13197 research outputs found
Sort by
Comparative evaluation of boundary-relaxed annotation for Entity Linking performance
No full textEntity Linking performance has a strong reliance on having a large quantity of high-quality annotated training data available. Yet, manual annotation of named entities, especially their boundaries, is ambiguous, error-prone, and raises many inconsistencies between annotators. While imprecise boundary annotation can degrade a model’s performance, there are applications where accurate extraction of entities’ surface form is not necessary. For those cases, a lenient annotation guideline could relieve the annotators’ workload and speed up the process. This paper presents a case study designed to verify the feasibility of such annotation process and evaluate the impact of boundary-relaxed annotation in an Entity Linking pipeline. We first generate a set of noisy versions of the widely used AIDA CoNLL-YAGO dataset by expanding the boundaries subsets of annotated entity mentions and then train three Entity Linking models on this data and evaluate the relative impact of imprecise annotation on entity recognition and disambiguation performances. We demonstrate that the magnitude of effects caused by noise in the Named Entity Recognition phase is dependent on both model complexity and noise ratio, while Entity Disambiguation components are susceptible to entity boundary imprecision due to strong vocabulary dependency.conference pape
テンシャ カイシテン ヤ ポリA フカブイ ノ タヨウセイ ヤ ジョウケンカン ヘンカ ノ セイギョ キコウ ニ カカワル シスハイレツ ノ カイセキ
Get PDF奈良先端科学技術大学院大学修士(工学)master thesi
Application of in vivo CMOS imaging device with microdialysis and optogenetics for studying serotonergic neurons in pain modulation
Get PDF奈良先端科学技術大学院大学博士（理学）doctoral thesi
Lightweight Intrusion Detection Using Multiple Entropies of Traffic Behavior in IoT Networks
Get PDFSince Mirai malware first appeared in 2016, different variants have been created. The variants infect Internet of Things (IoT) devices such as home routers and webcams. The scale of DDoS attacks using Mirai-infected IoT devices has exceeded 600 Gbps. There has been a lot of researches on intrusion detection methods using machine learning for IoT networks. However, the existing method needs a lot of computational resources. Therefore, it is difficult to run such intrusion detection systems on resource-limited IoT gateways. In this research, we focus on the communication behavior of IoT devices, such as periodic communication with a specific server during benign operations. We propose a new intrusion detection method that represents the communication behavior of each host using multiple entropy features such as destination port number, source port number, and transmission time interval. The proposed method can achieve performance comparable to the existing intrusion detection method even if using a lightweight machine learning algorithm with fewer features. The evaluation of the results shows that the proposed method can reduce the detection processing time by 28.7 ms and memory usage by up to 331 MiB compared to the existing method, and the proposed method can achieve a detection accuracy of 99.8%, which is almost the same as the existing method.conference pape
General Software Platform and Content Description Format for Assembly and Maintenance Task Based on Augmented Reality
No full textAugmented reality (AR) support systems have proven to be effective in supporting various tasks, but are not yet widely used in real-world applications due to their high development costs. Although several AR authoring tools have been proposed for content development to solve this problem, most AR systems developed with these tools are not compatible with industrial environments and task types, and have inflexible visualization styles. This study provides a systematic solution of combining the AR task support software platform and the general contents description format by expanding the degree of freedom of the visualization methods and improving the design process. The proposed solution adapts to different activities and environments, and optimizes content implementation and user experience. The user study was conducted to evaluate the task performance including processing time and errors and the user experience including the perceived cognitive load and usability of our solution by comparing with a conventional system. The results show that the proposed software platform could improve the compatibility with industrial environments and tasks, and reduce the workload and cognitive effort, although the task performance was the same as the conventional AR system.journal articl
Development of serial X-ray fluorescence holography for radiation-sensitive protein crystals
No full textX-ray fluorescence holography (XFH) is a powerful atomic resolution technique capable of directly imaging the local atomic structure around atoms of a target element within a material. Although it is theoretically possible to use XFH to study the local structures of metal clusters in large protein crystals, the experiment has proven difficult to perform, especially on radiation-sensitive proteins. Here, the development of serial X-ray fluorescence holography to allow the direct recording of hologram patterns before the onset of radiation damage is reported. By combining a 2D hybrid detector and the serial data collection used in serial protein crystallography, the X-ray fluorescence hologram can be directly recorded in a fraction of the measurement time needed for conventional XFH measurements. This approach was demonstrated by obtaining the Mn Kα hologram pattern from the protein crystal Photosystem II without any X-ray-induced reduction of the Mn clusters. Furthermore, a method to interpret the fluorescence patterns as real-space projections of the atoms surrounding the Mn emitters has been developed, where the surrounding atoms produce large dark dips along the emitter$2013scatterer bond directions. This new technique paves the way for future experiments on protein crystals that aim to clarify the local atomic structures of their functional metal clusters, and for other related XFH experiments such as valence-selective XFH or time-resolved XFH.research repor
Imaging techniques to analyze molecular mechanics for cell migration and axon guidance
No full textvideo/mp4講演日: 2023年7月20日オンライン開催vide
AI ノ シンカ オ ササエル ハンドウタイ プロセス デバイス ギジュツ
No full textvideo/mp4AIの急激な発達によって、我々の生活は便利になりつつあると同時に大きな変化をもたらしています。しかし、この発展を支えているのは、半導体技術の進化によるものです。この講座では、微細化、3次元構造化など半導体の最先端技術を紹介します。講演日2023年10月21日講演場所: ミレニアムホール講演者所属: 物質創成科学領域vide
FOREWORD : Special Section on the Architectures, Protocols, and Applications for the Future Internet
Get PDFThe Internet has been established as a social infrastructure with rapid growth from just a communication tool for academic researchers, the emergence of the World Wide Web, and the spread of multimedia information processing. PCs and smartphones, smart meters, in-vehicle devices, sensors, and many other things are connected to the Internet; a wide variety of traffic goes through the Internet. In addition, artificial intelligence technologies, such as deep learning, are being applied in various fields and becoming commoditized, changing the architecture of the Internet. Because of the importance of proposing and demonstrating new architectures and application technologies necessary for the Internet to fulfill its role in the future, we have planned this special section.journal articl