scholarly journals Image Enhanced Event Detection in News Articles

2020 ◽  
Vol 34 (05) ◽  
pp. 9040-9047
Author(s):  
Meihan Tong ◽  
Shuai Wang ◽  
Yixin Cao ◽  
Bin Xu ◽  
Juanzi Li ◽  
...  
Keyword(s):  
Event Detection ◽  
Event Extraction ◽  
Superior Performance ◽  
Context Information ◽  
The Core ◽  
Image Dataset ◽  
Image Modality ◽  
State Of Art

Event detection is a crucial and challenging sub-task of event extraction, which suffers from a severe ambiguity issue of trigger words. Existing works mainly focus on using textual context information, while there naturally exist many images accompanied by news articles that are yet to be explored. We believe that images not only reflect the core events of the text, but are also helpful for the disambiguation of trigger words. In this paper, we first contribute an image dataset supplement to ED benchmarks (i.e., ACE2005) for training and evaluation. We then propose a novel Dual Recurrent Multimodal Model, DRMM, to conduct deep interactions between images and sentences for modality features aggregation. DRMM utilizes pre-trained BERT and ResNet to encode sentences and images, and employs an alternating dual attention to select informative features for mutual enhancements. Our superior performance compared to six state-of-art baselines as well as further ablation studies demonstrate the significance of image modality and effectiveness of the proposed architecture. The code and image dataset are avaliable at https://github.com/shuaiwa16/image-enhanced-event-extraction.

scholarly journals Zero-Shot Ingredient Recognition by Multi-Relational Graph Convolutional Network

2020 ◽  
Vol 34 (07) ◽  
pp. 10542-10550 ◽  
Author(s):  
Jingjing Chen ◽  
Liangming Pan ◽  
Zhipeng Wei ◽  
Xiang Wang ◽  
Chong-Wah Ngo ◽  
...  
Keyword(s):  
Dietary Assessment ◽  
Training Data ◽  
Superior Performance ◽  
Convolutional Network ◽  
Food Ingredients ◽  
The Core ◽  
Training Samples ◽  
Robust Recognition ◽  
The Common ◽  
Shed Light

Recognizing ingredients for a given dish image is at the core of automatic dietary assessment, attracting increasing attention from both industry and academia. Nevertheless, the task is challenging due to the difficulty of collecting and labeling sufficient training data. On one hand, there are hundred thousands of food ingredients in the world, ranging from the common to rare. Collecting training samples for all of the ingredient categories is difficult. On the other hand, as the ingredient appearances exhibit huge visual variance during the food preparation, it requires to collect the training samples under different cooking and cutting methods for robust recognition. Since obtaining sufficient fully annotated training data is not easy, a more practical way of scaling up the recognition is to develop models that are capable of recognizing unseen ingredients. Therefore, in this paper, we target the problem of ingredient recognition with zero training samples. More specifically, we introduce multi-relational GCN (graph convolutional network) that integrates ingredient hierarchy, attribute as well as co-occurrence for zero-shot ingredient recognition. Extensive experiments on both Chinese and Japanese food datasets are performed to demonstrate the superior performance of multi-relational GCN and shed light on zero-shot ingredients recognition.

scholarly journals Reactive Navigation Framework for Mobile Robots by Heuristically Evaluated Pre-sampled Trajectories

2021 ◽  
Vol 3 (1) ◽  
pp. 47-68
Author(s):  
Neset Unver Akmandor ◽  
Taskın Padir
Keyword(s):  
Mobile Robots ◽  
Superior Performance ◽  
Grid Structure ◽  
Systematic Method ◽  
Reactive Navigation ◽  
Benchmark Data ◽  
Benchmark Tests ◽  
Unknown Environments ◽  
Previous Iteration ◽  
State Of Art

This paper describes and analyzes a reactive navigation framework for mobile robots in unknown environments. The approach does not rely on a global map and only considers the local occupancy in its robot-centered 3D grid structure. The proposed algorithm enables fast navigation by heuristic evaluations of pre-sampled trajectories on-the-fly. At each cycle, these paths are evaluated by a weighted cost function, based on heuristic features such as closeness to the goal, previously selected trajectories, and nearby obstacles. This paper introduces a systematic method to calculate a feasible pose on the selected trajectory, before sending it to the controller for the motion execution. Defining the structures in the framework and providing the implementation details, the paper also explains how to adjust its offline and online parameters. To demonstrate the versatility and adaptability of the algorithm in unknown environments, physics-based simulations on various maps are presented. Benchmark tests show the superior performance of the proposed algorithm over its previous iteration and another state-of-art method. The open-source implementation of the algorithm and the benchmark data can be found at https://github.com/RIVeR-Lab/tentabot.

