Unsupervised latent event representation learning and storyline extraction from news articles based on neural networks

2021 ◽  
Vol 25 (3) ◽  
pp. 589-603
Author(s):  
Jiasheng Si ◽  
Linsen Guo ◽  
Deyu Zhou
Keyword(s):  
Neural Network ◽  
Graphical Models ◽  
Representation Learning ◽  
Event Extraction ◽  
Event Representation ◽  
Practical Applications ◽  
Unsupervised Neural Network ◽  
Long Time ◽  
Event Representations ◽  
Over Time

Storyline extraction aims to generate concise summaries of related events unfolding over time from a collection of temporally-ordered news articles. Some existing approaches to storyline extraction are typically built on probabilistic graphical models that jointly model the extraction of events and the storylines from news published in different periods. However, their parameter inference procedures are often complex and require a long time to converge, which hinders their use in practical applications. More recently, a neural network-based approach has been proposed to tackle such limitations. However, event representations of documents, which are important for the quality of the generated storylines, are not learned. In this paper, we propose a novel unsupervised neural network-based approach to extract latent events and link patterns of storylines jointly from documents over time. Specifically, event representations are learned by a stacked autoencoder and clustered for event extraction, then a fusion component is incorporated to link the related events across consecutive periods for storyline extraction. The proposed model has been evaluated on three news corpora and the experimental results show that it outperforms state-of-the-art approaches with significant improvements.

scholarly journals A long time ago, in a galaxy far, far away: How temporal are episodic contents?

2021 ◽  
Author(s):  
Johannes Mahr ◽  
Joshua D. Greene ◽  
Daniel L. Schacter
Keyword(s):  
Time Of Day ◽  
Temporal Orientation ◽  
Mental Event ◽  
Event Representation ◽  
Retrieval Processes ◽  
Human Ability ◽  
Long Time ◽  
Human Adults ◽  
Successful Recall ◽  
Event Representations

A prominent feature of mental event (i.e. ‘episodic’) simulations is their temporality: human adults can generate episodic representations directed towards the past or the future. The ability to entertain event representations with different temporal orientations allows these representations to play various cognitive roles. Here, we investigated how the temporal orientation of imagined events relates to the contents (i.e. ‘what is happening’) of these events. Is the temporal orientation of an episode part of its contents? Or are the processes for assigning temporality to an event representation distinct from those generating its contents? In three experiments (N = 360), we asked participants to generate and later recall a series of imagined events differing in (1) location (indoors vs. outdoors), (2) time of day (daytime vs. nighttime), (3) temporal orientation (past vs. future), and (4) weekday (Monday vs. Friday). We then tested to what extent successful recall of episodic content (i.e. (1) and (2)) would predict recall of temporality and/or weekday information. Results showed that while recall of temporal orientation was predicted by content recall, weekday recall was not. However, temporal orientation was only weakly integrated with episodic contents. This finding suggests that episodic simulations are unlikely to be intrinsically temporal in nature. Instead, similar to other forms of temporal information, temporal orientation might be determined from such contents by reconstructive post-retrieval processes. These results have implications for how the human ability to ‘mentally travel’ in time is cognitively implemented.

scholarly journals Automated Ventricular System Segmentation in Paediatric Patients Treated for Hydrocephalus Using Deep Learning Methods

2019 ◽  
Vol 2019 ◽  
pp. 1-9 ◽  
Author(s):  
Michał Klimont ◽  
Mateusz Flieger ◽  
Jacek Rzeszutek ◽  
Joanna Stachera ◽  
Aleksandra Zakrzewska ◽  
...  
Keyword(s):  
Neural Network ◽  
Data Augmentation ◽  
Ct Images ◽  
Policy Transfer ◽  
Training Data ◽  
Intraobserver Variability ◽  
Practical Applications ◽  
Brain Scans ◽  
Rate Policy ◽  
Ct Brain

Hydrocephalus is a common neurological condition that can have traumatic ramifications and can be lethal without treatment. Nowadays, during therapy radiologists have to spend a vast amount of time assessing the volume of cerebrospinal fluid (CSF) by manual segmentation on Computed Tomography (CT) images. Further, some of the segmentations are prone to radiologist bias and high intraobserver variability. To improve this, researchers are exploring methods to automate the process, which would enable faster and more unbiased results. In this study, we propose the application of U-Net convolutional neural network in order to automatically segment CT brain scans for location of CSF. U-Net is a neural network that has proven to be successful for various interdisciplinary segmentation tasks. We optimised training using state of the art methods, including “1cycle” learning rate policy, transfer learning, generalized dice loss function, mixed float precision, self-attention, and data augmentation. Even though the study was performed using a limited amount of data (80 CT images), our experiment has shown near human-level performance. We managed to achieve a 0.917 mean dice score with 0.0352 standard deviation on cross validation across the training data and a 0.9506 mean dice score on a separate test set. To our knowledge, these results are better than any known method for CSF segmentation in hydrocephalic patients, and thus, it is promising for potential practical applications.

