MDPI - Publisher of Open Access Journals

11 pages, 647 KiB

Open AccessArticle

Transformer Architecture-Based Transfer Learning for Politeness Prediction in Conversation

by Shakir Khan, Mohd Fazil, Agbotiname Lucky Imoize, Bayan Ibrahimm Alabduallah, Bader M. Albahlal, Saad Abdullah Alajlan, Abrar Almjally and Tamanna Siddiqui

Sustainability 2023, 15(14), 10828; https://doi.org/10.3390/su151410828 - 10 Jul 2023

Cited by 18 | Viewed by 2937

Abstract

Politeness is an essential part of a conversation. Like verbal communication, politeness in textual conversation and social media posts is also stimulating. Therefore, the automatic detection of politeness is a significant and relevant problem. The existing literature generally employs classical machine learning-based models like naive Bayes and Support Vector-based trained models for politeness prediction. This paper exploits the state-of-the-art (SOTA) transformer architecture and transfer learning for respectability prediction. The proposed model employs the strengths of context-incorporating large language models, a feed-forward neural network, and an attention mechanism for representation learning of natural language requests. The trained representation is further classified using a softmax function into polite, impolite, and neutral classes. We evaluate the presented model employing two SOTA pre-trained large language models on two benchmark datasets. Our model outperformed the two SOTA and six baseline models, including two domain-specific transformer-based models using both the BERT and RoBERTa language models. The ablation investigation shows that the exclusion of the feed-forward layer displays the highest impact on the presented model. The analysis reveals the batch size and optimization algorithms as effective parameters affecting the model performance. Full article

(This article belongs to the Special Issue Machine Learning, IoT and Artificial Intelligence for Sustainable Development)

► Show Figures

Figure 1

17 pages, 6503 KiB

Open AccessFeature PaperArticle

Hydrothermally Derived Mg-Doped TiO₂ Nanostructures for Enhanced H₂ Evolution Using Photo- and Electro-Catalytic Water Splitting

by Mohd Fazil, Saad M. Alshehri, Yuanbing Mao and Tokeer Ahmad

Catalysts 2023, 13(5), 893; https://doi.org/10.3390/catal13050893 - 15 May 2023

Cited by 15 | Viewed by 3929

Abstract

Mg-doped TiO₂ nano-structures in different compositions (1, 2.5 and 5%) were successfully synthesized by low-temperature hydrothermal route. X-ray diffraction and electron microscopic studies were used to investigate the crystal structure, surface morphology and particle size of the as-synthesized materials. Raman studies were carried out to elucidate the phase identification and the modes of vibrations to determine the impact of dopant ion on the crystal structures. The band gap was estimated using UV-DRS studies whereas, BET surface area analysis revealed an increase in the surface area of increasing Mg²⁺ ions concentration in TiO₂ nanostructures. Among the synthesized various composition of nano-structures, 5% Mg-doped TiO₂ photocatalyst showed maximum hydrogen evolution activity (38.96 mmol

g_{cat}^{- 1}

) in an 8-h analysis cycle. Moreover, the 2.5% Mg-doped TiO₂ nanocatalyst with tafel slopes of 123.5 and 126.7 mV/dec showed strong activity for both HER in 0.5 N H₂SO₄ and 0.1 N KOH, with an onset potential of 0.96 V (at 10 mA/cm²) and −1.38 V (at 1 mA/cm²) for HER, respectively. Experimental investigations deduced that the incorporation of Mg²⁺ ions in the TiO₂ resulted in the increase of hydrogen generation catalytic activity of titanium dioxide owing to the synergistic effect provided by the remarkable surface area and the presence of defects introduced by doping. Full article

(This article belongs to the Special Issue Advanced Nanostructured Materials for Modern Catalysis Applications)

► Show Figures

Graphical abstract

21 pages, 8264 KiB

Open AccessArticle

Pristine TiO₂ and Sr-Doped TiO₂ Nanostructures for Enhanced Photocatalytic and Electrocatalytic Water Splitting Applications

by Mohd Fazil and Tokeer Ahmad

Catalysts 2023, 13(1), 93; https://doi.org/10.3390/catal13010093 - 2 Jan 2023

Cited by 32 | Viewed by 4711

Abstract

Pristine TiO₂ and Sr-doped TiO₂ (1%, 2.5% and 5%) nanoparticles were synthesized at low temperatures via an eco-friendly hydrothermal route for water-splitting applications. XRD, EDAX and Raman analysis were performed to analyze the crystallinity, purity and structure of the as-synthesized materials. TEM, SEM, BET and UV-DRS studies were carried out to elucidate the size, morphology, surface area and optoelectronic properties of the nanoparticles. High surface areas of 169, 182, 178 and 141.16 m² g⁻¹ for pristine TiO₂ (12 ± 0.6 nm) and 1% (11.1 ± 0.6 nm), 2.5% (12.1 ± 0.6 nm) and 5% (13 ± 0.7 nm) Sr-doped TiO₂ nanoparticles were obtained, respectively. One-percent Sr-doped TiO₂ nanoparticles were found to be active photocatalysts, as they showed higher hydrogen production (26.30 mmol

g_{c a t}^{- 1}

). Furthermore, electrocatalysis was investigated for HER and OER in 0.5 N H₂SO₄ and 0.1 N KOH electrolytic solutions using calomel as a reference electrode, revealing that 1% and 5% Sr-doped TiO₂ showed maximum current density for both HER (≈10 mA/cm²) and OER (≈2.49 mA/cm²), with an onset potential of 0.96 V for HER and 1.55 V for OER, and Tafel slopes of 84.09 and 91.60 mV/dec, respectively. Full article

(This article belongs to the Special Issue Advances in Photoassisted and Photocatalytic Processes for Water Remediation)

► Show Figures

Graphical abstract

Search Results (3)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI

Saved Queries

Search Filter Reset All

Years

Feature Papers

Subjects

Journals

Article Types

Countries / Regions

Search Results (3)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI