-
Thermodynamics of Giant Molecular Clouds: The Effects of Dust Grain Size
Authors:
Nadine H. Soliman,
Philip F. Hopkins,
Michael Y. Grudić
Abstract:
The dust grain size distribution (GSD) likely varies significantly across different star-forming environments in the Universe, but the overall impact of this variation on star formation remains unclear. This ambiguity arises because the GSD interacts non-linearly with processes like heating/cooling, radiation, and chemistry, which have competing effects and different environmental dependencies. In…
▽ More
The dust grain size distribution (GSD) likely varies significantly across different star-forming environments in the Universe, but the overall impact of this variation on star formation remains unclear. This ambiguity arises because the GSD interacts non-linearly with processes like heating/cooling, radiation, and chemistry, which have competing effects and different environmental dependencies. In this study, we investigate the effects of GSD variation on the thermochemistry and evolution of giant molecular clouds (GMCs). To achieve this, we conducted radiation-dust-magnetohydrodynamic simulations spanning a range of cloud masses and grain sizes, which explicitly incorporate the dynamics of dust grains within the full-physics framework of the STARFORGE project. We find that differences in grain size significantly alter the thermochemistry of GMCs. Specifically, we show that the leading-order effect is that larger grains, under fixed dust mass and dust-to-gas ratio conditions, result in lower dust opacities. This reduced opacity permits ISRF photons to penetrate more deeply and allows internal radiation field photons to permeate more extensively into the cloud, resulting in rapid gas heating and the inhibition of star formation. We find that star formation efficiency is highly sensitive to grain size, with an order of magnitude reduction in efficiency when grain size increases from 0.1 $\rmμm$ to 10 $\rmμm$. Additionally, we note that warmer gas suppresses the formation of low-mass stars. Moreover, as a consequence of the decreased opacities, we observe a greater proportion of gas residing in diffuse ionized structures.
△ Less
Submitted 12 July, 2024;
originally announced July 2024.
-
Dust-Evacuated Zones Near Massive Stars: Consequences of Dust Dynamics on Star-forming Regions
Authors:
Nadine H. Soliman,
Philip F. Hopkins,
Michael Y. Grudić
Abstract:
Stars form within dense cores composed of both gas and dust within molecular clouds. However, despite the crucial role that dust plays in the star formation process, its dynamics is frequently overlooked, with the common assumption being a constant, spatially uniform dust-to-gas ratio and grain size spectrum. In this study, we introduce a set of radiation-dust-magnetohydrodynamic simulations of st…
▽ More
Stars form within dense cores composed of both gas and dust within molecular clouds. However, despite the crucial role that dust plays in the star formation process, its dynamics is frequently overlooked, with the common assumption being a constant, spatially uniform dust-to-gas ratio and grain size spectrum. In this study, we introduce a set of radiation-dust-magnetohydrodynamic simulations of star forming molecular clouds from the {\small STARFORGE} project. These simulations expand upon the earlier radiation MHD models, which included cooling, individual star formation, and feedback. Notably, they explicitly address the dynamics of dust grains, considering radiation, drag, and Lorentz forces acting on a diverse size spectrum of live dust grains. We find that interactions between radiation and dust significantly influence the properties of gas surrounding and accreting onto massive stars. Specifically, we find that once stars exceed a certain mass threshold ($\sim 2 M_{\odot}$), their emitted radiation can evacuate dust grains from their vicinity, giving rise to a dust-suppressed zone of size $\sim 100$ AU. Commencing during the early accretion stages and preceding the Main-sequence phase, this process results in a mass-dependent depletion in the accreted dust-to-gas (ADG) mass ratio within both the circumstellar disc and the star. We predict massive stars ($\gtrsim 10 M_{\odot}$) would exhibit ADG ratios that are approximately one order of magnitude lower than that of their parent clouds. Consequently, stars, their discs, and circumstellar environments would display notable deviations in the abundances of elements commonly associated with dust grains, such as carbon and oxygen.
△ Less
Submitted 13 June, 2024;
originally announced June 2024.
-
Microphysical Regulation of Non-Ideal MHD in Weakly-Ionized Systems: Does the Hall Effect Matter?
Authors:
Philip F. Hopkins,
Jonathan Squire,
Raphael Skalidis,
Nadine H. Soliman
Abstract:
The magnetohydrodynamics (MHD) equations plus 'non-ideal' (Ohmic, Hall, ambipolar) resistivities are widely used to model weakly-ionized astrophysical systems. We show that if gradients in the magnetic field become too steep, the implied charge drift speeds become much faster than microphysical signal speeds, invalidating the assumptions used to derive both the resistivities and MHD equations them…
▽ More
The magnetohydrodynamics (MHD) equations plus 'non-ideal' (Ohmic, Hall, ambipolar) resistivities are widely used to model weakly-ionized astrophysical systems. We show that if gradients in the magnetic field become too steep, the implied charge drift speeds become much faster than microphysical signal speeds, invalidating the assumptions used to derive both the resistivities and MHD equations themselves. Generically this situation will excite microscale instabilities that suppress the drift and current. We show this could be relevant at low ionization fractions especially if Hall terms appear significant, external forces induce supersonic motions, or dust grains become a dominant charge carrier. Considering well-established treatments of super-thermal drifts in laboratory, terrestrial, and Solar plasmas as well as conduction and viscosity models, we generalize a simple prescription to rectify these issues, where the resistivities are multiplied by a correction factor that depends only on already-known macroscopic quantities. This is generalized for multi-species and weakly-ionized systems, and leaves the equations unchanged in the drift limits for which they are derived, but restores physical behavior (driving the system back towards slow drift by diffusing away small-scale gradients in the magnetic field) if the limits are violated. This has important consequences: restoring intuitive behaviors such as the system becoming hydrodynamic in the limit of zero ionization; suppressing magnetic structure on scales below a critical length which can comparable to circumstellar disk sizes; limiting the maximum magnetic amplification; and suppressing the effects of the Hall term in particular. This likely implies that the Hall term does not become dynamically important under most conditions of interest in these systems.
△ Less
Submitted 9 May, 2024;
originally announced May 2024.
-
Co-Evolution vs. Co-existence: The Effect of Accretion Modelling on the Evolution of Black Holes and Host Galaxies
Authors:
Nadine H. Soliman,
Andrea V. Macciò,
Marvin Blank
Abstract:
We append two additional black hole (BH) accretion models, namely viscous disc and gravitational torque-driven accretion, into the Numerical Investigation of a Hundred Astrophysical Objects (NIHAO) project of galaxy simulations. We show that these accretion models, characterized by a weaker dependence on the BH mass compared to the commonly used Bondi-Hoyle accretion, naturally create a common evo…
▽ More
We append two additional black hole (BH) accretion models, namely viscous disc and gravitational torque-driven accretion, into the Numerical Investigation of a Hundred Astrophysical Objects (NIHAO) project of galaxy simulations. We show that these accretion models, characterized by a weaker dependence on the BH mass compared to the commonly used Bondi-Hoyle accretion, naturally create a common evolutionary track (co-existence) between the mass of the BH and the stellar mass of the galaxy, even without any direct coupling via feedback (FB). While FB is indeed required to control the final BH and stellar mass of the galaxies, our results suggest that FB might not be the leading driver of the cosmic co-evolution between these two quantities; in these models, co-evolution is simply determined by the shared central gas supply. Conversely, simulations using Bondi-Hoyle accretion show a two-step evolution, with an early growth of stellar mass followed by exponential growth of the central supermassive black hole (SMBH). Our results show that the modelling of BH accretion (sometimes overlooked) is an extremely important part of BH evolution and can improve our understanding of how scaling relations emerge and evolve, and whether SMBH and stellar mass co-exist or co-evolve through cosmic time.
△ Less
Submitted 25 July, 2023;
originally announced July 2023.
-
Android Malware Detection using Machine learning: A Review
Authors:
Md Naseef-Ur-Rahman Chowdhury,
Ahshanul Haque,
Hamdy Soliman,
Mohammad Sahinur Hossen,
Tanjim Fatima,
Imtiaz Ahmed
Abstract:
Malware for Android is becoming increasingly dangerous to the safety of mobile devices and the data they hold. Although machine learning(ML) techniques have been shown to be effective at detecting malware for Android, a comprehensive analysis of the methods used is required. We review the current state of Android malware detection us ing machine learning in this paper. We begin by providing an ove…
▽ More
Malware for Android is becoming increasingly dangerous to the safety of mobile devices and the data they hold. Although machine learning(ML) techniques have been shown to be effective at detecting malware for Android, a comprehensive analysis of the methods used is required. We review the current state of Android malware detection us ing machine learning in this paper. We begin by providing an overview of Android malware and the security issues it causes. Then, we look at the various supervised, unsupervised, and deep learning machine learning approaches that have been utilized for Android malware detection. Addi tionally, we present a comparison of the performance of various Android malware detection methods and talk about the performance evaluation metrics that are utilized to evaluate their efficacy. Finally, we draw atten tion to the drawbacks and difficulties of the methods that are currently in use and suggest possible future directions for research in this area. In addition to providing insights into the current state of Android malware detection using machine learning, our review provides a comprehensive overview of the subject.
△ Less
Submitted 15 March, 2023;
originally announced July 2023.
-
Wireless Sensor Networks anomaly detection using Machine Learning: A Survey
Authors:
Ahsnaul Haque,
Md Naseef-Ur-Rahman Chowdhury,
Hamdy Soliman,
Mohammad Sahinur Hossen,
Tanjim Fatima,
Imtiaz Ahmed
Abstract:
Wireless Sensor Networks (WSNs) have become increasingly valuable in various civil/military applications like industrial process control, civil engineering applications such as buildings structural strength monitoring, environmental monitoring, border intrusion, IoT (Internet of Things), and healthcare. However, the sensed data generated by WSNs is often noisy and unreliable, making it a challenge…
▽ More
Wireless Sensor Networks (WSNs) have become increasingly valuable in various civil/military applications like industrial process control, civil engineering applications such as buildings structural strength monitoring, environmental monitoring, border intrusion, IoT (Internet of Things), and healthcare. However, the sensed data generated by WSNs is often noisy and unreliable, making it a challenge to detect and diagnose anomalies. Machine learning (ML) techniques have been widely used to address this problem by detecting and identifying unusual patterns in the sensed data. This survey paper provides an overview of the state of the art applications of ML techniques for data anomaly detection in WSN domains. We first introduce the characteristics of WSNs and the challenges of anomaly detection in WSNs. Then, we review various ML techniques such as supervised, unsupervised, and semi-supervised learning that have been applied to WSN data anomaly detection. We also compare different ML-based approaches and their performance evaluation metrics. Finally, we discuss open research challenges and future directions for applying ML techniques in WSNs sensed data anomaly detection.
△ Less
Submitted 15 March, 2023;
originally announced March 2023.