scholarly journals An Exploratory Study of COVID-19 Information on Twitter in the Greater Region

2021 ◽  
Vol 5 (1) ◽  
pp. 5
Author(s):  
Ninghan Chen ◽  
Zhiqiang Zhong ◽  
Jun Pang
Keyword(s):  
Machine Learning ◽  
Social Networks ◽  
Online Social Networks ◽  
Exploratory Study ◽  
Representation Learning ◽  
Data Driven ◽  
Learning Methods ◽  
Twitter Users ◽  
Over Time

The outbreak of the COVID-19 led to a burst of information in major online social networks (OSNs). Facing this constantly changing situation, OSNs have become an essential platform for people expressing opinions and seeking up-to-the-minute information. Thus, discussions on OSNs may become a reflection of reality. This paper aims to figure out how Twitter users in the Greater Region (GR) and related countries react differently over time through conducting a data-driven exploratory study of COVID-19 information using machine learning and representation learning methods. We find that tweet volume and COVID-19 cases in GR and related countries are correlated, but this correlation only exists in a particular period of the pandemic. Moreover, we plot the changing of topics in each country and region from 22 January 2020 to 5 June 2020, figuring out the main differences between GR and related countries.

scholarly journals Adaptive Robust Controller Design-Based RBF Neural Network for Aerial Robot Arm Model

Electronics ◽  
2021 ◽  
Vol 10 (7) ◽  
pp. 831
Author(s):  
Izzat Al-Darraji ◽  
Dimitrios Piromalis ◽  
Ayad A. Kakei ◽  
Fazal Qudus Khan ◽  
Milos Stojemnovic ◽  
...  
Keyword(s):  
Neural Network ◽  
Controller Design ◽  
Sliding Mode ◽  
Rbf Neural Network ◽  
Robot Arm ◽  
Robust Controller ◽  
Practical Applications ◽  
Control Loops ◽  
Modeling Errors ◽  
Aerial Robot

Aerial Robot Arms (ARAs) enable aerial drones to interact and influence objects in various environments. Traditional ARA controllers need the availability of a high-precision model to avoid high control chattering. Furthermore, in practical applications of aerial object manipulation, the payloads that ARAs can handle vary, depending on the nature of the task. The high uncertainties due to modeling errors and an unknown payload are inversely proportional to the stability of ARAs. To address the issue of stability, a new adaptive robust controller, based on the Radial Basis Function (RBF) neural network, is proposed. A three-tier approach is also followed. Firstly, a detailed new model for the ARA is derived using the Lagrange–d'Alembert principle. Secondly, an adaptive robust controller, based on a sliding mode, is designed to manipulate the problem of uncertainties, including modeling errors. Last, a higher stability controller, based on the RBF neural network, is implemented with the adaptive robust controller to stabilize the ARAs, avoiding modeling errors and unknown payload issues. The novelty of the proposed design is that it takes into account high nonlinearities, coupling control loops, high modeling errors, and disturbances due to payloads and environmental conditions. The model was evaluated by the simulation of a case study that includes the two proposed controllers and ARA trajectory tracking. The simulation results show the validation and notability of the presented control algorithm.

scholarly journals Dynamics of life expectancy and life span equality

2020 ◽  
Vol 117 (10) ◽  
pp. 5250-5259 ◽  
Author(s):  
José Manuel Aburto ◽  
Francisco Villavicencio ◽  
Ugofilippo Basellini ◽  
Søren Kjærgaard ◽  
James W. Vaupel
Keyword(s):  
Life Expectancy ◽  
Life Span ◽  
Health Policies ◽  
Public Health Policies ◽  
Dynamic Relationship ◽  
Mortality Increase ◽  
Long Time ◽  
Underlying Causes ◽  
The Relationship ◽  
Over Time

As people live longer, ages at death are becoming more similar. This dual advance over the last two centuries, a central aim of public health policies, is a major achievement of modern civilization. Some recent exceptions to the joint rise of life expectancy and life span equality, however, make it difficult to determine the underlying causes of this relationship. Here, we develop a unifying framework to study life expectancy and life span equality over time, relying on concepts about the pace and shape of aging. We study the dynamic relationship between life expectancy and life span equality with reliable data from the Human Mortality Database for 49 countries and regions with emphasis on the long time series from Sweden. Our results demonstrate that both changes in life expectancy and life span equality are weighted totals of rates of progress in reducing mortality. This finding holds for three different measures of the variability of life spans. The weights evolve over time and indicate the ages at which reductions in mortality increase life expectancy and life span equality: the more progress at the youngest ages, the tighter the relationship. The link between life expectancy and life span equality is especially strong when life expectancy is less than 70 y. In recent decades, life expectancy and life span equality have occasionally moved in opposite directions due to larger improvements in mortality at older ages or a slowdown in declines in midlife mortality. Saving lives at ages below life expectancy is the key to increasing both life expectancy and life span equality.

