scholarly journals Constraining the astrophysics and cosmology from 21 cm tomography using deep learning with the SKA

2020 ◽  
Vol 494 (4) ◽  
pp. 5761-5774 ◽  
Author(s):  
Sultan Hassan ◽  
Sambatra Andrianomena ◽  
Caitlin Doughty
Keyword(s):  
Thermal Noise ◽  
Cosmological Parameters ◽  
Hubble Constant ◽  
Great Accuracy ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Density Parameter ◽  
Huge Data ◽  
Learning Techniques ◽  
Fluctuation Amplitude

ABSTRACT Future Square Kilometre Array (SKA) surveys are expected to generate huge data sets of 21 cm maps on cosmological scales from the Epoch of Reionization. We assess the viability of exploiting machine learning techniques, namely, convolutional neural networks (CNNs), to simultaneously estimate the astrophysical and cosmological parameters from 21 cm maps from seminumerical simulations. We further convert the simulated 21 cm maps into SKA-like mock maps using the detailed SKA antennae distribution, thermal noise, and a recipe for foreground cleaning. We successfully design two CNN architectures (VGGNet-like and ResNet-like) that are both efficiently able to extract simultaneously three astrophysical parameters, namely the photon escape fraction (fesc), the ionizing emissivity power dependence on halo mass (Cion), and the ionizing emissivity redshift evolution index (Dion), and three cosmological parameters, namely the matter density parameter (Ωm), the dimensionless Hubble constant (h), and the matter fluctuation amplitude (σ8), from 21 cm maps at several redshifts. With the presence of noise from SKA, our designed CNNs are still able to recover these astrophysical and cosmological parameters with great accuracy ($R^{2} \gt 92{{\ \rm per\ cent}}$), improving to $R^{2} \gt 99{{\ \rm per\ cent}}$ towards low-redshift and low neutral fraction values. Our results show that future 21 cm observations can play a key role to break degeneracy between models and tightly constrain the astrophysical and cosmological parameters, using only few frequency channels.

scholarly journals A Study on Machine Learning in Big Data

2017 ◽  
Vol 10 (3) ◽  
pp. 660-663
Author(s):  
L. Dhanapriya ◽  
Dr. S. MANJU
Keyword(s):  
Machine Learning ◽  
Big Data ◽  
Large Data ◽  
Machine Learning Algorithms ◽  
Large Data Sets ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Huge Data ◽  
Learning Techniques ◽  
Market Needs

In the recent development of IT technology, the capacity of data has surpassed the zettabyte, and improving the efficiency of business is done by increasing the ability of predictive through an efficient analysis on these data which has emerged as an issue in the current society. Now the market needs for methods that are capable of extracting valuable information from large data sets. Recently big data is becoming the focus of attention, and using any of the machine learning techniques to extract the valuable information from the huge data of complex structures has become a concern yet an urgent problem to resolve. The aim of this work is to provide a better understanding of this Machine Learning technique for discovering interesting patterns and introduces some machine learning algorithms to explore the developing trend.

Improving Reliability Estimation for Individual Numeric Predictions: A Machine Learning Approach

2021 ◽  
Author(s):  
Gediminas Adomavicius ◽  
Yaqiong Wang
Keyword(s):  
Machine Learning ◽  
General Purpose ◽  
Reliability Estimation ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Real World Data ◽  
Learning Techniques ◽  
Reliability Indicator ◽  
Machine Learning Approach ◽  
Prediction Reliability