-
Cross-Domain Neural Entity Linking
Authors:
Hassan Soliman
Abstract:
Entity Linking is the task of matching a mention to an entity in a given knowledge base (KB). It contributes to annotating a massive amount of documents existing on the Web to harness new facts about their matched entities. However, existing Entity Linking systems focus on develo** models that are typically domain-dependent and robust only to a particular knowledge base on which they have been t…
▽ More
Entity Linking is the task of matching a mention to an entity in a given knowledge base (KB). It contributes to annotating a massive amount of documents existing on the Web to harness new facts about their matched entities. However, existing Entity Linking systems focus on develo** models that are typically domain-dependent and robust only to a particular knowledge base on which they have been trained. The performance is not as adequate when being evaluated on documents and knowledge bases from different domains.
Approaches based on pre-trained language models, such as Wu et al. (2020), attempt to solve the problem using a zero-shot setup, illustrating some potential when evaluated on a general-domain KB. Nevertheless, the performance is not equivalent when evaluated on a domain-specific KB. To allow for more accurate Entity Linking across different domains, we propose our framework: Cross-Domain Neural Entity Linking (CDNEL). Our objective is to have a single system that enables simultaneous linking to both the general-domain KB and the domain-specific KB. CDNEL works by learning a joint representation space for these knowledge bases from different domains. It is evaluated using the external Entity Linking dataset (Zeshel) constructed by Logeswaran et al. (2019) and the Reddit dataset collected by Botzer et al. (2021), to compare our proposed method with the state-of-the-art results. The proposed framework uses different types of datasets for fine-tuning, resulting in different model variants of CDNEL. When evaluated on four domains included in the Zeshel dataset, these variants achieve an average precision gain of 9%.
△ Less
Submitted 28 September, 2022;
originally announced October 2022.
-
Dust Dynamics in AGN Winds: A New Mechanism For Multiwavelength AGN Variability
Authors:
Nadine H. Soliman,
Philip F. Hopkins
Abstract:
Partial dust obscuration in active galactic nuclei (AGN) has been proposed as a potential explanation for some cases of AGN variability. The dust-gas mixture present in AGN torii is accelerated by radiation pressure leading to the launching of an AGN wind. Dust under these conditions has been shown to be unstable to a generic class of fast growing resonant drag instabilities (RDIs). We present the…
▽ More
Partial dust obscuration in active galactic nuclei (AGN) has been proposed as a potential explanation for some cases of AGN variability. The dust-gas mixture present in AGN torii is accelerated by radiation pressure leading to the launching of an AGN wind. Dust under these conditions has been shown to be unstable to a generic class of fast growing resonant drag instabilities (RDIs). We present the first set of numerical simulations of radiation driven outflows that include explicit dust dynamics in conditions resembling AGN winds and discuss the implications of the RDIs on the morphology of the AGN torus, AGN variability, and the ability of the radiation to effectively launch a wind. We find that the RDIs rapidly develop reaching saturation at times much shorter than the global timescales of the outflows, resulting in the formation of filamentary structure on box-size scales with strong dust clum** and super-Alfvénic velocity dispersions on micro-scales. This results in 10-20% fluctuations in dust opacity and gas column density, integrated along mock observed lines-of-sight to the quasar accretion disk, over year to decade timescales with a red-noise power spectrum which is commonly observed for AGN. Additionally, all our simulations show that the radiation is sufficiently coupled to the dust-gas mixture launching highly super-sonic winds, which entrain 70-90% of gas, with a factor of $\lesssim 3$ photon momentum loss relative to the ideal case. Therefore, the RDIs could play an important role in generating the clumpy nature of AGN torii and driving AGN variability consistent with observations.
△ Less
Submitted 24 October, 2022;
originally announced October 2022.
-
Identifying Difficult exercises in an eTextbook Using Item Response Theory and Logged Data Analysis
Authors:
Ahmed Abd Elrahman,
Ahmed I. Taloba,
Mohammed F. Farghally,
Taysir Hassan A Soliman
Abstract:
The growing dependence on eTextbooks and Massive Open Online Courses (MOOCs) has led to an increase in the amount of students' learning data. By carefully analyzing this data, educators can identify difficult exercises, and evaluate the quality of the exercises when teaching a particular topic. In this study, an analysis of log data from the semester usage of the OpenDSA eTextbook was offered to i…
▽ More
The growing dependence on eTextbooks and Massive Open Online Courses (MOOCs) has led to an increase in the amount of students' learning data. By carefully analyzing this data, educators can identify difficult exercises, and evaluate the quality of the exercises when teaching a particular topic. In this study, an analysis of log data from the semester usage of the OpenDSA eTextbook was offered to identify the most difficult data structure course exercises and to evaluate the quality of the course exercises. Our study is based on analyzing students' responses to the course exercises. We applied item response theory (IRT) analysis and a latent trait mode (LTM) to identify the most difficult exercises .To evaluate the quality of the course exercises we applied IRT theory. Our findings showed that the exercises that related to algorithm analysis topics represented the most difficult exercises, and there existing six exercises were classified as poor exercises which could be improved or need some attention.
△ Less
Submitted 25 November, 2022; v1 submitted 11 October, 2022;
originally announced October 2022.
-
Data Augmentation using Feature Generation for Volumetric Medical Images
Authors:
Khushboo Mehra,
Hassan Soliman,
Soumya Ranjan Sahoo
Abstract:
Medical image classification is one of the most critical problems in the image recognition area. One of the major challenges in this field is the scarcity of labelled training data. Additionally, there is often class imbalance in datasets as some cases are very rare to happen. As a result, accuracy in classification task is normally low. Deep Learning models, in particular, show promising results…
▽ More
Medical image classification is one of the most critical problems in the image recognition area. One of the major challenges in this field is the scarcity of labelled training data. Additionally, there is often class imbalance in datasets as some cases are very rare to happen. As a result, accuracy in classification task is normally low. Deep Learning models, in particular, show promising results on image segmentation and classification problems, but they require very large datasets for training. Therefore, there is a need to generate more of synthetic samples from the same distribution. Previous work has shown that feature generation is more efficient and leads to better performance than corresponding image generation. We apply this idea in the Medical Imaging domain. We use transfer learning to train a segmentation model for the small dataset for which gold-standard class annotations are available. We extracted the learnt features and use them to generate synthetic features conditioned on class labels, using Auxiliary Classifier GAN (ACGAN). We test the quality of the generated features in a downstream classification task for brain tumors according to their severity level. Experimental results show a promising result regarding the validity of these generated features and their overall contribution to balancing the data and improving the classification class-wise accuracy.
