scholarly journals Deep neural networks for climate relation extraction

2021 ◽  
Keyword(s):  
Time Series ◽  
Data Distribution ◽  
Relation Extraction ◽  
Feature Space ◽  
Good Alternative ◽  
Climate Data ◽  
Background Space ◽  
Proposed Model ◽  
Hidden Layer ◽  
Carbon Dioxide Co2

<p>Climate data composes of time series and space series with unknown. These unknown series contains complex co-variation relations of climate data. The extraction of these relations is essential for further revealing the complex representations between time series and space series in climate data. As an important application, through extracting these co-variation relations, we can further predict the change of climate to provide early warning for natural disasters, e.g., Greenhouse effect. Hence, it is a challenge to explore the relations between climate data. To address this, this work propose a deep neural network. Based on Brenier theorem, the loss function is derived. Since Brenier theorem rigorously proves that the data distribution in background space is consistent with the data distribution in the feature space with greatest probability, ensuring that the relations extracted from the latent space are as close to that of in background space as possible. Then, the parameters of time series consisting of eight variables are encoded by the first hidden-layer in the proposed model. The remaining two hidden-layers encode the latitude and longitude in spatial series, respectively. Experimental results show that the proposed method outperforms the state-of-the-art methods with respect to climate relations extracted. Hence, the proposed method is considered a good alternative in capturing relations between climate variables, as well as, between carbon dioxide (CO2) and surface temperature</p>

scholarly journals Clustering for Data-driven Unraveling Artificial Neural Networks

2020 ◽  
Author(s):  
Felipe Farias ◽  
Teresa Ludermir ◽  
Carmelo Bastos-Filho
Keyword(s):  
Neural Networks ◽  
Statistical Significance ◽  
Feature Space ◽  
Data Driven ◽  
P Value ◽  
Proposed Model ◽  
Data Driven Approach ◽  
Benchmark Datasets ◽  
Hidden Layer ◽  
Fold Cross Validation

This work presents an investigation on how to define Neural Networks (NN) architectures adopting a data-driven approach using clustering to create sub-labels to facilitate the learning process and to discover the number of neurons needed to compose the layers. We also increase the depth of the model aiming to represent the samples better, the more in-depth it flows into the model. We hypothesize that the clustering process identifies sub-regions in the feature space in which the samples belonging to the same cluster have strong similarities. We used seven benchmark datasets to validate our hypothesis using 10-fold cross validation 3 times. The proposed model increased the performance, while never decreased it, with statistical significance considering the p-value $< 0.05$ in comparison with a Multi-Layer Perceptron with a single hidden layer with approximately the same number of parameters of the architectures found by our approach.

scholarly journals A Neutrosophic Forecasting Model for Time Series Based on First-Order State and Information Entropy of High-Order Fluctuation

Entropy ◽  
2019 ◽  
Vol 21 (5) ◽  
pp. 455 ◽  
Author(s):  
Hongjun Guan ◽  
Zongli Dai ◽  
Shuang Guan ◽  
Aiwu Zhao
Keyword(s):  
Time Series ◽  
Information Entropy ◽  
Stock Exchange ◽  
Time Series Forecasting ◽  
High Order ◽  
Stock Index ◽  
Neutrosophic Set ◽  
Forecasting Model ◽  
Forecasting Models ◽  
Proposed Model