QUALITY AND SAFETY - TWO SIDES OF THE SAME COIN IN HOSPITALITY AND TOURISM

2009 ◽  
Vol 15 (2) ◽  
pp. 193-204
Author(s):  
Anders Steene
Keyword(s):  
Quality And Safety ◽  
Quality Aspects ◽  
Long Time ◽  
Hospitality And Tourism ◽  
Two Sides ◽  
Over Time

This paper will discuss some methodical aspects in doing research in the field of hospitality and tourism. Quality aspects have been dominant subjects for a long time in the industry, safety and security aspects were more or less not on the agenda in the early 90-ies. According to how society has developed, the experience of risk and danger has changed in the society over time and nowadays both safety and security as well as quality aspects has become important elements in the tourism products. The question is, if those two different approaches can be used as mutual methods.

Systematic Synthesis and Applications of Novel Multi-Degree-of-Freedom Differential Systems

1992 ◽  
Author(s):  
Sridhar Kota ◽  
Srinivas Bidare
Keyword(s):  
Degrees Of Freedom ◽  
Building Blocks ◽  
Degree Of Freedom ◽  
Differential Systems ◽  
Practical Applications ◽  
Epicyclic Gear ◽  
Differential Mechanism ◽  
Long Time ◽  
Gear Trains ◽  
Drive Systems

Abstract A two-degree-of-freedom differential system has been known for a long time and is widely used in automotive drive systems. Although higher degree-of-freedom differential systems have been developed in the past based on the well-known standard differential, the number of degrees-of-freedom has been severely restricted to 2n. Using a standard differential mechanism and simple epicyclic gear trains as differential building blocks, we have developed novel whiffletree-like differential systems that can provide n-degrees of freedom, where n is any integer greater than two. Symbolic notation for representing these novel differentials is also presented. This paper presents a systematic method of deriving multi-degree-of-freedom differential systems, a three and four output differential systems and some of their practical applications.

Coal gangue recognition using multichannel auditory spectrogram of hydraulic support sound in convolutional neural network

2021 ◽  
Author(s):  
xu chen ◽  
Shibo Wang ◽  
Houguang Liu ◽  
Jianhua Yang ◽  
Songyong Liu ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Spectral Characteristics ◽  
Coal Gangue ◽  
Hydraulic Support ◽  
Practical Applications ◽  
Network Training ◽  
Auditory Recognition ◽  
Auditory Signals ◽  
Fully Connected

Abstract Many data-driven coal gangue recognition (CGR) methods based on the vibration or sound of collapsed coal and gangue have been proposed to achieve automatic CGR, which is important for realizing intelligent top-coal caving. However, the strong background noise and complex environment in underground coal mines render this task challenging in practical applications. Inspired by the fact that workers distinguish coal and gangue from underground noise by listening to the hydraulic support sound, we propose an auditory model based CGR method that simulates human auditory recognition by combining an auditory spectrogram with a convolutional neural network (CNN). First, we adjust the characteristic frequency (CF) distribution of the auditory peripheral model (APM) based on the spectral characteristics of collapsed sound signals from coal and gangue and then process the sound signals using the adjusted APM to obtain inferior colliculus auditory signals with multiple CFs. Subsequently, the auditory signals of all CFs are converted into gray images separately and then concatenated into a multichannel auditory spectrum along the channel dimension. Finally, we input the multichannel auditory spectrum as a feature map to the two-dimensional CNN, whose convolutional layers are used to automatically extract features, and the fully connected layer and softmax layer are used to flatten features and predict the recognition result, respectively. The CNN is optimized for the CGR based on a comparison study of four typical types of CNN structures with different network training hyperparameters. The experimental results show that this method affords an accurate CGR with a recognition accuracy of 99.5%. Moreover, this method offers excellent noise immunity compared with typically used CGR methods under various noisy conditions.

Megnn: Meta-path extracted graph neural network for heterogeneous graph representation learning

2021 ◽  
pp. 107611
Author(s):  
Yaomin Chang ◽  
Chuan Chen ◽  
Weibo Hu ◽  
Zibin Zheng ◽  
Xiaocong Zhou ◽  
...  
Keyword(s):  
Neural Network ◽  
Representation Learning ◽  
Graph Representation ◽  
Meta Path
Sign in / Sign up

Export Citation Format

Share Document