scholarly journals A Prior Information Enhanced Extraction Framework for Document-level Financial Event Extraction

2021 ◽  
pp. 1-12
Author(s):  
Haitao Wang ◽  
Tong Zhu ◽  
Mingtao Wang ◽  
Guoliang Zhang ◽  
Wenliang Chen
Keyword(s):  
Event Detection ◽  
Prior Information ◽  
Event Extraction ◽  
Language Models ◽  
Event Type ◽  
Event Argument ◽  
Implicit Information ◽  
Share Task ◽  
Financial Domain ◽  
Document Level

Abstract Document-level financial event extraction (DFEE) is the task of detecting event and extracting the corresponding event arguments in financial documents, which plays an important role in information extraction in the financial domain. This task is challenging as the financial documents are generally long text and event arguments of one event may be scattered in different sentences. To address this issue, we propose a novel Prior Information Enhanced Extraction framework (PIEE) for DFEE, leveraging prior information from both event types and pre-trained language models. Specifically, PIEE consists of three components: event detection, event argument extraction, and event table filling. In event detection, we identify the event type. Then, the event type is explicitly used for event argument extraction. Meanwhile, the implicit information within language models also provides considerable cues for event arguments localization. Finally, all the event arguments are filled in an event table by a set of predefined heuristic rules. To demonstrate the effectiveness of our proposed framework, we participate the share task of CCKS2020 Task5-2: Document-level Event Arguments Extraction. On both Leaderboard A and Leaderboard B, PIEE takes the first place and significantly outperforms the other systems.

scholarly journals New leads for fragment-based design of rhenium/technetium radiopharmaceutical agents

IUCrJ ◽  
2017 ◽  
Vol 4 (3) ◽  
pp. 283-290 ◽  
Author(s):  
Alice Brink ◽  
John R. Helliwell
Keyword(s):  
Binding Sites ◽  
Target Identification ◽  
Peak Height ◽  
Wavelength Tuning ◽  
Amino Acid Residues ◽  
X Ray ◽  
The Core ◽  
Height Increase ◽  
Radiopharmaceutical Agents ◽  
State Of Art

Multiple possibilities for the coordination offac-[Re(CO)3(H2O)3]+to a protein have been determined and include binding to Asp, Glu, Arg and His amino-acid residues as well as to the C-terminal carboxylate in the vicinity of Leu and Pro. The large number of rhenium metal complex binding sites that have been identified on specific residues thereby allow increased target identification for the design of future radiopharmaceuticals. The core experimental concept involved the use of state-of-art tuneable synchrotron radiation at the Diamond Light Source to optimize the rhenium anomalous dispersion signal to a large value (f′′ of 12.1 electrons) at itsLIabsorption edge with a selected X-ray wavelength of 0.9763 Å. At the Cu Kα X-ray wavelength (1.5418 Å) thef′′ for rhenium is 5.9 electrons. The expected peak-height increase owing to the optimization of the Ref′′ was therefore 2.1. This X-ray wavelength tuning methodology thereby showed the lower occupancy rhenium binding sites as well as the occupancies of the higher occupancy rhenium binding sites.

scholarly journals A pesquisa de representação social na área de ensino-aprendizagem: elementos do estado da arte

2019 ◽  
Vol 79 (193) ◽  
Author(s):  
Mary Rangel
Keyword(s):  
Social Psychology ◽  
Social Representations ◽  
Social Representation ◽  
The Core ◽  
The Social ◽  
Social Representation Theory ◽  
Final Segment ◽  
Research In Education ◽  
Teaching Learning ◽  
State Of Art