Numerical predictive modeling is widely used in different application domains. Although many modeling techniques have been proposed, and a number of different aggregate accuracy metrics exist for evaluating the overall performance of predictive models, other important aspects, such as the reliability (or confidence and uncertainty) of individual predictions, have been underexplored. We propose to use estimated absolute prediction error as the indicator of individual prediction reliability, which has the benefits of being intuitive and providing highly interpretable information to decision makers, as well as allowing for more precise evaluation of reliability estimation quality. As importantly, the proposed reliability indicator allows the reframing of reliability estimation itself as a canonical numeric prediction problem, which makes the proposed approach general-purpose (i.e., it can work in conjunction with any outcome prediction model), alleviates the need for distributional assumptions, and enables the use of advanced, state-of-the-art machine learning techniques to learn individual prediction reliability patterns directly from data. Extensive experimental results on multiple real-world data sets show that the proposed machine learning-based approach can significantly improve individual prediction reliability estimation as compared with a number of baselines from prior work, especially in more complex predictive scenarios.

scholarly journals Decision Tree: A Machine Learning for Intrusion Detection

2019 ◽  
Vol 8 (6S4) ◽  
pp. 1126-1130
Keyword(s):  
Machine Learning ◽  
Intrusion Detection ◽  
Detection System ◽  
Research Work ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Legitimate User ◽  
Learning Techniques ◽  
Three Stages

The Intrusion is a major threat to unauthorized data or legal network using the legitimate user identity or any of the back doors and vulnerabilities in the network. IDS mechanisms are developed to detect the intrusions at various levels. The objective of the research work is to improve the Intrusion Detection System performance by applying machine learning techniques based on decision trees for detection and classification of attacks. The methodology adapted will process the datasets in three stages. The experimentation is conducted on KDDCUP99 data sets based on number of features. The Bayesian three modes are analyzed for different sized data sets based upon total number of attacks. The time consumed by the classifier to build the model is analyzed and the accuracy is done.

scholarly journals Using quasar X-ray and UV flux measurements to constrain cosmological model parameters

2020 ◽  
Vol 497 (1) ◽  
pp. 263-278 ◽  
Author(s):  
Narayan Khadka ◽  
Bharat Ratra
Keyword(s):  
Cosmological Parameters ◽  
Hubble Constant ◽  
Acoustic Oscillation ◽  
Joint Analysis ◽  
Model Parameters ◽  
Density Parameter ◽  
Current Standard ◽  
X Ray ◽  
Flux Measurements ◽  
Parameter Constraints

ABSTRACT Risaliti and Lusso have compiled X-ray and UV flux measurements of 1598 quasars (QSOs) in the redshift range 0.036 ≤ z ≤ 5.1003, part of which, z ∼ 2.4 − 5.1, is largely cosmologically unprobed. In this paper we use these QSO measurements, alone and in conjunction with baryon acoustic oscillation (BAO) and Hubble parameter [H(z)] measurements, to constrain cosmological parameters in six different cosmological models, each with two different Hubble constant priors. In most of these models, given the larger uncertainties, the QSO cosmological parameter constraints are mostly consistent with those from the BAO + H(z) data. A somewhat significant exception is the non-relativistic matter density parameter Ωm0 where QSO data favour Ωm0 ∼ 0.5 − 0.6 in most models. As a result, in joint analyses of QSO data with H(z) + BAO data the 1D Ωm0 distributions shift slightly towards larger values. A joint analysis of the QSO + BAO + H(z) data is consistent with the current standard model, spatially-flat ΛCDM, but mildly favours closed spatial hypersurfaces and dynamical dark energy. Since the higher Ωm0 values favoured by QSO data appear to be associated with the z ∼ 2 − 5 part of these data, and conflict somewhat with strong indications for Ωm0 ∼ 0.3 from most z < 2.5 data as well as from the cosmic microwave background anisotropy data at z ∼ 1100, in most models, the larger QSO data Ωm0 is possibly more indicative of an issue with the z ∼ 2 − 5 QSO data than of an inadequacy of the standard flat ΛCDM model.

scholarly journals Cosmological constraints from H ii starburst galaxy apparent magnitude and other cosmological measurements

2020 ◽  
Vol 497 (3) ◽  
pp. 3191-3203 ◽  
Author(s):  
Shulei Cao ◽  
Joseph Ryan ◽  
Bharat Ratra
Keyword(s):  
Dark Energy ◽  
Cosmological Parameters ◽  
Angular Size ◽  
Hubble Constant ◽  
Joint Analysis ◽  
Apparent Magnitude ◽  
Starburst Galaxy ◽  
Density Parameter ◽  
Acoustic Oscillations ◽  
Combined Data

