E-mail spam continues to become a problem on the Internet. Spammed e-mail may contain many copies of the same message, commercial advertisement or other irrelevant posts like pornographic content. In previous research, different filtering techniques are used to detect these e-mails such as using Random Forest, Naïve Bayesian, Support Vector Machine (SVM) and Neutral Network. In this research, we test Naïve Bayes algorithm for e-mail spam filtering on two datasets and test its performance, i.e., Spam Data and SPAMBASE datasets [8]. The performance of the datasets is evaluated based on their accuracy, recall, precision and F-measure. Our research use WEKA tool for the evaluation of Naïve Bayes algorithm for e-mail spam filtering on both datasets. The result shows that the type of email and the number of instances of the dataset has an influence towards the performance of Naïve Bayes.

The rapid growth of digital content in Urdu has created an urgent need for effective automatic text summarization (ATS) systems. While extractive methods have been widely studied, abstractive summarization for Urdu remains largely unexplored due to the language’s complex morphology and rich literary tradition. This paper systematically evaluates four transformer-based language models (BERT-Urdu, BART, mT5, and GPT-2) for Urdu abstractive summarization, comparing their performance against conventional machine learning and deep learning approaches. Using multiple Urdu datasets—including the Urdu Summarization Corpus, Fake News Dataset, and Urdu-Instruct-News—we show that fine-tuned Transformer Language Models (TLMs) consistently outperform traditional methods, with the multilingual mT5 model achieving a 0.42 absolute improvement in F1-score over the best baseline. Our analysis reveals that mT5’s architecture is particularly effective at handling Urdu-specific challenges such as right-to-left script processing, diacritic interpretation, and complex verb–noun compounding. Furthermore, we present empirically validated hyperparameter configurations and training strategies for Urdu ATS, establishing transformer-based approaches as the new state-of-the-art for Urdu summarization. Notably, mT5 outperforms Seq2Seq baselines by up to 20% in ROUGE-L, underscoring the efficacy of Transformer-based models for low-resource languages. This work contributes both a systematic review of prior research and a novel empirical benchmark for advancing Urdu abstractive summarization.

Carbon sequestration through forestry represents a promising approach to partially counteract anthropogenic greenhouse gas emissions driving climate change. Tree growth naturally removes carbon dioxide from the atmosphere, storing it as biomass. Sustainably managed forests can effectively function as carbon sinks. However, determining optimal forestry policies involves balancing complex ecological dynamics with economic constraints. This study develops differential equation models quantitatively capturing forest growth, timber harvesting, and carbon sequestration dynamics. Logistic models are first adapted to simulate biomass accumulation of representative tree species. Lifecycle growth patterns spanning juvenile to mature phases are incorporated, along with climate effects. Biomass levels are proportionally related to carbon dioxide removal rates from the atmosphere. Deforestation impacts are analyzed by incorporating harvesting-induced biomass reductions. Sustainability constraints are implemented to ensure minimum viable tree densities across harvest rotations. Optimization techniques then identify management guidelines maximizing economic returns given ecological stability considerations. The goal is providing quantitative insights into rotation lengths, planting densities, and allowable cuts upholding both climate change mitigation and commercial demands. Findings can inform science-based forestry policies to leverage forests as sustainable natural carbon sinks.