Search

In the internet era, almost every business entity is trying to have its digital footprint in digital media and other social media platforms. For these entities, word of mouse is also very important. Particularly, this is quite crucial for the hospitality sector dealing with hotels, restaurants etc. Consumers do read other consumers reviews before making final decisions. This is where it becomes very important to understand which aspects are affecting most in the minds of the consumers while giving their ratings. The current study focuses on the consumer reviews of Indian hotels to extract aspects important for final ratings. The study involves gathering data using web scraping methods, analyzing the texts using Latent Dirichlet Allocation for topic extraction and sentiment analysis for aspect-specific sentiment mapping. Finally, it incorporates Random Forest to understand the importance of the aspects in predicting the final rating of a user., Comment: This is the pre-print of the paper that was accepted for oral presentation and publication in the proceedings of IEEE ICCCNT 2024 which was organized as IIT Mandi, India from June 24 to 28, 2024. The paper is 5 pages long and it contains 4 figures and 6 tables. The is not the final version of the paper

The Mg II k \& h line intensity ratios can be used to probe the characteristics of the plasma in the solar atmosphere. In this study, using the observations recorded by the Interface Region Imaging Spectrometer (IRIS), we study the variation of the Mg II k \& h intensity ratio for three flares belonging to X-class, M-class, and C-class, throughout their evolution. We also study the k-to-h intensity ratio as a function of magnetic flux density obtained from the line-of-sight magnetograms recorded by the Helioseismic and Magnetic Imager (HMI) on board the Solar Dynamics Observatory (SDO). Our results reveal that while the intensity ratios are independent of magnetic flux density, they show significant changes during the evolution of the C-class and M-class flares. The intensity ratios start to increase at the start of the flare and peak during the impulsive phase before the flare peak and decrease rapidly thereafter. The values of the ratios fall even below the pre-flare level during the peak and decline phases of the flare. These results are important in light of the heating and cooling of localized plasma and provide further constraints on the understanding of flare physics., Comment: Accepted for publication in ApJ

This paper proposes a simple but highly efficient expansion-based model for continual learning. The recent feature transformation, masking and factorization-based methods are efficient, but they grow the model only over the global or shared parameter. Therefore, these approaches do not fully utilize the previously learned information because the same task-specific parameter forgets the earlier knowledge. Thus, these approaches show limited transfer learning ability. Moreover, most of these models have constant parameter growth for all tasks, irrespective of the task complexity. Our work proposes a simple filter and channel expansion based method that grows the model over the previous task parameters and not just over the global parameter. Therefore, it fully utilizes all the previously learned information without forgetting, which results in better knowledge transfer. The growth rate in our proposed model is a function of task complexity; therefore for a simple task, the model has a smaller parameter growth while for complex tasks, the model requires more parameters to adapt to the current task. Recent expansion based models show promising results for task incremental learning (TIL). However, for class incremental learning (CIL), prediction of task id is a crucial challenge; hence, their results degrade rapidly as the number of tasks increase. In this work, we propose a robust task prediction method that leverages entropy weighted data augmentations and the models gradient using pseudo labels. We evaluate our model on various datasets and architectures in the TIL, CIL and generative continual learning settings. The proposed approach shows state-of-the-art results in all these settings. Our extensive ablation studies show the efficacy of the proposed components., Comment: To be Appeared in WACV, 2024

The thermal Sunyaev-Zeldovich (SZ) effect serves as a direct potential probe of the energetic outflows from quasars that are responsible for heating the intergalactic medium. In this work, we use the GIZMO meshless finite mass hydrodynamic cosmological simulation SIMBA (Dave et al. 2019), which includes different prescriptions for quasar feedback, to compute the SZ effect arising from different feedback modes. From these theoretical simulations, we perform mock observations of the Atacama Large Millimeter Array (ALMA) in four bands (320 GHz, 135 GHZ, 100 GHz and 42 GHz) to characterize the feasibility of direct detection of the quasar SZ signal. Our results show that for all the systems we get an enhancement of the SZ signal, when there is radiative feedback, while the signal gets suppressed when the jet mode of feedback is introduced in the simulations. Our mock ALMA maps reveal that, with the current prescription of jet feedback, the signal goes below the detection threshold of ALMA. We also find that the signal is higher for high redshift systems, making it possible for ALMA and cross SZ-X-ray studies to disentangle the varying modes of quasar feedback and their relative importance in the cosmological context., Comment: Accepted for publication in ApJ