△ Less
Submitted 28 September, 2022;
originally announced September 2022.
-
Offensive Language Detection on Twitter
Authors:
Nikhil Chilwant,
Syed Taqi Abbas Rizvi,
Hassan Soliman
Abstract:
Detection of offensive language in social media is one of the key challenges for social media. Researchers have proposed many advanced methods to accomplish this task. In this report, we try to use the learnings from their approach and incorporate our ideas to improve upon them. We have successfully achieved an accuracy of 74% in classifying offensive tweets. We also list upcoming challenges in th…
▽ More
Detection of offensive language in social media is one of the key challenges for social media. Researchers have proposed many advanced methods to accomplish this task. In this report, we try to use the learnings from their approach and incorporate our ideas to improve upon them. We have successfully achieved an accuracy of 74% in classifying offensive tweets. We also list upcoming challenges in the abusive content detection in the social media world.
△ Less
Submitted 28 September, 2022;
originally announced September 2022.
-
Effective General-Domain Data Inclusion for the Machine Translation Task by Vanilla Transformers
Authors:
Hassan Soliman
Abstract:
One of the vital breakthroughs in the history of machine translation is the development of the Transformer model. Not only it is revolutionary for various translation tasks, but also for a majority of other NLP tasks. In this paper, we aim at a Transformer-based system that is able to translate a source sentence in German to its counterpart target sentence in English. We perform the experiments on…
▽ More
One of the vital breakthroughs in the history of machine translation is the development of the Transformer model. Not only it is revolutionary for various translation tasks, but also for a majority of other NLP tasks. In this paper, we aim at a Transformer-based system that is able to translate a source sentence in German to its counterpart target sentence in English. We perform the experiments on the news commentary German-English parallel sentences from the WMT'13 dataset. In addition, we investigate the effect of the inclusion of additional general-domain data in training from the IWSLT'16 dataset to improve the Transformer model performance. We find that including the IWSLT'16 dataset in training helps achieve a gain of 2 BLEU score points on the test set of the WMT'13 dataset. Qualitative analysis is introduced to analyze how the usage of general-domain data helps improve the quality of the produced translation sentences.
△ Less
Submitted 28 September, 2022;
originally announced September 2022.
-
A Mutually Exciting Latent Space Hawkes Process Model for Continuous-time Networks
Authors:
Zhipeng Huang,
Hadeel Soliman,
Subhadeep Paul,
Kevin S. Xu
Abstract:
Networks and temporal point processes serve as fundamental building blocks for modeling complex dynamic relational data in various domains. We propose the latent space Hawkes (LSH) model, a novel generative model for continuous-time networks of relational events, using a latent space representation for nodes. We model relational events between nodes using mutually exciting Hawkes processes with ba…
▽ More
Networks and temporal point processes serve as fundamental building blocks for modeling complex dynamic relational data in various domains. We propose the latent space Hawkes (LSH) model, a novel generative model for continuous-time networks of relational events, using a latent space representation for nodes. We model relational events between nodes using mutually exciting Hawkes processes with baseline intensities dependent upon the distances between the nodes in the latent space and sender and receiver specific effects. We demonstrate that our proposed LSH model can replicate many features observed in real temporal networks including reciprocity and transitivity, while also achieving superior prediction accuracy and providing more interpretable fits than existing models.
△ Less
Submitted 6 July, 2022; v1 submitted 18 May, 2022;
originally announced May 2022.
-
The Multivariate Community Hawkes Model for Dependent Relational Events in Continuous-time Networks
Authors:
Hadeel Soliman,
Lingfei Zhao,
Zhipeng Huang,
Subhadeep Paul,
Kevin S. Xu
Abstract:
The stochastic block model (SBM) is one of the most widely used generative models for network data. Many continuous-time dynamic network models are built upon the same assumption as the SBM: edges or events between all pairs of nodes are conditionally independent given the block or community memberships, which prevents them from reproducing higher-order motifs such as triangles that are commonly o…
▽ More
The stochastic block model (SBM) is one of the most widely used generative models for network data. Many continuous-time dynamic network models are built upon the same assumption as the SBM: edges or events between all pairs of nodes are conditionally independent given the block or community memberships, which prevents them from reproducing higher-order motifs such as triangles that are commonly observed in real networks. We propose the multivariate community Hawkes (MULCH) model, an extremely flexible community-based model for continuous-time networks that introduces dependence between node pairs using structured multivariate Hawkes processes. We fit the model using a spectral clustering and likelihood-based local refinement procedure. We find that our proposed MULCH model is far more accurate than existing models both for predictive and generative tasks.
△ Less
Submitted 6 July, 2022; v1 submitted 2 May, 2022;
originally announced May 2022.
