Muhammad Babar

Intelligent transportation systems (ITS) bring down several advantages, like traffic management, congestion control, road safety, economic and environmental benefits, and several others. However, ITS faces several challenges, such as data collection and integration, scalability, performance, and privacy concerns. This study proposes a privacy-preserving framework for ITSs, and a secure hybrid DBSCAN algorithm has been introduced in a trusted execution environment (TEE) to improve the performance and privacy in ITS. The algorithm combines the features of hierarchical, streaming, and distributed DBSCAN variants that generate clusters in real time for the identification of congestion zones, abnormal events, and incident hotspots with high granularity, while protecting sensitive data through secure enclave processing. A simulation testbed is deployed using SUMO and NS-3. The hybrid Density-Based Spatial Clustering of Applications with Noise (DBSCAN) outperforms standard models, achieving 97.5 to 99.1% congestion detection accuracy and 96.2 to 98.3% incident detection accuracy while handling low-to-severe traffic conditions. Along with noteworthy reductions in edge processing latency of up to 68% and network overhead lowered by 75%, witnessed in severe traffic congestion in a simulated 2 km × 2 km urban testbed. The framework contributes to integrating secure edge computing mechanisms with intelligent traffic management systems. It offers a promising framework for future smart city deployments, evaluated through high-fidelity co-simulation and laboratory-scale validation on commercial edge hardware.

There is diminished transparency, fragmented information exchange, and lack of trust among geographically dispersed stakeholders, which increasingly challenge global supply chains. The classic centralized systems of supply chain management are not always capable of being able to offer real-time traceability and data integrity which is dependable and effective in contract enforcement. The proposed study is a blockchain-based smart contract design that is focused on ensuring increased transparency, traceability and trust in global supply chain management. The suggested framework will combine automated smart contracts, cryptographic provenance tracking, permissioned blockchain consensus, and a decentralized trust score evaluation mechanism to overcome some of the major operation and governance challenges. A simulated assessment with a multi-tier global supply chain setting of 15 blockchain nodes and 12,000 transactions was performed through experimentation. The findings show that the proposed system attained an average transaction delay of 210 ms, which is very low compared to centralized systems (520 ms), with throughput being raised to 120 transactions per minute. End-to-end traceability performance also improved significantly, with a reduction in trace-back time to 8 s compared with 95s this represents a 100% tampering detection rate. The consensus mechanism ensured that the ledger integrity failed only at a rate of less than 1.1%, even when more than 30% of nodes were faulty. Risk-wise, the trust evaluation algorithm dynamically enhanced reliable supplier scores up to 12%, which facilitated the selection of reliable partners. On the whole, the results prove that smart contracts based on blockchains can drastically enhance the efficiency of operations, data integrity, and confidence in global supply chains, with the platform capable of providing a resilient and scalable backbone for the future supply chain management model.

Natural language processing is the technology used to interact with computers using human languages. An overlapping technology is Information Retrieval (IR), in which a user searches for the demanded or required documents from among a number of documents that are already stored. The required document is retrieved according to the relevance of the query of the user, and the results are presented in descending order. Many of the languages have their own IR systems, whereas a dedicated IR system for Sindhi still needs attention. Various approaches to effective information retrieval have been proposed. As Sindhi is an old language with a rich history and literature, it needs IR. For the development of Sindhi IR, a document database is required so that the documents can be retrieved accordingly. Many Sindhi documents were identified and collected from various sources, such as books, journal, magazines, and newspapers. These documents were identified as having potential for use in indexing and other forms of processing. Probabilistic modeling and pattern discovery were used to find patterns and for effective retrieval and relevancy. The results for Sindhi Information Retrieval systems are promising and presented more than 90% relevancy. The time elapsed was recorded as ranging from 0.2 to 4.8 s for a single word and 4.6 s with a Sindhi sentence, with the same starting time of 0.2 s. The IR system for Sindhi can be fine-tuned and utilized for other languages with the same characteristics, which adopt Arabic script.

Knowledge discovery helps mitigate the shortcomings of classical machine learning, especially those so-called imbalanced, high-dimensional, and noisy data challenges. Adaptive combination of multiple models, voting and other data fusion strategies, and the incorporation of other disparate information fusion methods characterize ensemble learning, which addresses the improvement of a predictive model’s accuracy, stability, and generalization. This paper provides a summary of the important approaches to ensemble learning and their real-world uses, emphasizing challenges and opportunities for future work. This paper also discusses how ensemble learning integrates with emergent areas such as deep learning and reinforcement learning. This paper also describes the most important machine learning methods for predicting heart disease, which include decision trees, support vector machines, artificial neural networks, Naïve Bayes, random forest, and K-nearest neighbors.

GraphRAG-Causal introduces an innovative framework that combines graph-based retrieval with large language models to enhance causal reasoning in news analysis. Traditional NLP approaches often struggle with identifying complex, implicit causal links, especially in low-data scenarios. Our approach addresses these challenges by transforming annotated news headlines into structured causal knowledge graphs. It then employs a hybrid retrieval system that merges semantic embeddings with graph-based structural cues leveraging Neo4j to accurately match and retrieve relevant events. The framework is built on a three-stage pipeline: First, during Data Preparation, news sentences are meticulously annotated and converted into causal graphs capturing cause, effect, and trigger relationships. Next, the Graph Retrieval stage stores these graphs along with their embeddings in a Neo4j database and utilizes hybrid Cypher queries to efficiently identify events that share both semantic and structural similarities with a given query. Finally, the LLM Inference stage utilizes these retrieved causal graphs in a few-shot learning setup with XML-based prompting, enabling robust classification and tagging of causal relationships. Experimental evaluations demonstrate that GraphRAG-Causal achieves an impressive F1-score of 82.1% on causal classification using just 20 few-shot examples. This approach significantly boosts accuracy and consistency, making it highly suitable for real-time applications in news reliability assessment, misinformation detection, and policy analysis.