In time series forecasting, information presentation directly affects prediction efficiency. Most existing time series forecasting models follow logical rules according to the relationships between neighboring states, without considering the inconsistency of fluctuations for a related period. In this paper, we propose a new perspective to study the problem of prediction, in which inconsistency is quantified and regarded as a key characteristic of prediction rules. First, a time series is converted to a fluctuation time series by comparing each of the current data with corresponding previous data. Then, the upward trend of each of fluctuation data is mapped to the truth-membership of a neutrosophic set, while a falsity-membership is used for the downward trend. Information entropy of high-order fluctuation time series is introduced to describe the inconsistency of historical fluctuations and is mapped to the indeterminacy-membership of the neutrosophic set. Finally, an existing similarity measurement method for the neutrosophic set is introduced to find similar states during the forecasting stage. Then, a weighted arithmetic averaging (WAA) aggregation operator is introduced to obtain the forecasting result according to the corresponding similarity. Compared to existing forecasting models, the neutrosophic forecasting model based on information entropy (NFM-IE) can represent both fluctuation trend and fluctuation consistency information. In order to test its performance, we used the proposed model to forecast some realistic time series, such as the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX), the Shanghai Stock Exchange Composite Index (SHSECI), and the Hang Seng Index (HSI). The experimental results show that the proposed model can stably predict for different datasets. Simultaneously, comparing the prediction error to other approaches proves that the model has outstanding prediction accuracy and universality.

scholarly journals A Novel Time-Sensitive Composite Similarity Model for Multivariate Time-Series Correlation Analysis

Entropy ◽  
2021 ◽  
Vol 23 (6) ◽  
pp. 731
Author(s):  
Mengxia Liang ◽  
Xiaolong Wang ◽  
Shaocong Wu
Keyword(s):  
Time Series ◽  
Correlation Analysis ◽  
Stock Price ◽  
Multivariate Time Series ◽  
Temporal Features ◽  
Proposed Model ◽  
Time Series Segmentation ◽  
Similarity Model ◽  
Dynamic Time ◽  
Investment Portfolios

Finding the correlation between stocks is an effective method for screening and adjusting investment portfolios for investors. One single temporal feature or static nontemporal features are generally used in most studies to measure the similarity between stocks. However, these features are not sufficient to explore phenomena such as price fluctuations similar in shape but unequal in length which may be caused by multiple temporal features. To research stock price volatilities entirely, mining the correlation between stocks should be considered from the point view of multiple features described as time series, including closing price, etc. In this paper, a time-sensitive composite similarity model designed for multivariate time-series correlation analysis based on dynamic time warping is proposed. First, a stock is chosen as the benchmark, and the multivariate time series are segmented by the peaks and troughs time-series segmentation (PTS) algorithm. Second, similar stocks are screened out by similarity. Finally, the rate of rising or falling together between stock pairs is used to verify the proposed model’s effectiveness. Compared with other models, the composite similarity model brings in multiple temporal features and is generalizable for numerical multivariate time series in different fields. The results show that the proposed model is very promising.

Hierarchical Concept-Driven Language Model

2021 ◽  
Vol 15 (6) ◽  
pp. 1-22
Author(s):  
Yashen Wang ◽  
Huanhuan Zhang ◽  
Zhirun Liu ◽  
Qiang Zhou
Keyword(s):  
Language Model ◽  
Generation Process ◽  
Data Generation ◽  
Modeling Framework ◽  
Long Distance ◽  
Short Text ◽  
Proposed Model ◽  
Scalable Inference ◽  
End To End ◽  
Hidden Layer

For guiding natural language generation, many semantic-driven methods have been proposed. While clearly improving the performance of the end-to-end training task, these existing semantic-driven methods still have clear limitations: for example, (i) they only utilize shallow semantic signals (e.g., from topic models) with only a single stochastic hidden layer in their data generation process, which suffer easily from noise (especially adapted for short-text etc.) and lack of interpretation; (ii) they ignore the sentence order and document context, as they treat each document as a bag of sentences, and fail to capture the long-distance dependencies and global semantic meaning of a document. To overcome these problems, we propose a novel semantic-driven language modeling framework, which is a method to learn a Hierarchical Language Model and a Recurrent Conceptualization-enhanced Gamma Belief Network, simultaneously. For scalable inference, we develop the auto-encoding Variational Recurrent Inference, allowing efficient end-to-end training and simultaneously capturing global semantics from a text corpus. Especially, this article introduces concept information derived from high-quality lexical knowledge graph Probase, which leverages strong interpretability and anti-nose capability for the proposed model. Moreover, the proposed model captures not only intra-sentence word dependencies, but also temporal transitions between sentences and inter-sentence concept dependence. Experiments conducted on several NLP tasks validate the superiority of the proposed approach, which could effectively infer meaningful hierarchical concept structure of document and hierarchical multi-scale structures of sequences, even compared with latest state-of-the-art Transformer-based models.

scholarly journals Fully Automated Detection of Supraglacial Lake Area for Northeast Greenland Using Sentinel-2 Time-Series

2021 ◽  
Vol 13 (2) ◽  
pp. 205
Author(s):  
Philipp Hochreuther ◽  
Niklas Neckel ◽  
Nathalie Reimann ◽  
Angelika Humbert ◽  
Matthias Braun
Keyword(s):  
Time Series ◽  
Observation Interval ◽  
Lake Area ◽  
Climate Data ◽  
Melt Pond ◽  
Automated Processing ◽  
Average Observation ◽  
Average Size ◽  
Lake Size ◽  
Sentinel 2

The usability of multispectral satellite data for detecting and monitoring supraglacial meltwater ponds has been demonstrated for western Greenland. For a multitemporal analysis of large regions or entire Greenland, largely automated processing routines are required. Here, we present a sequence of algorithms that allow for an automated Sentinel-2 data search, download, processing, and generation of a consistent and dense melt pond area time-series based on open-source software. We test our approach for a ~82,000 km2 area at the 79°N Glacier (Nioghalvfjerdsbrae) in northeast Greenland, covering the years 2016, 2017, 2018 and 2019. Our lake detection is based on the ratio of the blue and red visible bands using a minimum threshold. To remove false classification caused by the similar spectra of shadow and water on ice, we implement a shadow model to mask out topographically induced artifacts. We identified 880 individual lakes, traceable over 479 time-steps throughout 2016–2019, with an average size of 64,212 m2. Of the four years, 2019 had the most extensive lake area coverage with a maximum of 333 km2 and a maximum individual lake size of 30 km2. With 1.5 days average observation interval, our time-series allows for a comparison with climate data of daily resolution, enabling a better understanding of short-term climate-glacier feedbacks.

scholarly journals Onboard Radio Frequency Interference as the Origin of Inter-Satellite Biases for Microwave Humidity Sounders

2019 ◽  
Vol 11 (7) ◽  
pp. 866 ◽  
Author(s):  
Imke Hans ◽  
Martin Burgdorf ◽  
Stefan A. Buehler
Keyword(s):  
Time Series ◽  
Radio Frequency ◽  
Time Dependent ◽  
Radio Frequency Interference ◽  
Climate Variables ◽  
Compelling Evidence ◽  
Climate Data ◽  
The Earth ◽  
Correction Scheme

Understanding the causes of inter-satellite biases in climate data records from observations of the Earth is crucial for constructing a consistent time series of the essential climate variables. In this article, we analyse the strong scan- and time-dependent biases observed for the microwave humidity sounders on board the NOAA-16 and NOAA-19 satellites. We find compelling evidence that radio frequency interference (RFI) is the cause of the biases. We also devise a correction scheme for the raw count signals for the instruments to mitigate the effect of RFI. Our results show that the RFI-corrected, recalibrated data exhibit distinctly reduced biases and provide consistent time series.

scholarly journals An Attention-Based Model Using Character Composition of Entities in Chinese Relation Extraction

Information ◽  
2020 ◽  
Vol 11 (2) ◽  
pp. 79 ◽  
Author(s):  
Xiaoyu Han ◽  
Yue Zhang ◽  
Wenkai Zhang ◽  
Tinglei Huang
Keyword(s):  
Language Processing ◽  
Large Scale ◽  
Named Entity Recognition ◽  
Relation Extraction ◽  
Entity Recognition ◽  
Additional Information ◽  
Named Entity ◽  
Proposed Model ◽  
The Relationship ◽  
Crucial Part