-
NIHAO XXVII: Crossing the green valley
Authors:
Marvin Blank,
Andrea V. Macciò,
Xi Kang,
Keri L. Dixon,
Nadine H. Soliman
Abstract:
The transition of high-mass galaxies from being blue and star forming to being red and dead is a crucial step in galaxy evolution, yet not fully understood. In this work, we use the NIHAO suite of galaxy simulations to investigate the relation between the transition time through the green valley and other galaxy properties. The typical green valley crossing time of our galaxies is approximately 40…
▽ More
The transition of high-mass galaxies from being blue and star forming to being red and dead is a crucial step in galaxy evolution, yet not fully understood. In this work, we use the NIHAO suite of galaxy simulations to investigate the relation between the transition time through the green valley and other galaxy properties. The typical green valley crossing time of our galaxies is approximately 400 Myr, somewhat shorter than observational estimates. The crossing of the green valley is triggered by the onset of AGN feedback and the subsequent shut down of star formation. Interestingly the time spent in the green valley is not related to any other galaxy properties, such as stellar age or metallicity, or the time at which the star formation quenching takes place. The crossing time is set by two main contributions: the ageing of the current stellar population and the residual star formation in the green valley. These effects are of comparable magnitude, while major and minor mergers have a negligible contribution. Most interestingly, we find the time that a galaxy spends to travel through the green valley is twice the $e$-folding time of the star formation quenching. This result is stable against galaxy properties and the exact numerical implementation of AGN feedback in the simulation. Assuming a typical crossing time of about one Gyr inferred from observations, our results imply that any mechanism or process aiming to quench star formation, must do it on a typical timescale of 500 Myr.
△ Less
Submitted 25 April, 2022;
originally announced April 2022.
-
A Predictive Model for Student Performance in Classrooms Using Student Interactions With an eTextbook
Authors:
Ahmed Abd Elrahman,
Taysir Hassan A Soliman,
Ahmed I. Taloba,
Mohammed F. Farghally
Abstract:
With the rise of online eTextbooks and Massive Open Online Courses (MOOCs), a huge amount of data has been collected related to students' learning. With the careful analysis of this data, educators can gain useful insights into the performance of their students and their behavior in learning a particular topic. This paper proposes a new model for predicting student performance based on an analysis…
▽ More
With the rise of online eTextbooks and Massive Open Online Courses (MOOCs), a huge amount of data has been collected related to students' learning. With the careful analysis of this data, educators can gain useful insights into the performance of their students and their behavior in learning a particular topic. This paper proposes a new model for predicting student performance based on an analysis of how students interact with an interactive online eTextbook. By being able to predict students' performance early in the course, educators can easily identify students at risk and provide a suitable intervention. We considered two main issues the prediction of good/bad performance and the prediction of the final exam grade. To build the proposed model, we evaluated the most popular classification and regression algorithms on data from a data structures and algorithms course (CS2) offered in a large public research university. Random Forest Regression and Multiple Linear Regression have been applied in Regression. While Logistic Regression, decision tree, Random Forest Classifier, K Nearest Neighbors, and Support Vector Machine have been applied in classification.
△ Less
Submitted 16 February, 2022;
originally announced March 2022.
-
Arabic aspect sentiment polarity classification using BERT
Authors:
Mohammed M. Abdelgwad,
Taysir Hassan A Soliman,
Ahmed I. Taloba
Abstract:
Aspect-based sentiment analysis(ABSA) is a textual analysis methodology that defines the polarity of opinions on certain aspects related to specific targets. The majority of research on ABSA is in English, with a small amount of work available in Arabic. Most previous Arabic research has relied on deep learning models that depend primarily on context-independent word embeddings (e.g.word2vec), whe…
▽ More
Aspect-based sentiment analysis(ABSA) is a textual analysis methodology that defines the polarity of opinions on certain aspects related to specific targets. The majority of research on ABSA is in English, with a small amount of work available in Arabic. Most previous Arabic research has relied on deep learning models that depend primarily on context-independent word embeddings (e.g.word2vec), where each word has a fixed representation independent of its context. This article explores the modeling capabilities of contextual embeddings from pre-trained language models, such as BERT, and making use of sentence pair input on Arabic aspect sentiment polarity classification task. In particular, we develop a simple but effective BERT-based neural baseline to handle this task. Our BERT architecture with a simple linear classification layer surpassed the state-of-the-art works, according to the experimental results on three different Arabic datasets. Achieving an accuracy of 89.51% on the Arabic hotel reviews dataset, 73% on the Human annotated book reviews dataset, and 85.73% on the Arabic news dataset.
△ Less
Submitted 10 March, 2023; v1 submitted 28 July, 2021;
originally announced July 2021.
-
Dust in the Wind with Resonant Drag Instabilities: I. The Dynamics of Dust-Driven Outflows in GMCs and HII Regions
Authors:
Philip F. Hopkins,
Anna L. Rosen,
Jonathan Squire,
Georgia V. Panopoulou,
Nadine H. Soliman,
Darryl Seligman,
Ulrich P. Steinwandel
Abstract:
Radiation-dust driven outflows, where radiation pressure on dust grains accelerates gas, occur in many astrophysical environments. Almost all previous numerical studies of these systems have assumed that the dust was perfectly-coupled to the gas. However, it has recently been shown that the dust in these systems is unstable to a large class of resonant drag instabilities (RDIs) which de-couple the…
▽ More
Radiation-dust driven outflows, where radiation pressure on dust grains accelerates gas, occur in many astrophysical environments. Almost all previous numerical studies of these systems have assumed that the dust was perfectly-coupled to the gas. However, it has recently been shown that the dust in these systems is unstable to a large class of resonant drag instabilities (RDIs) which de-couple the dust and gas dynamics and could qualitatively change the nonlinear outcome of these outflows. We present the first simulations of radiation-dust driven outflows in stratified, inhomogeneous media, including explicit grain dynamics and a realistic spectrum of grain sizes and charge, magnetic fields and Lorentz forces on grains (which dramatically enhance the RDIs), Coulomb and Epstein drag forces, and explicit radiation transport allowing for different grain absorption and scattering properties. In this paper we consider conditions resembling giant molecular clouds (GMCs), HII regions, and distributed starbursts, where optical depths are modest, single-scattering effects dominate radiation-dust coupling, Lorentz forces dominate over drag on grains, and the fastest-growing RDIs are similar, such as magnetosonic and fast-gyro RDIs. These RDIs generically produce strong size-dependent dust clustering, growing nonlinear on timescales that are much shorter than the characteristic times of the outflow. The instabilities produce filamentary and plume-like or 'horsehead' nebular morphologies that are remarkably similar to observed dust structures in GMCs and HII regions. Additionally, in some cases they strongly alter the magnetic field structure and topology relative to filaments. Despite driving strong micro-scale dust clum** which leaves some gas behind, an order-unity fraction of the gas is always efficiently entrained by dust.
