Search

The purpose of the Bangla grammatical error correction task is to spontaneously identify and correct syntactic, morphological, semantic, and punctuation mistakes in written Bangla text using computational models, ultimately enhancing language precision and eloquence. The significance of the task encompasses bolstering linguistic acumen, fostering efficacious communication, and ensuring utmost lucidity and meticulousness in written expression, thereby mitigating the potential for obfuscation or dissemination of fallacious connotations. Prior endeavors have centered around surmounting the constraints inherent in rule-based and statistical methods through the exploration of machine learning and deep learning methods, aiming to enhance accuracy by apprehending intricate linguistic patterns, comprehending contextual cues, and discerning semantic nuances. In this study, we address the absence of a baseline for the task by developing a large-scale parallel corpus comprising 7.7M source-target pairs and exploring the untapped potential of transformers. Alongside the corpus, we introduce a Vaswani-style efficient monolingual transformer-based method named Bangla grammatical error corrector, Panini by leveraging transfer learning, which has become the state-of-the-art method for the task by surpassing the performance of both BanglaT5 and T5-Small by 18.81% and 23.8% of accuracy scores, and 11.5 and 15.6 of SacreBLEU scores, respectively. The empirical findings of the method substantiate its superiority over other approaches when it comes to capturing intricate linguistic rules and patterns. Moreover, the efficacy of our proposed method has been compared with the Bangla paraphrase task, showcasing its superior capability by outperforming the previous state-of-the-art method for the task as well. The BanglaGEC corpus and Panini, along with the baselines of BGEC and the Bangla paraphrase task, have been made publicly accessible at https://tinyurl.com/BanglaGEC. [ABSTRACT FROM AUTHOR]

Spelling error correction is the task of identifying and rectifying misspelled words in texts. It is a potential and active research topic in Natural Language Processing because of numerous applications in human language understanding. The phonetically or visually similar yet semantically distinct characters make it an arduous task in any language. Earlier efforts on spelling error correction in Bangla and resource-scarce Indic languages focused on rule-based, statistical, and machine learning-based methods which we found rather inefficient. In particular, machine learning-based approaches, which exhibit superior performance to rule-based and statistical methods, are ineffective as they correct each character regardless of its appropriateness. In this paper, we propose a novel detector-purificator-corrector framework, DPCSpell based on denoising transformers by addressing previous issues. In addition to that, we present a method for large-scale corpus creation from scratch which in turn resolves the resource limitation problem of any left-to-right scripted language. The empirical outcomes demonstrate the effectiveness of our approach, which outperforms previous state-of-the-art methods by attaining an exact match (EM) score of 94.78%, a precision score of 0.9487, a recall score of 0.9478, an f1 score of 0.948, an f0.5 score of 0.9483, and a modified accuracy (MA) score of 95.16% for Bangla spelling error correction. The models and corpus are publicly available at https://tinyurl.com/DPCSpell. • A transformer based detection-correction framework. • Augmentation technique for generating datasets. • Works for multiple language spell checking. [ABSTRACT FROM AUTHOR]

Classical methods of medical image super-resolution (MISR) utilize open-loop architecture with implicit under-resolution (UR) unit and explicit super-resolution (SR) unit. The UR unit can always be given, assumed, or estimated, while the SR unit is elaborately designed according to various SR algorithms. The closed-loop feedback mechanism is widely employed in current MISR approaches and can efficiently improve their performance. The feedback mechanism may be divided into two categories: local feedback and global feedback. Therefore, this paper proposes a global feedback-based closed-cycle framework, circular MISR (CMISR), with unambiguous UR and advanced SR elements. Mathematical model and closed-loop equation of CMISR are built. Mathematical proof with Taylor-series approximation indicates that CMISR has zero recovery error in steady-state. In addition, CMISR holds plug-and-play characteristic that fuses model-based and learning-based approaches and can be established on any existing MISR algorithms. Five CMISR algorithms are respectively proposed based on the state-of-the-art open-loop MISR algorithms. Experimental results with three scale factors and on three open medical image datasets show that CMISR is superior to MISR in reconstruction performance and is particularly suited to medical images with strong edges or intense contrast. [ABSTRACT FROM AUTHOR]

Classical methods of medical image super-resolution (MISR) utilize open-loop architecture with implicit under-resolution (UR) unit and explicit super-resolution (SR) unit. The UR unit can always be given, assumed, or estimated, while the SR unit is elaborately designed according to various SR algorithms. The closed-loop feedback mechanism is widely employed in current MISR approaches and can efficiently improve their performance. The feedback mechanism may be divided into two categories: local feedback and global feedback. Therefore, this paper proposes a global feedback-based closed-cycle framework, circular MISR (CMISR), with unambiguous UR and advanced SR elements. Mathematical model and closed-loop equation of CMISR are built. Mathematical proof with Taylor-series approximation indicates that CMISR has zero recovery error in steady-state. In addition, CMISR holds plug-and-play characteristic that fuses model-based and learning-based approaches and can be established on any existing MISR algorithms. Five CMISR algorithms are respectively proposed based on the state-of-the-art open-loop MISR algorithms. Experimental results with three scale factors and on three open medical image datasets show that CMISR is superior to MISR in reconstruction performance and is particularly suited to medical images with strong edges or intense contrast. [ABSTRACT FROM AUTHOR]

The occurrence of events affecting project performance cannot likely be eliminated, and hence their early identification and communication become critical to efficiently mitigate their impact. This research devised a taxonomy for the classification of events that can influence a project with 36 event types grouped in 10 event categories. The frequency of occurrence of events, and their impact on cost, schedule, and predictability performances were statistically determined based on information gathered from 135 recently completed projects. The statistical results demonstrate and quantify the long-held perception that the frequency of an event is often inversely proportional to its impact on project performance. In addition, qualitative evidence collected from subject-matter experts was leveraged to provide complementary insights and explanations beyond the statistical results. The results of this study are of interest to both practitioners and academics. [ABSTRACT FROM AUTHOR]

With the emergence of multidrug-resistant ‘superbug’, conventional treatments become obsolete. Quorum quenching (QQ), enzyme-dependent alteration of quorum sensing (QS), is now considered as a promising antimicrobial therapy because of its potentiality to impede virulence gene expression without resulting in growth inhibition and antibiotic resistance. In our study, we intended to compare between two major QQ enzyme groups (i.e., AHL lactonases and AHL acylases) in terms of their structural and functional aspects. The amino acid composition-based principal component analysis (PCA) suggested that probably there is no structural and functional overlapping between the two groups of enzymes as well as within the lactonase enzymes but the acylases may functionally be affected by one another. In subcellular localization analysis, we also found that most lactonases are cytoplasmic while acylases are periplasmic. Investigation on the secondary structural features showed random coil dominates over α-helix and β-sheet in all evaluated enzymes. For structural comparison, the tertiary structures of the selected proteins were modelled and submitted to the PMDB database (Accession ID: PM0081007 to PM0081018). Interestingly, sequence alignment revealed the presence of several conserved domains important for functions in both protein groups. In addition, three amino acid residues, namely aspartic acid, histidine, and isoleucine, were common in the active sites of all protein models while most frequent ligands were found to be 3C7, FEO, and PAC. Importantly, binding interactions of predicted ligands were similar to that of native QS signal molecules. Furthermore, hydrogen bonds analysis suggested six proteins are more stable than others. We believe that the knowledge of this comparative study could be useful for further research in the development of QS-based universal antibacterial strategies. [ABSTRACT FROM AUTHOR]

The energy sector is the main contributor to GHG emissions in Saudi Arabia. The tremendous growth of GHG emissions poses serious challenges for the Kingdom in terms of their reduction targets, and also the mitigation of the associated climate changes. The rising trend of population and urbanization affects the energy demand, which results in a faster rate of increase in GHG emissions. The major energy sector sources that contribute to GHG emissions include the electricity generation, road transport, desalination plants, petroleum refining, petrochemical, cement, iron and steel, and fertilizer industries. In recent years, the energy sector has become the major source, accounting for more than 90% of national CO2emissions. Although a substantial amount of research has been conducted on renewable energy resources, a sustainable shift from petroleum resources is yet to be achieved. Public awareness, access to energy-efficient technology, and the development and implementation of a legislative framework, energy pricing policies, and renewable and alternative energy policies are not mature enough to ensure a significant reduction in GHG emissions from the energy sector. An innovative and integrated solution that best serves the Kingdom's long-term needs and exploits potential indigenous, renewable, and alternative energy resources while maintaining its sustainable development stride is essential. Policy relevance The main contributor to GHG emissions in Saudi Arabia is the energy sector that accounts for more than 90% of the national CO2emissions. Tremendous growth of GHG emissions poses serious challenges for the Kingdom in their reduction and mitigating the associated climate changes. This study examines the changing patterns of different activities associated with energy sector, the pertinent challenges, and the opportunities that promise reduction of GHG emissions while providing national energy and economic security. The importance of achieving timely, sustained, and increasing reductions in GHG emissions means that a combination of policies may be needed. This study points to the long-term importance of making near- and medium-term policy choices on a well-informed, strategic basis. This analytical paper is expected to provide useful information to the national policy makers and other decision makers. It may also contribute to the GHG emission inventories and the climate change negotiations. [ABSTRACT FROM AUTHOR]