Relation extraction is a vital task in natural language processing. It aims to identify the relationship between two specified entities in a sentence. Besides information contained in the sentence, additional information about the entities is verified to be helpful in relation extraction. Additional information such as entity type getting by NER (Named Entity Recognition) and description provided by knowledge base both have their limitations. Nevertheless, there exists another way to provide additional information which can overcome these limitations in Chinese relation extraction. As Chinese characters usually have explicit meanings and can carry more information than English letters. We suggest that characters that constitute the entities can provide additional information which is helpful for the relation extraction task, especially in large scale datasets. This assumption has never been verified before. The main obstacle is the lack of large-scale Chinese relation datasets. In this paper, first, we generate a large scale Chinese relation extraction dataset based on a Chinese encyclopedia. Second, we propose an attention-based model using the characters that compose the entities. The result on the generated dataset shows that these characters can provide useful information for the Chinese relation extraction task. By using this information, the attention mechanism we used can recognize the crucial part of the sentence that can express the relation. The proposed model outperforms other baseline models on our Chinese relation extraction dataset.

scholarly journals Systematic Water Fraction Estimation for a Global and Daily Surface Water Time-Series

2021 ◽  
Vol 13 (14) ◽  
pp. 2675
Author(s):  
Stefan Mayr ◽  
Igor Klein ◽  
Martin Rutzinger ◽  
Claudia Kuenzer
Keyword(s):  
Time Series ◽  
Surface Water ◽  
Spatial Resolution ◽  
Pure Water ◽  
Feature Space ◽  
Observation Time ◽  
Temporal Information ◽  
Landsat 8 ◽  
Water Fraction ◽  
Classification Probability

Fresh water is a vital natural resource. Earth observation time-series are well suited to monitor corresponding surface dynamics. The DLR-DFD Global WaterPack (GWP) provides daily information on globally distributed inland surface water based on MODIS (Moderate Resolution Imaging Spectroradiometer) images at 250 m spatial resolution. Operating on this spatiotemporal level comes with the drawback of moderate spatial resolution; only coarse pixel-based surface water quantification is possible. To enhance the quantitative capabilities of this dataset, we systematically access subpixel information on fractional water coverage. For this, a linear mixture model is employed, using classification probability and pure pixel reference information. Classification probability is derived from relative datapoint (pixel) locations in feature space. Pure water and non-water reference pixels are located by combining spatial and temporal information inherent to the time-series. Subsequently, the model is evaluated for different input sets to determine the optimal configuration for global processing and pixel coverage types. The performance of resulting water fraction estimates is evaluated on the pixel level in 32 regions of interest across the globe, by comparison to higher resolution reference data (Sentinel-2, Landsat 8). Results show that water fraction information is able to improve the product’s performance regarding mixed water/non-water pixels by an average of 11.6% (RMSE). With a Nash-Sutcliffe efficiency of 0.61, the model shows good overall performance. The approach enables the systematic provision of water fraction estimates on a global and daily scale, using only the reflectance and temporal information contained in the input time-series.

Relationship between Selic rate and Basel III parameters - A statistics approach and a fuzzy forecasting model

2021 ◽  
pp. 1-14
Author(s):  
Thiago Henrique Barbosa de Carvalho Tavares ◽  
Bruno Pérez Ferreira ◽  
Eduardo Mazoni Andrade Marçal Mendes
Keyword(s):  
Cross Correlation ◽  
Data Distribution ◽  
Forecasting Model ◽  
Basel Iii ◽  
Output Data ◽  
Basel Accords ◽  
Proposed Model ◽  
The Relationship ◽  
The Universe ◽  
Universe Of Discourse

In this work the relationship between the Selic rate and some bank parameters defined by the so-called Basel Accords is studied. The cross-correlation between the Selic rate and the parameters is used to explain how these parameters affect the Selic rate and vice-versa so as to define the predictability of the Selic rate using (some of) these parameters as inputs. A model is then proposed for predicting the Selic rate based on some specific parameters using fuzzy logic ideas, which dealt with a partitioning of the universe of discourse using clusters related to the output data distribution. The proposed model is compared to four other known models in the literature and showed to have better performance in average compared to all other models.

Sign in / Sign up

Export Citation Format

Share Document