△ Less
Submitted 8 November, 2022; v1 submitted 9 July, 2021;
originally announced July 2021.
-
Arabic aspect based sentiment analysis using bidirectional GRU based models
Authors:
Mohammed M. Abdelgwad,
Taysir Hassan A Soliman,
Ahmed I. Taloba,
Mohamed Fawzy Farghaly
Abstract:
Aspect-based Sentiment analysis (ABSA) accomplishes a fine-grained analysis that defines the aspects of a given document or sentence and the sentiments conveyed regarding each aspect. This level of analysis is the most detailed version that is capable of exploring the nuanced viewpoints of the reviews. The bulk of study in ABSA focuses on English with very little work available in Arabic. Most pre…
▽ More
Aspect-based Sentiment analysis (ABSA) accomplishes a fine-grained analysis that defines the aspects of a given document or sentence and the sentiments conveyed regarding each aspect. This level of analysis is the most detailed version that is capable of exploring the nuanced viewpoints of the reviews. The bulk of study in ABSA focuses on English with very little work available in Arabic. Most previous work in Arabic has been based on regular methods of machine learning that mainly depends on a group of rare resources and tools for analyzing and processing Arabic content such as lexicons, but the lack of those resources presents another challenge. In order to address these challenges, Deep Learning (DL)-based methods are proposed using two models based on Gated Recurrent Units (GRU) neural networks for ABSA. The first is a DL model that takes advantage of word and character representations by combining bidirectional GRU, Convolutional Neural Network (CNN), and Conditional Random Field (CRF) making up the (BGRU-CNN-CRF) model to extract the main opinionated aspects (OTE). The second is an interactive attention network based on bidirectional GRU (IAN-BGRU) to identify sentiment polarity toward extracted aspects. We evaluated our models using the benchmarked Arabic hotel reviews dataset. The results indicate that the proposed methods are better than baseline research on both tasks having 39.7% enhancement in F1-score for opinion target extraction (T2) and 7.58% in accuracy for aspect-based sentiment polarity classification (T3). Achieving F1 score of 70.67% for T2, and accuracy of 83.98% for T3.
△ Less
Submitted 6 October, 2021; v1 submitted 22 January, 2021;
originally announced January 2021.
-
RANK: AI-assisted End-to-End Architecture for Detecting Persistent Attacks in Enterprise Networks
Authors:
Hazem M. Soliman,
Geoff Salmon,
Dušan Sovilj,
Mohan Rao
Abstract:
Advanced Persistent Threats (APTs) are sophisticated multi-step attacks, planned and executed by skilled adversaries targeting modern government and enterprise networks. Intrusion Detection Systems (IDSs) and User and Entity Behavior Analytics (UEBA) are commonly employed to aid a security analyst in the detection of APTs. The prolonged nature of APTs, combined with the granular focus of UEBA and…
▽ More
Advanced Persistent Threats (APTs) are sophisticated multi-step attacks, planned and executed by skilled adversaries targeting modern government and enterprise networks. Intrusion Detection Systems (IDSs) and User and Entity Behavior Analytics (UEBA) are commonly employed to aid a security analyst in the detection of APTs. The prolonged nature of APTs, combined with the granular focus of UEBA and IDS, results in overwhelming the analyst with an increasingly impractical number of alerts. Consequent to this abundance of data, and together with the crucial importance of the problem as well as the high cost of the skilled personnel involved, the problem of APT detection becomes a perfect candidate for automation through Artificial Intelligence (AI). In this paper, we provide, up to our knowledge, the first study and implementation of an end-to-end AI-assisted architecture for detecting APTs -- RANK. The goal of the system is not to replace the analyst, rather, it is to automate the complete pipeline from data sources to a final set of incidents for analyst review. The architecture is composed of four consecutive steps: 1) alert templating and merging, 2) alert graph construction, 3) alert graph partitioning into incidents, and 4) incident scoring and ordering. We evaluate our architecture against the 2000 DARPA Intrusion Detection dataset, as well as a read-world private dataset from a medium-scale enterprise. Extensive results are provided showing a three order of magnitude reduction in the amount of data to be reviewed by the analyst, innovative extraction of incidents and security-wise scoring of extracted incidents.
△ Less
Submitted 6 January, 2021;
originally announced January 2021.
