Textual emotion detection aims to discover human emotions from written text. Textual emotion detection is a significant challenge due to the unavailability of facial and voice expressions. Considerable research has been done to identify textual emotions in high-resource languages such as English, French, Chinese, and others. Despite having over 300 million speakers and large volumes of literature available online, Urdu has not been properly investigated for the textual emotion detection task. To address this gap, this study makes two contributions: (1) the creation of a novel dialog-based corpus for Urdu (Contextual Urdu Text Emotion Detection Corpus). CUTEC contains 30,160 training and 5,509 testing labelled dialogues, where each dialogue consists of three Urdu contextual sentences. In addition, all dialogues are labelled using four emotion classes, i.e., Happy, Sad, Angry, and Other. As a second contribution (2) five deep learning models, i.e., RNN, LSTM, Bi- LSTM, GRU, and Bi-GRU have been trained and tested using CUTEC with different parametric settings. The highest results (Accuracy = 87.28 and F1 = 0.87) are attained using a GRU-based architecture.

The variety of e-commerce and online business environments makes spotting fraud or fake reviews challenging. The goal of this research is to create methods that could ensure the authenticity of reviews. Numerous endeavors have already been made by the research community to address this problem. Unfortunately, these efforts were confined to a specific number of languages, such as English, Arabic, and a few others. Roman Urdu is being used across the subcontinent. For this task, this language hasn't been properly investigated. On the other hand, Transformer-based models have demonstrated tremendous success over the past few years for a variety of Natural Language Processing applications. Transformer-based models, however, have not been investigated for the goal of examining fake Roman Urdu reviews. This study compares multiple transfer-based models, including BERT_base and BERT_large with five deep learning models Simple RNN, LSTM, Bidirectional LSTM, GRU, and Bidirectional GRU aims to close this research gap. The evaluation was conducted using frequently applied metrics, i.e., ROC-Accuracy, F_1, Precision, and Recall. The BERT_large with Bi-LSTM model produced the best results (ROC-ACC=0.94 and F_1=0.94). Results from a previous study using the same corpus were improved by 6% in this investigation.

The prime objective of word sense disambiguation (WSD) is to develop such machines that can automatically recognize the actual meaning (sense) of ambiguous words in a sentence. WSD can improve various NLP and HCI challenges. Researchers explored a wide variety of methods to resolve this issue of sense ambiguity. However, majorly, their focus was on English and some other well-reputed languages. Urdu with more than 300 million users and a large amount of electronic text available on the web is still unexplored. In recent years, for a variety of Natural Language Processing tasks, word embedding methods have proven extremely successful. This study evaluates, compares, and applies a variety of word embedding approaches to Urdu Word embedding (both Lexical Sample and All-Words), including pre-trained (Word2Vec, Glove, and FastText) as well as custom-trained (Word2Vec, Glove, and FastText trained on the Ur-Mono corpus). Two benchmark corpora are used for the evaluation in this study: (1) the UAW-WSD-18 corpus and (2) the ULS-WSD-18 corpus. For Urdu All-Words WSD tasks, top results have been achieved (Accuracy=60.07 and F1=0.45) using pre-trained FastText. For the Lexical Sample, WSD has been achieved (Accuracy=70.93 and F1=0.60) using custom-trained GloVe word embedding method.

Cyberbullying on social media isn't a monolithic problem; it strikes people from all sides, including age, gender, ethnicity, and religion. As a result, moderation workflows clearly need fine-grained detection. Using the cyberbullying_tweets dataset (47,692 samples in stratified 70/15/15 splits), this paper benchmarks transformer-based learning models in the space of 6-class cyberbullying detection. The comparison examines two different paths. There's supervised fine-tuning via a QLoRA-style recipe with LoRA adapters and 4-bit NF4 quantization. Second is few-shot in-context learning based on chain-of-thought prompting and 12 examples per class (72 overall). This was done in eight instruction-tuned LLMs, including Qwen2.5 7B/8B, DeepSeek 7B/8B, Llama-3 7B/8B, Mistral-7B, and Mixtral-8x7B. Macro-F1 scores under supervised fine-tuning ranged from 0.809 to 0.857, which are identical to the accuracy range. The excellent model Mixtral-8x7B-Instruct reached macro-F1 0.857, with MCC 0.808, AUPRC 0.911, and a small calibration error (ECE 0.021). Few-shot prompting wasn't far behind, with macro-F1 between 0.770 and 0.820. The difference is usually only 0.03-0.04 compared to fine-tuning, but it significantly reduces training compute. In general, these results assess the trade-off between cost and performance and establish a reproducible baseline for transformer-based moderation.

Breast cancer is a prominent contributor to female deaths due to its prevalence, nearly twice as high as that of other cancers combined. Early detection of this disease is key to survival. This research examines the performance of the Deep Learning (DL) models such as RNN, LSTM and GRU along with BERT based models such as BERT-Base and BERT-Large in breast cancer detection application on structured medical datasets. The outcomes of the experiments showed that the model with 1-Layer RNN achieved the highest performance in terms of accuracy (97.36%), precision (97.56%), recall (95.24%), F1 score (96.38%) and AUC (99.21%), while the model with 3-Layer RNN achieved the highest performance in terms of AUC (99.40%). The transformer-based models performed at 92.11% accuracy and 98.88% AUC for BERT-Base (Seed 21) while BERT-Large had inconsistent performance due to its higher complexity. The results show that the light-weight DL models are more reliable and efficient in a structured breast cancer data set.

Topic modeling is a commonly used text-mining tool for discovering hidden semantic structures within a text corpus. This paper introduces an unsupervised learning-based topic modeling approach for Urdu documents, a language with limited resources. By leveraging unsupervised learning techniques such as Latent Dirichlet Allocation (LDA) and Unsupervised Latent Semantic Indexing (ULSI), specific and accurate topics are extracted from Urdu texts. The experimental results illustrate the dominance of our recommended ULSI model and LDA model, achieving 99% and 98% accuracy, and 44% and 37% coherence values in LDA and ULSI, respectively. The experimental results demonstrate the superiority of the proposed ULSI and LDA models, which achieve high accuracy and coherence values.