A pesquisa de pósdoutorado, Teoria de representação social: o quadro teórico da Psicologia Social e aplicações atuais à pesquisa na educação (Rangel, 1997), teve como um dos seus objetivos a análise de dissertações e teses contemporâneas (anos 90) que aplicaram a teoria, trazendo contribuições particularmente à área de ensino-aprendizagem. Este estudo, então, possibilitou - nos limites das pesquisas alcançadas - identificar elementos do estado da arte, sem perder de vista a crítica à Teoria de Representação Social (TRS), com particular consideração a Spink (1996). Ainda, tratando-se da TRS, procurou-se observar, pela sua importância na estrutura das representações, componentes do núcleo central da sua aplicação nas pesquisas. No segmento conclusivo, apresentam-se sugestões ao avanço e refinamento da construção teórica. Palavras-Chave: representação social; pesquisa; ensinoaprendizagem; estado da arte. Abstract The research of pos-doctorate, Theory of the Social Representation: the theoretical table of the Social Psychology and its applications of the present time to the research in Education (Rangel, 1997), had as one of the aims the analysis of dissertations and the contemporary theses (90) that applied the theory, bringing contributions, especially in teaching-apprenticeship. In this study could be identified (in the limits of the research) the elements of the state of art – with critical to the social representations theory (SRT), considering Spink (1996). About the Social Representation Theory, some components of the core of its application in research were observed by the importance in the structure of the representations. In the final segment, suggestions are presented to the advancement and refinement of the theorical construction. Keywords: social representations; research; teaching-learning; state of art.

scholarly journals Current State of Art Design Education in Colleges and a New Hybrid Learning Mode

2021 ◽  
Vol 16 (17) ◽  
pp. 120
Author(s):  
Jing Liang
Keyword(s):  
Design Education ◽  
Teaching Method ◽  
Hybrid Learning ◽  
Teaching Quality ◽  
Project Based Learning ◽  
The Core ◽  
Current State ◽  
Art Design ◽  
Core Literacy ◽  
State Of Art

Art design education contributes immensely to social and economic development. To train art designers that suit social development and industrial needs, colleges must substantially reform their art design education. From the perspective of cultivating the core literacy, this paper carries out a survey on the current state of art design education in colleges. The survey reveals the following problems of art design education in Chinese colleges: the students have a poor cultural literacy, a low interest in professional learning, and a weak understanding of the core literacy; the faculty has an unreasonable structure, an overall weak competence, a monotonous teaching method, and the detachment of theories from practices. Based on the survey results, a hybrid learning mode was derived from project-based learning, and the application effect of the proposed mode in art design education was analyzed from the perspective of the cultivation of core literacy. The results show that, the proposed learning mode can effectively enhance teaching quality, and improve the core literacy and professional ability of students. The research provides a valuable theoretical and practical reference for the reform of art design education.

scholarly journals StyHighNet: Semi-Supervised Learning Height Estimation from a Single Aerial Image via Unified Style Transferring

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2272
Author(s):  
Qian Gao ◽  
Xukun Shen
Keyword(s):  
Supervised Learning ◽  
Domain Adaptation ◽  
Superior Performance ◽  
Aerial Image ◽  
Distribution Maps ◽  
The Core ◽  
Public Data ◽  
Learning Modes ◽  
New City ◽  
The One

Recovering height information from a single aerial image is a key problem in the fields of computer vision and remote sensing. At present, supervised learning methods have achieved impressive results, but, due to domain bias, the trained model cannot be directly applied to a new scene. In this paper, we propose a novel semi-supervised framework, StyHighNet, for accurately estimating the height of a single aerial image in a new city that requires only a small number of labeled data. The core is to transfer multi-source images to a unified style, making the unlabeled data provide the appearance distribution as additional supervision signals. The framework mainly contains three sub-networks: (1) the style transferring sub-network maps multi-source images into unified style distribution maps (USDMs); (2) the height regression sub-network, with the function of predicting the height maps from USDMs; and (3) the style discrimination sub-network, used to distinguish the sources of USDMs. Among them, the style transferring sub-network shoulders dual responsibilities: On the one hand, it needs to compute USDMs with obvious characteristics, so that the height regression sub-network can accurately estimate the height maps. On the other hand, it is necessary that the USDMs have consistent distribution to confuse the style discrimination sub-network, so as to achieve the goal of domain adaptation. Unlike previous methods, our style distribution function is learned unsupervised, thus it is of greater flexibility and better accuracy. Furthermore, when the style discrimination sub-network is shielded, this framework can also be used for supervised learning. We performed qualitatively and quantitative evaluations on two sets of public data, Vaihingen and Potsdam. Experiments show that the framework achieved superior performance in both supervised and semi-supervised learning modes.

scholarly journals EVALUATION OF VISION-BASED LOCALIZATION AND MAPPING TECHNIQUES IN A SUBSEA METROLOGY SCENARIO

2019 ◽  
Vol XLII-2/W10 ◽  
pp. 127-134
Author(s):  
F. Menna ◽  
A. Torresani ◽  
E. Nocerino ◽  
M. M. Nawaf ◽  
J. Seinturier ◽  
...  
Keyword(s):  
Measured Data ◽  
Remotely Operated Vehicle ◽  
The Core ◽  
Localization And Mapping ◽  
Offshore Industry ◽  
Mapping Techniques ◽  
Vision Based Localization ◽  
And Control ◽  
State Of Art ◽  
Underwater Structures

