• 제목/요약/키워드: fine grained classification

검색결과 32건 처리시간 0.027초

Classifying Articles in Chinese Wikipedia with Fine-Grained Named Entity Types

  • Zhou, Jie;Li, Bicheng;Tang, Yongwang
    • Journal of Computing Science and Engineering
    • /
    • 제8권3호
    • /
    • pp.137-148
    • /
    • 2014
  • Named entity classification of Wikipedia articles is a fundamental research area that can be used to automatically build large-scale corpora of named entity recognition or to support other entity processing, such as entity linking, as auxiliary tasks. This paper describes a method of classifying named entities in Chinese Wikipedia with fine-grained types. We considered multi-faceted information in Chinese Wikipedia to construct four feature sets, designed different feature selection methods for each feature, and fused different features with a vector space using different strategies. Experimental results show that the explored feature sets and their combination can effectively improve the performance of named entity classification.

영상 처리와 CNN을 이용한 애완동물 영상 세부 분류 비교 (Comparison of Fine Grained Classification of Pet Images Using Image Processing and CNN)

  • 김지혜;고정환;권철희
    • 방송공학회논문지
    • /
    • 제26권2호
    • /
    • pp.175-183
    • /
    • 2021
  • 영상의 세부 분류에 대한 연구는 계속적으로 발전하고 있지만, 다형성의 성질을 갖는 동물에 대한 객체인식 연구는 더디게 진행되고 있다. 본 논문은 개와 고양이에 해당하는 애완동물 이미지만을 이용하여, 세부 분류인 동물의 종을 분류하는 방법 중 영상처리를 이용한 방법과 딥러닝을 이용한 방법을 비교하는 것을 목표로 한다. 본 논문에서 영상처리를 이용한 방법으로 객체 분리를 위해 Grab-cut 알고리즘을 사용하고, 영상 인코딩을 위해 Fisher Vector를 사용한 방법을 제안한다. 다른 방법으로는 기계학습으로 여러 분야에서 좋은 성과를 얻고 있는 딥러닝을 이용하였으며, 그 중에서도 이미지 인식 분야에서 뛰어난 성능을 보인 Convolutional Neural Network(CNN)과 구글에서 제공하는 오픈소스 기반 딥러닝 프레임워크인 Tensorflow를 활용하였다. 제안하는 각각의 방법에 대해 37종의 애완동물 이미지, 총 7,390장에 대해 실험하여 그 효과를 검증 및 비교하였다.

Compressibility of fine-grained sediments based on pore water salinity changes

  • Junbong Jang;Handikajati Kusuma Marjadi
    • Geomechanics and Engineering
    • /
    • 제33권1호
    • /
    • pp.113-120
    • /
    • 2023
  • Coastal and offshore structures such as ports and offshore wind farms will often need to be built on fine-grained sediments. Geotechnical properties associated with sediment compressibility are key parameters for marine construction designs especially on soft grounds, which involve clay-mineral dominated fines that can consolidate and settle significantly in response to engineered and environmental loads. We conduct liquid limit tests and 1D consolidation tests with fine-grained soils (silica silt, mica, kaolin and bentonite) and biogenic soils (diatom). The pore fluids for the liquid limit tests include deionized water and a series of brines with NaCl salt concentrations of 0.001 m, 0.01 m, 0.1 m, 0.6 m and 2.0 m, and the pore fluids for the consolidation tests deionized water, 0.01 m, 0.6 m, 2 m. The salt concentrations help the liquid limits of kaolin and bentonite decrease, but those of diatom slightly increase. The silica silt and mica show minimal changes in liquid limit due to salt concentrations. Accordingly, compression indices of soils follow the trend of the liquid limit as the liquid limit determined the initial void ratio of the consolidation test. Diatoms are more likely to be broken than clastic sediments during to loading, and diatom-rich sediment is therefore generally more compressible than clastic-rich sediment.