ABSTRACT We use H ii starburst galaxy apparent magnitude measurements to constrain cosmological parameters in six cosmological models. A joint analysis of H ii galaxy, quasar angular size, baryon acoustic oscillations peak length scale, and Hubble parameter measurements result in relatively model-independent and restrictive estimates of the current values of the non-relativistic matter density parameter $\Omega _{\rm m_0}$ and the Hubble constant H0. These estimates favour a 2.0–3.4σ (depending on cosmological model) lower H0 than what is measured from the local expansion rate. The combined data are consistent with dark energy being a cosmological constant and with flat spatial hypersurfaces, but do not strongly rule out mild dark energy dynamics or slightly non-flat spatial geometries.

scholarly journals Malicious web domain identification using online credibility and performance data by considering the class imbalance issue

2019 ◽  
Vol 119 (3) ◽  
pp. 676-696 ◽  
Author(s):  
Zhongyi Hu ◽  
Raymond Chiong ◽  
Ilung Pranata ◽  
Yukun Bao ◽  
Yuqing Lin
Keyword(s):  
Machine Learning ◽  
Class Imbalance ◽  
Performance Data ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Real World Data ◽  
Content Type ◽  
Domain Identification ◽  
Learning Techniques ◽  
And Performance

Purpose Malicious web domain identification is of significant importance to the security protection of internet users. With online credibility and performance data, the purpose of this paper to investigate the use of machine learning techniques for malicious web domain identification by considering the class imbalance issue (i.e. there are more benign web domains than malicious ones). Design/methodology/approach The authors propose an integrated resampling approach to handle class imbalance by combining the synthetic minority oversampling technique (SMOTE) and particle swarm optimisation (PSO), a population-based meta-heuristic algorithm. The authors use the SMOTE for oversampling and PSO for undersampling. Findings By applying eight well-known machine learning classifiers, the proposed integrated resampling approach is comprehensively examined using several imbalanced web domain data sets with different imbalance ratios. Compared to five other well-known resampling approaches, experimental results confirm that the proposed approach is highly effective. Practical implications This study not only inspires the practical use of online credibility and performance data for identifying malicious web domains but also provides an effective resampling approach for handling the class imbalance issue in the area of malicious web domain identification. Originality/value Online credibility and performance data are applied to build malicious web domain identification models using machine learning techniques. An integrated resampling approach is proposed to address the class imbalance issue. The performance of the proposed approach is confirmed based on real-world data sets with different imbalance ratios.

scholarly journals A fusion fuzzy model for detecting phony accounts in social networks

2018 ◽  
Vol 7 (4) ◽  
pp. 2738
Author(s):  
P. Srinivas Rao ◽  
Jayadev Gyani ◽  
G. Narsimha
Keyword(s):  
Social Network ◽  
Fuzzy Model ◽  
Machine Learning Techniques ◽  
Personal Attributes ◽  
Data Sets ◽  
Learning Techniques ◽  
The Social ◽  
Using Data ◽  
Removal Technique ◽  
Normal Rank

In online social network’s phony account detection is one of the major task among the ability of genuine user from forged user account. The fundamental objective of detection of phony account framework is to detect fake account and removal technique in Social network user sites. This work concentrates on detection of phony account in which it depends on normal basis framework, transformative Algorithms and fuzzy technique. Initially, the most essential attributes including personal attributes, comparability techniques and various real user review, tweets, or comments are extricated. A direct blend of these attributes demonstrates the significance of each reviews tweets comments etc. To compute closeness measure, a consolidated strategy in view of artificial honey bee state Algorithm and fuzzy technique are utilized. Second approach is proposed to alter the best weights of the normal user attributes utilizing the social network activities/transaction and inherited Algorithm. Finally, a normal rank rationale framework is utilized to calculate the final scoring of normal user activities. The decision making of proposed approach to find phony account are variation with existing techniques user behavioral analysis using data sets and machine learning techniques such as crowdflower_sample and genuine_accounts_sample dataset of facebook and Twitter. The outcomes demonstrate that proposed strategy overcomes the previously mentioned strategies. 