-
Objective Diagnosis for Histopathological Images Based on Machine Learning Techniques: Classical Approaches and New Trends
Authors:
Naira Elazab,
Hassan Soliman,
Shaker El-Sappagh,
S. M. Riazul Islam,
Mohammed Elmogy
Abstract:
Histopathology refers to the examination by a pathologist of biopsy samples. Histopathology images are captured by a microscope to locate, examine, and classify many diseases, such as different cancer types. They provide a detailed view of different types of diseases and their tissue status. These images are an essential resource with which to define biological compositions or analyze cell and tis…
▽ More
Histopathology refers to the examination by a pathologist of biopsy samples. Histopathology images are captured by a microscope to locate, examine, and classify many diseases, such as different cancer types. They provide a detailed view of different types of diseases and their tissue status. These images are an essential resource with which to define biological compositions or analyze cell and tissue structures. This imaging modality is very important for diagnostic applications. The analysis of histopathology images is a prolific and relevant research area supporting disease diagnosis. In this paper, the challenges of histopathology image analysis are evaluated. An extensive review of conventional and deep learning techniques which have been applied in histological image analyses is presented. This review summarizes many current datasets and highlights important challenges and constraints with recent deep learning techniques, alongside possible future research avenues. Despite the progress made in this research area so far, it is still a significant area of open research because of the variety of imaging techniques and disease-specific characteristics.
△ Less
Submitted 10 November, 2020;
originally announced November 2020.
-
A Graph Neural Network Approach for Scalable and Dynamic IP Similarity in Enterprise Networks
Authors:
Hazem M. Soliman,
Geoff Salmon,
Dusan Sovilij,
Mohan Rao
Abstract:
Measuring similarity between IP addresses is an important task in the daily operations of any enterprise network. Applications that depend on an IP similarity measure include measuring correlation between security alerts, building baselines for behavioral modelling, debugging network failures and tracking persistent attacks. However, IPs do not have a natural similarity measure by definition. Deep…
▽ More
Measuring similarity between IP addresses is an important task in the daily operations of any enterprise network. Applications that depend on an IP similarity measure include measuring correlation between security alerts, building baselines for behavioral modelling, debugging network failures and tracking persistent attacks. However, IPs do not have a natural similarity measure by definition. Deep Learning architectures are a promising solution here since they are able to learn numerical representations for IPs directly from data, allowing various distance measures to be applied on the calculated representations. Current works have utilized Natural Language Processing (NLP) techniques for learning IP embeddings. However, these approaches have no proper way to handle out-of-vocabulary (OOV) IPs not seen during training. In this paper, we propose a novel approach for IP embedding using an adapted graph neural network (GNN) architecture. This approach has the advantages of working on the raw data, scalability and, most importantly, induction, i.e. the ability to measure similarity between previously unseen IPs. Using data from an enterprise network, our approach is able to identify similarities between local DNS servers and root DNS servers even though some of these machines are never encountered during the training phase.
△ Less
Submitted 9 October, 2020;
originally announced October 2020.
-
NIHAO XXVI: Nature versus nurture, the Star Formation Main Sequence and the origin of its scatter
Authors:
Marvin Blank,
Liam E. Meier,
Andrea V. Macciò,
Aaron A. Dutton,
Keri L. Dixon,
Nadine H. Soliman,
Xi Kang
Abstract:
We investigate how the NIHAO galaxies match the observed star formation main sequence (SFMS) and what the origin of its scatter is. The NIHAO galaxies reproduce the SFMS and generally agree with observations, but the slope is about unity and thus significantly larger than observed values. This is because observed galaxies at large stellar masses, although still being part of the SFMS, are already…
▽ More
We investigate how the NIHAO galaxies match the observed star formation main sequence (SFMS) and what the origin of its scatter is. The NIHAO galaxies reproduce the SFMS and generally agree with observations, but the slope is about unity and thus significantly larger than observed values. This is because observed galaxies at large stellar masses, although still being part of the SFMS, are already influenced by quenching. This partial suppression of star formation by AGN feedback leads to lower star formation rates and therefore to lower observed slopes. We confirm that including the effects of AGN in our galaxies leads to slopes in agreement with observations. We find the deviation of a galaxy from the SFMS is correlated with its $z=0$ dark matter halo concentration and thus with its halo formation time. This means galaxies with a higher-than-average star formation rate (SFR) form later and vice versa. We explain this apparent correlation with the SFR by re-interpreting galaxies that lie above the SFMS (higher-than-average SFR) as lying to the left of the SFMS (lower-than-average stellar mass) and vice versa. Thus later forming haloes have a lower-than-average stellar mass, this is simply because they have had less-than-average time to form stars, and vice versa. It is thus the nature, i.e. how and when these galaxies form, that sets the path of a galaxy in the SFR versus stellar mass plane.
△ Less
Submitted 6 September, 2020; v1 submitted 31 August, 2020;
originally announced August 2020.
-
Low Complexity V-BLAST MIMO-OFDM Detector by Successive Iterations Reduction
Authors:
Karam Ahmed,
Sherif Abuelenin,
Heba Soliman,
Khairy Al-Barbary
Abstract:
V-BLAST detection method suffers large computational complexity due to its successive detection of symbols. In this paper, we propose a modified V-BLAST algorithm to decrease the computational complexity by reducing the number of detection iterations required in MIMO communication systems. We begin by showing the existence of a maximum number of iterations, beyond which, no significant improvement…
▽ More
V-BLAST detection method suffers large computational complexity due to its successive detection of symbols. In this paper, we propose a modified V-BLAST algorithm to decrease the computational complexity by reducing the number of detection iterations required in MIMO communication systems. We begin by showing the existence of a maximum number of iterations, beyond which, no significant improvement is obtained. We establish a criterion for the number of maximum effective iterations. We propose a modified algorithm that uses the measured SNR to dynamically set the number of iterations to achieve an acceptable bit-error rate. Then, we replace the feedback algorithm with an approximate linear function to reduce the complexity. Simulations show that significant reduction in computational complexity is achieved compared to the ordinary V-BLAST, while maintaining a good BER performance.