직접전단장비를 이용한 토석류의 전단강도 및 유변학적 특성에 대한 입도분포의 영향 연구 (Effects of Grain Size Distribution on the Shear Strength and Rheological Properties of Debris Flow Using Direct Shear Apparatus)

  • 박근우;홍원택;홍영호;정승원;이종섭
    • 한국지반공학회논문집
    • /
    • 제33권12호
    • /
    • pp.7-20
    • /
    • 2017
  • 본 연구에서는 전단특성 및 유변학적 정수를 모두 산정할 수 있는 직접전단실험 장비를 이용하여 조립토와 세립토에 대하여 전단강도 및 유변학적 특성에 대한 입도분포의 영향을 조사하고자 하였다. 최대입경 0.075mm의 세립토와 최대입경이 0.425mm이고, 세립분 함량이 17%인 조립토를 건조상태와 액성한계상태로 조성하여, 산사태 분류기준에 따라 재활성 산사태(reactivated landslide) 혹은 붕괴직후 토석류 속도에 해당하는 전단속도에 대하여 전단강도를 산정하였다. 또한, 유변학적 특성 평가를 위해 액성한계상태로 조성된 조립토와 세립토에 대하여 서로 다른 세 가지의 전단변형률속도로 반복적으로 전단하며 잔류전단강도를 측정하였다. 측정된 잔류전단강도와 전단변형률속도와의 관계를 통해 빙햄모델의 소성 점도와 항복응력을 산정하였다. 건조 및 액성한계상태에서 조성된 시료에 대하여 첨두전단강도에서 산정한 점착력의 경우, 세립토에서 조립토보다 더 크게 산정되었으며, 내부마찰각은 조립토에서 더 크게 산정되었다. 유변학 정수의 경우, 소성 점도와 항복응력이 조립토보다 세립토에서 더 큰 것으로 나타났다. 본 연구는 재활성 산사태 혹은 붕괴직후 토석류의 거동예측에 효과적으로 활용될 것으로 기대된다.

Assessing the impact of nanoclay on the permeability and geotechnical properties of fine-grained soils in landfill liners

  • Mahdi Nikbakht;Fariba Behrooz Sarand;Rouzbeh Dabiri;Masoud Hajialilue Bonab
    • Advances in materials Research
    • /
    • 제13권1호
    • /
    • pp.19-35
    • /
    • 2024
  • Presented Article evaluates the effect of nanoclay on permeability, compressive strength, and plasticity behavior of fine-grained soil related to the Tabriz landfill site. In this regard, comprehensive experimental study was performed on taken soil samples (42 specimens) with aim of design high-performance liners for Tabriz landfill. The samples was mixed by 0% (control) 3%, 6% and 9% nanoclay and prepared in 1, 7, 14 and 28 days before testing stage. Index tests like particle-size, permeability, atterberg limits, and uniaxial compressive strength (UCS) was conducted on samples. The results show that studied soil is classified as CL in USCS classification and atterberg limits measured as LL is 37, PL is 20.67, and PI is 16.33 which increase into 75, 45, and 30. The assessment presented the LL was increased about 20.27% based on increasing in nanoclay from 0% to 9%. These variations for PL and PI were 21.77% and 18.37%, respectively. Also, the and soil's compressive strength is increase from 120 kPa to 188 kPa and permeability is estimated as 4.25×10-6 m/s which reduced into the 6.34×10-9 m/s with respect the naboclay content increases form 0% to 9%.

One-dimensional CNN Model of Network Traffic Classification based on Transfer Learning

  • Lingyun Yang;Yuning Dong;Zaijian Wang;Feifei Gao
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제18권2호
    • /
    • pp.420-437
    • /
    • 2024
  • There are some problems in network traffic classification (NTC), such as complicated statistical features and insufficient training samples, which may cause poor classification effect. A NTC architecture based on one-dimensional Convolutional Neural Network (CNN) and transfer learning is proposed to tackle these problems and improve the fine-grained classification performance. The key points of the proposed architecture include: (1) Model classification--by extracting normalized rate feature set from original data, plus existing statistical features to optimize the CNN NTC model. (2) To apply transfer learning in the classification to improve NTC performance. We collect two typical network flows data from Youku and YouTube, and verify the proposed method through extensive experiments. The results show that compared with existing methods, our method could improve the classification accuracy by around 3-5%for Youku, and by about 7 to 27% for YouTube.

Black box-assisted fine-grained hierarchical access control scheme for epidemiological survey data

  • Xueyan Liu;Ruirui Sun;Linpeng Li;Wenjing Li;Tao Liu
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제17권9호
    • /
    • pp.2550-2572
    • /
    • 2023
  • Epidemiological survey is an important means for the prevention and control of infectious diseases. Due to the particularity of the epidemic survey, 1) epidemiological survey in epidemic prevention and control has a wide range of people involved, a large number of data collected, strong requirements for information disclosure and high timeliness of data processing; 2) the epidemiological survey data need to be disclosed at different institutions and the use of data has different permission requirements. As a result, it easily causes personal privacy disclosure. Therefore, traditional access control technologies are unsuitable for the privacy protection of epidemiological survey data. In view of these situations, we propose a black box-assisted fine-grained hierarchical access control scheme for epidemiological survey data. Firstly, a black box-assisted multi-attribute authority management mechanism without a trusted center is established to avoid authority deception. Meanwhile, the establishment of a master key-free system not only reduces the storage load but also prevents the risk of master key disclosure. Secondly, a sensitivity classification method is proposed according to the confidentiality degree of the institution to which the data belong and the importance of the data properties to set fine-grained access permission. Thirdly, a hierarchical authorization algorithm combined with data sensitivity and hierarchical attribute-based encryption (ABE) technology is proposed to achieve hierarchical access control of epidemiological survey data. Efficiency analysis and experiments show that the scheme meets the security requirements of privacy protection and key management in epidemiological survey.