Hybrid System based on Rough Sets and Genetic Algorithms for Medical Data Classifications

2013 ◽  
Vol 3 (4) ◽  
pp. 31-46 ◽  
Author(s):  
Hanaa Ismail Elshazly ◽  
Ahmad Taher Azar ◽  
Aboul Ella Hassanien ◽  
Abeer Mohamed Elkorany
Keyword(s):  
Genetic Algorithm ◽  
Hybrid System ◽  
Selection Process ◽  
Decision Rules ◽  
Medical Data ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Biomedical Data ◽  
Hybrid Techniques ◽  
Learning Techniques

Computational intelligence provides the biomedical domain by a significant support. The application of machine learning techniques in medical applications have been evolved from the physician needs. Screening, medical images, pattern classification, prognosis are some examples of health care support systems. Typically medical data has its own characteristics such as huge size and features, continuous and real attributes that refer to patients' investigations. Therefore, discretization and feature selection process are considered a key issue in improving the extracted knowledge from patients' investigations records. In this paper, a hybrid system that integrates Rough Set (RS) and Genetic Algorithm (GA) is presented for the efficient classification of medical data sets of different sizes and dimensionalities. Genetic Algorithm is applied with the aim of reducing the dimension of medical datasets and RS decision rules were used for efficient classification. Furthermore, the proposed system applies the Entropy Gain Information (EI) for discretization process. Four biomedical data sets are tested by the proposed system (EI-GA-RS), and the highest score was obtained through three different datasets. Other different hybrid techniques shared the proposed technique the highest accuracy but the proposed system preserves its place as one of the highest results systems four three different sets. EI as discretization technique also is a common part for the best results in the mentioned datasets while RS as an evaluator realized the best results in three different data sets.

Application of multi-sensor unmanned aerial system for identification of hydrothermal alteration zones

2020 ◽  
Author(s):  
Yosoon Choi ◽  
Jieun Baek ◽  
Jangwon Suh ◽  
Sung-Min Kim
Keyword(s):  
Machine Learning ◽  
Classification Accuracy ◽  
Training Data ◽  
Sensor Data ◽  
Machine Learning Techniques ◽  
Integrated Analysis ◽  
Unmanned Aerial System ◽  
Data Sets ◽  
Learning Techniques ◽  
Hydrothermal Alteration Zones

<p>In this study, we proposed a method to utilize a multi-sensor Unmanned Aerial System (UAS) for exploration of hydrothermal alteration zones. This study selected an area (10m × 20m) composed mainly of the andesite and located on the coast, with wide outcrops and well-developed structural and mineralization elements. Multi-sensor (visible, multispectral, thermal, magnetic) data were acquired in the study area using UAS, and were studied using machine learning techniques. For utilizing the machine learning techniques, we applied the stratified random method to sample 1000 training data in the hydrothermal zone and 1000 training data in the non-hydrothermal zone identified through the field survey. The 2000 training data sets created for supervised learning were first classified into 1500 for training and 500 for testing. Then, 1500 for training were classified into 1200 for training and 300 for validation. The training and validation data for machine learning were generated in five sets to enable cross-validation. Five types of machine learning techniques were applied to the training data sets: k-Nearest Neighbors (k-NN), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Deep Neural Network (DNN). As a result of integrated analysis of multi-sensor data using five types of machine learning techniques, RF and SVM techniques showed high classification accuracy of about 90%. Moreover, performing integrated analysis using multi-sensor data showed relatively higher classification accuracy in all five machine learning techniques than analyzing magnetic sensing data or single optical sensing data only.</p>

Sign in / Sign up

Export Citation Format

Share Document