<p><strong>Abstract.</strong> Metrology is fundamental in all the applications that require to qualify, verify and validate measured data according to standards or, in other words, to assess their compliance with predefined tolerances. At sea, metrology is commonly associated with the process of measuring underwater structures, mainly pipeline elements widely used in offshore industry. Subsea operations are very expensive; optimizing time and money resources are the core factors driving innovation in the subsea metrology industry. In this study, the authors investigate the use of state-of-art vision-based algorithms, i.e. ORB-SLAM2 and Visual Odometry, as a navigation tool to assist and control a Remotely Operated Vehicle (ROV) while performing subsea metrology operations. In particular, the manuscript will focus on methods for assessing the accuracy of both trajectory and tie points provided by the tested approaches and evaluating whether the preliminary real time reconstruction meets the tolerances defined in typical subsea metrology scenarios.</p>

scholarly journals An integrative network-based approach for drug target indication expansion

PLoS ONE ◽  
2021 ◽  
Vol 16 (7) ◽  
pp. e0253614
Author(s):  
Yingnan Han ◽  
Clarence Wang ◽  
Katherine Klinger ◽  
Deepak K. Rajpal ◽  
Cheng Zhu
Keyword(s):  
Drug Discovery ◽  
Drug Targets ◽  
Interaction Network ◽  
Superior Performance ◽  
Systematic Evaluation ◽  
Network Algorithms ◽  
Drug Discovery And Development ◽  
Traditional Drug ◽  
Literature Evidence ◽  
State Of Art

Background The identification of a target-indication pair is regarded as the first step in a traditional drug discovery and development process. Significant investment and attrition occur during discovery and development before a molecule is shown to be safe and efficacious for the selected indication and becomes an approved drug. Many drug targets are functionally pleiotropic and might be good targets for multiple indications. Methodologies that leverage years of scientific contributions on drug targets to allow systematic evaluation of other indication opportunities are critical for both patients and drug discovery and development scientists. Methods We introduced a network-based approach to systematically screen and prioritize disease indications for drug targets. The approach fundamentally integrates disease genomics data and protein interaction network. Further, the methodology allows for indication identification by leveraging state-of-art network algorithms to generate and compare the target and disease subnetworks. Results We first evaluated the performance of our method on recovering FDA approved indications for 15 randomly selected drug targets. The results showed superior performance when compared with other state-of-art approaches. Using this approach, we predicted novel indications supported by literature evidence for several highly pursued drug targets such as IL12/IL23 combination. Conclusions Our results demonstrated a potential global approach for indication expansion strategies. The proposed methodology enables rapid and systematic evaluation of both individual and combined drug targets for novel indications. Additionally, this approach provides novel insights on expanding the role of genes and pathways for developing therapeutic intervention strategies.

scholarly journals Improvement of deep cross-modal retrieval by generating real-valued representation

2021 ◽  
Vol 7 ◽  
pp. e491
Author(s):  
Nikita Bhatt ◽  
Amit Ganatra
Keyword(s):  
State Of The Art ◽  
Representation Learning ◽  
Training Model ◽  
Performance Measure ◽  
Text And Image ◽  
Average Precision ◽  
Common Solution ◽  
The Core ◽  
The Mean ◽  
Image Modality

The cross-modal retrieval (CMR) has attracted much attention in the research community due to flexible and comprehensive retrieval. The core challenge in CMR is the heterogeneity gap, which is generated due to different statistical properties of multi-modal data. The most common solution to bridge the heterogeneity gap is representation learning, which generates a common sub-space. In this work, we propose a framework called “Improvement of Deep Cross-Modal Retrieval (IDCMR)”, which generates real-valued representation. The IDCMR preserves both intra-modal and inter-modal similarity. The intra-modal similarity is preserved by selecting an appropriate training model for text and image modality. The inter-modal similarity is preserved by reducing modality-invariance loss. The mean average precision (mAP) is used as a performance measure in the CMR system. Extensive experiments are performed, and results show that IDCMR outperforms over state-of-the-art methods by a margin 4% and 2% relatively with mAP in the text to image and image to text retrieval tasks on MSCOCO and Xmedia dataset respectively.

Sign in / Sign up

Export Citation Format

Share Document