인천 해안지역 지반의 국지적 침식특성 평가 (A Case Study on Local Erosion Characteristics Evaluation of the Inchon Coast)

  • 곽기석;이주형;박재현;정문경
    • 한국지반공학회:학술대회논문집
    • /
    • 한국지반공학회 2006년도 춘계 학술발표회 논문집
    • /
    • pp.455-465
    • /
    • 2006
  • In this study, the effect of scour was evaluated by regional and geotechnical characteristics and back data were accumulate for the design against scour through the local erosion characteristics evaluation of the Inchon coast. The erosion characteristics for the undisturbed soil samples collected near the main locations at the Incheon 2nd bridge, the Hwangyeong bridge, and a coast road in Songdo, are determined quantitatively through the scour rate tests. On the basic soil properties test, the bed around the Inchon coast chiefly consists of fined grained soils, and the soil samples were classified as silty clay(ML) or clay(CL) under the Unified Soil Classification System. On the scour rate test, the critical shear stress increases when the undrained shear strength increases as of the general trend of fine grained soils, and the average scour rate for the maximum velocity by 100 year flood is 173mm/hr at the Incheon 2nd bridge, 67mm/hr at the Hwangyeong bridge and 10mm/hr at a coast road in Songdo, respectively. Comparing to the scour rate of coarse grained soil, that of the bed around the Inchon coast is turned out to be very low. Therefore, the relative ability of the bed around the Inchon coast to resist erosion is assumed to be very high.

  • PDF

Attention Capsule Network for Aspect-Level Sentiment Classification

  • Deng, Yu;Lei, Hang;Li, Xiaoyu;Lin, Yiou;Cheng, Wangchi;Yang, Shan
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제15권4호
    • /
    • pp.1275-1292
    • /
    • 2021
  • As a fine-grained classification problem, aspect-level sentiment classification predicts the sentiment polarity for different aspects in context. To address this issue, researchers have widely used attention mechanisms to abstract the relationship between context and aspects. Still, it is difficult to effectively obtain a more profound semantic representation, and the strong correlation between local context features and the aspect-based sentiment is rarely considered. In this paper, a hybrid attention capsule network for aspect-level sentiment classification (ABASCap) was proposed. In this model, the multi-head self-attention was improved, and a context mask mechanism based on adjustable context window was proposed, so as to effectively obtain the internal association between aspects and context. Moreover, the dynamic routing algorithm and activation function in capsule network were optimized to meet the task requirements. Finally, sufficient experiments were conducted on three benchmark datasets in different domains. Compared with other baseline models, ABASCap achieved better classification results, and outperformed the state-of-the-art methods in this task after incorporating pre-training BERT.

대입열 EH36-TM강의 Tandem EGW 용접부 미세조직 및 기계적 성질 (Mechanical Properties and Microstructures of High Heat Input Welded Tandem EGW Joint in EH36-TM Steel)

  • 정홍철;박영환;안영호;이종봉
    • Journal of Welding and Joining
    • /
    • 제25권1호
    • /
    • pp.57-62
    • /
    • 2007
  • In the coarse grained HAZ of conventional TiN steel, most TiN particles are dissolved and austenite grain growth easily occurrs during high heat input welding. To avoid this difficulty, thermal stability of TiN particles is improved by increasing nitrogen content in EH36-TM steel. Increased thermal stability of TiN particle is helpful for preventing austenite grain growth by the pinning effect. In this study, the mechanical properties and microstructures of high heat input welded Tandem EGW joint in EH36-TM steel with high nitrogen content were investigated. The austenite grain size in simulated HAZ of the steel at $1400^{\circ}C$ was much smaller than that of conventional TiN steel. Even for high heat input welding, the microstructure of coarse grained HAZ consisted of fine ferrite and pearlite and the mechanical properties of the joint were sufficient to meet all the requirements specified in classification rule.