△ Less
Submitted 7 April, 2015;
originally announced April 2015.
-
Current Services In Cloud Computing: A Survey
Authors:
Mohamed Magdy Mosbah,
Hany Soliman,
Mohamad Abou El-Nasr
Abstract:
Due to the fast development of the Cloud Computing technologies, the rapid increase of cloud services are became very remarkable. The fact of integration of these services with many of the modern enterprises cannot be ignored. Microsoft, Google, Amazon, SalesForce.com and the other leading IT companies are entered the field of develo** these services. This paper presents a comprehensive survey o…
▽ More
Due to the fast development of the Cloud Computing technologies, the rapid increase of cloud services are became very remarkable. The fact of integration of these services with many of the modern enterprises cannot be ignored. Microsoft, Google, Amazon, SalesForce.com and the other leading IT companies are entered the field of develo** these services. This paper presents a comprehensive survey of current cloud services, which are divided into eleven categories. Also the most famous providers for these services are listed. Finally, the Deployment Models of Cloud Computing are mentioned and briefly discussed.
△ Less
Submitted 13 November, 2013;
originally announced November 2013.
-
Designing a WISHBONE Protocol Network Adapter for an Asynchronous Network-on-Chip
Authors:
Ahmed H. M. Soliman,
E. M. Saad,
M. El-Bably,
Hesham M. A. M. Keshk
Abstract:
The Scaling of microchip technologies, from micron to submicron and now to deep sub-micron (DSM) range, has enabled large scale systems-on-chip (SoC). In future deep submicron (DSM) designs, the interconnect effect will definitely dominate performance. Network-on-Chip (NoC) has become a promising solution to bus-based communication infrastructure limitations. NoC designs usually targets Applicatio…
▽ More
The Scaling of microchip technologies, from micron to submicron and now to deep sub-micron (DSM) range, has enabled large scale systems-on-chip (SoC). In future deep submicron (DSM) designs, the interconnect effect will definitely dominate performance. Network-on-Chip (NoC) has become a promising solution to bus-based communication infrastructure limitations. NoC designs usually targets Application Specific Integrated Circuits (ASICs), however, the fabrication process costs a lot. Implementing a NoC on an FPGA does not only reduce the cost but also decreases programming and verification cycles. In this paper, an Asynchronous NoC has been implemented on a SPARTAN-3E\textregistered device. The NoC supports basic transactions of both widely used on-chip interconnection standards, the Open Core Protocol (OCP) and the WISHBONE Protocol. Although, FPGA devices are synchronous in nature, it has been shown that they can be used to prototype a Global Asynchronous Local Synchronous (GALS) systems, comprising an Asynchronous NoC connecting IP cores operating in different clock domains.
△ Less
Submitted 19 March, 2012;
originally announced March 2012.
-
Fair Allocation of Backhaul Resources in Multi-Cell MIMO Co-operative Networks
Authors:
Hazem M. Soliman,
Omar A. Nasr,
Mohamed M. Khairy
Abstract:
In this paper the problem of allocating the limited backhaul bandwidth among users in Multi-cell MIMO cooperative networks is considered. We approach the problem from both the sum-rate and fairness perspectives. First, we show that there are many allocations that can provide near maximum sumrate while varying significantly in fairness, which is assessed through the mean versus variance criteria. S…
▽ More
In this paper the problem of allocating the limited backhaul bandwidth among users in Multi-cell MIMO cooperative networks is considered. We approach the problem from both the sum-rate and fairness perspectives. First, we show that there are many allocations that can provide near maximum sumrate while varying significantly in fairness, which is assessed through the mean versus variance criteria. Second, Two novel schemes that achieve near maximum sum-rate while offering fair allocation of the backhaul bandwidth among users are proposed: the Equal Signal-to-Interference ratio (SIR) and the Equal Interference schemes. Simulation results show that, for the same mean rate among users, the proposed schemes can achieve more fairness when compared to the conventional scheme, which gives all users the same share of bandwidth. Moreover, we show that the Equal SIR scheme can achieve zero variance among users in a wide range of backhaul bandwidths while kee** very close to maximum sum rate. This is the most fair solution that can be used in Multi-cell MIMO in that range of backhaul bandwidths.
△ Less
Submitted 12 June, 2011;
originally announced June 2011.
-
Interaction Dependence Thermodynamical parameters of Harmonically Trapped Bose gas
Authors:
Ahmed S. Hassan,
Shemi S. M. Soliman,
Emad H. Soliman
Abstract:
In this paper the thermodynamical parameters of a condensed Boson gas are calculated from the partial derivative of the grand potential. In particular, the analytical expressions for some important parameters, such as the condensed fraction, specific heat, critical temperature, effective size, and release energy are investigated. The mean effects which can be altered the ideal Bose gas, such as…
▽ More
In this paper the thermodynamical parameters of a condensed Boson gas are calculated from the partial derivative of the grand potential. In particular, the analytical expressions for some important parameters, such as the condensed fraction, specific heat, critical temperature, effective size, and release energy are investigated. The mean effects which can be altered the ideal Bose gas, such as finite size, highly anisotropic of the external potential and interatomic interaction effects are considered simultaneously. Some new characteristics of the trapped interacting Bose gases in a highly anisotropic trap are revealed. The calculated result for the condensed fraction is compared with the ongoing Stuttgart experiment for 52Cr (Griesmaier et al., Phys. Rev. Lett. 94, 160401(2005)) directly. Good agreement between both the theoretical and experimental data are obtained.
△ Less
Submitted 25 September, 2009;
originally announced September 2009.
