명전기설비학회지:조명전기설비
• /
• 제11권1호
• /
• pp.106-113
• /
• 1997

기계기구의 설계나 개발시 개발제품의 표준화 및 양호환 호환성, 공작상 공기의 단축, 가격 저하 등의 장점을 갖는 계열화 설계기법으로 CAE에 의한 방법을 소개했다. 특히 대형기기에서부터 초정밀 기기 등에 대한 설계나 개발시 표본 모델을 선정하고 그로부터 물리적이고 기술적인 부분의 면밀한 분석을 통해서 고유한 특징을 추출하고 이에 대한 유추론적 수학의 해석으로서 추구하고자 하는 크기에 대한 특성 실험을 예측할 수 있고 최적 설계상의 데이터를 사전에 검증하므로서 수요자가 요구하는 설계 표본을 이끌어 낼 수 있다. 본 연구에서는 설계자가 계산기를 이용한 계열화 설계에 있어서의 유츄론적 알고리즘과 설계정수 및 제한요소들의 처리 방법을 제시하고 각종 임의의 계열화 기기에 대한 단계별 기능 및 기술적인 데이터의 유도 과정을 액츄에이터를 모델로 선정하여 분석했다.

• 산업경영시스템학회지
• /
• 제43권4호
• /
• pp.41-47
• /
• 2020

As modern industries are highly being developed, it is required that mechanical parts have to be manufactured with a high precision. In order to have precise parts, error-free designs have to be done before manufacturing with accuracy. For this intention being fulfilled, a mechanical analysis is essential for design proof. Nowadays, FEM simulation is a popular tool for verifying a machine design. In this paper, an impeller, being utilized in a compressor or an oil mixer as an actuator, is studied for an evaluation. The purpose of this study is to present a safety of an impeller for a proof of its mechanical stability. A static analysis for stress, strain, and deformation within a regular usage is examined. This simulation test shows 357.26×106 Pa for maximum equivalent stress and 0.207mm for total deformation. A fatigue test is carried to provide durability and its result shows that minimum safety factor is 3.2889, which guarantees that it runs without a fatigue failure in 106 cycles. The natural frequencies for the impeller is ranged from 228.09Hz to 1,253.6Hz for the 1st to the 6th mode. Total deformations at these natural frequencies are shown from 6.84mm to 12.631mm. Furthermore, Campbell diagram reveals that a critical speed is not found throughout regular rotational speeds. From the test results for the analysis, this paper concludes that the suggested impeller is proved for its mechanical safety and good to utilize at industries.

• 대한의용생체공학회:의공학회지
• /
• 제44권3호
• /
• pp.182-189
• /
• 2023

CT is a medical device that acquires medical images based on Attenuation coefficient of human organs related to X-rays. In addition, using this theory, it can acquire sagittal and coronal planes and 3D images of the human body. Then, CT is essential device for universal diagnostic test. But Exposure of CT scan is so high that it is regulated and managed with special medical equipment. As the special medical equipment, CT must implement quality control. In detail of quality control, Spatial resolution of existing phantom imaging tests, Contrast resolution and clinical image evaluation are qualitative tests. These tests are not objective, so the reliability of the CT undermine trust. Therefore, by applying an artificial intelligence classification model, we wanted to confirm the possibility of quantitative evaluation of the qualitative evaluation part of the phantom test. We used intelligence classification models (VGG19, DenseNet201, EfficientNet B2, inception_resnet_v2, ResNet50V2, and Xception). And the fine-tuning process used for learning was additionally performed. As a result, in all classification models, the accuracy of spatial resolution was 0.9562 or higher, the precision was 0.9535, the recall was 1, the loss value was 0.1774, and the learning time was from a maximum of 14 minutes to a minimum of 8 minutes and 10 seconds. Through the experimental results, it was concluded that the artificial intelligence model can be applied to CT implements quality control in spatial resolution and contrast resolution.