17 research outputs found
Part of speech (POS) tagging in Roman Urdu: datasets and models
No full textRoman Urdu is a prevalent medium of expression on social media, news websites, and text messages in the subcontinent, making it a valuable data source for social media and text analytics, particularly in the Indo-Pak perspective. However, despite the immense potential, limited efforts have been made in the area of Roman Urdu text analytics due to various complexities, such as a lack of a standard lexicon, the informal nature of the text, and the lack of text processing tools. The development of the Roman Urdu Part-of-Speech (POS) dataset and the implementation of a robust tagger hold immense importance for text analytics in Roman Urdu. In this work, we created a comprehensive, large-scale Roman Urdu POS dataset and developed a Roman Urdu POS tagger, laying the foundation for future advancements in advanced text analysis. Our approach involved the utilization of Hidden Markov Models, Neural Networks, state-of-the-art transformer models, and Large Language Models as baselines. In our work, we curated two distinct test datasets: one with lexical variation and the other without such variation. This approach allowed us to test the model’s robustness in handling different linguistic challenges posed by lexical variations. Our tagger yields high-quality output with an accuracy score of 96% without lexical variation and 86% on test data with lexical variations. We also evaluated state-of-the-art Large Language Models (GPT-4o and Llama-3-8B) in zero-shot and few-shot settings, with GPT-4o achieving up to 53.78% accuracy in the few-shot configuration, demonstrating a substantial performance gap compared to specialized models. This work establishes a comprehensive framework for Roman Urdu POS tagging that effectively addresses lexical variation challenges, providing essential resources and benchmarks for advancing Roman Urdu natural language processing research
Computational Cavitation Analysis of a Submerged Body at Different Depths
No full textTransient's analysis is performed to determine the cavitation inception on a submerged body and flow dynamics under different working conditions. Cavitation causing hydrodynamic, structural and noise problems is essential to avoid. Cavitation analysis is done for a particular geometry to investigate the location and determine the critical conditions for different depths at which cavitation take place. Simulation has been done using the commercial CFD code Fluent 6.1. Multiphase Mixture model and Standard K- ? turbulence model with standard wall function is used in the study. Analysis determines the region and critical velocity for a particular depth at which cavitation occurs. The time dependent analysis provides detailed insight into the hydrodynamics and highlights the capabilities and limitations of the cavitation model used
Computational Modeling of Gas Liquid Interfaces Using Different Multiphase Models
No full textA time dependent Computational Fluid Dynamics analysis of gas jets impinging onto liquid pools has been conducted. The aim of the study is to obtain a better understanding of highly complex, and industrially relevant flows in jetting system. Three different multiphase models, i.e., The Eulerian model, the volume of fluid model and the mixture model are used to analyze the surface deformations namely dimpling, splashing and penetration. The Standard k-? model is used to incorporate the Turbulence in the continuous phase. Two-dimensional axisymmetric geometries with different dimensions have been used in the study. Simulations are performed using commercial CFD code Fluent 6.1. PISO (pressure- implicit with splitting of operators) algorithm was employed to compute the pressure velocity coupling. The computed results are compared with experimental and theoretical data reported in the literature. Also the results of the study highlight and compare the discrepancies between the three multiphase models in capturing the flow structure and cavities formed at gas-liquid interfaces
A Survey of OCR in Arabic Language: Applications, Techniques, and Challenges
No full textOptical character recognition (OCR) is the process of extracting handwritten or printed text from a scanned or printed image and converting it to a machine-readable form for further data processing, such as searching or editing. Automatic text extraction using OCR helps to digitize documents for improved productivity and accessibility and for preservation of historical documents. This paper provides a survey of the current state-of-the-art applications, techniques, and challenges in Arabic OCR. We present the existing methods for each step of the complete OCR process to identify the best-performing approach for improved results. This paper follows the keyword-search method for reviewing the articles related to Arabic OCR, including the backward and forward citations of the article. In addition to state-of-art techniques, this paper identifies research gaps and presents future directions for Arabic OCR
Design Optimization of a Cavitating Submerged body using Computational Fluid Dynamics
No full textTransient's analysis is performed to determine the cavitation inception on a submerged body and flow dynamics under different working conditions. Cavitation analysis is done for a particular geometry to investigate the location and determine the critical conditions for different depths at which cavitation take place. Simulation has been done using the commercial CFD code Fluent 6.2.16. Multiphase Mixture model and Standard K- ? turbulence model with standard wall function is used in the study. Analysis determines the region and critical velocity for a particular depth at which cavitation occurs. The time dependent analysis provides detailed insight into the hydrodynamics and highlights the capabilities and limitations of the cavitation model used
GENERATIVE ARTIFICIAL INTELLIGENCE AND DIGITAL TRANSFORMATION IN CONTACT CENTER BUSINESSES
No full textIn today’s business landscape, digital transformation is critical to competitiveness, with generative AI (GAI) offering significant potential to improve customer interactions and operational efficiency - yet its adoption in contact centres faces challenges such as technical complexity, data privacy concerns, and resistance to change. In this study, we explore these barriers by surveying contact centre personnel. Our findings reveal that security risks, potential misuse, and process complexity drive reluctance toward GAI adoption, along with a knowledge gap to maximize its benefits. The study also uncovers obstacles at all levels of employees and stresses the need for strong governance, multi-stakeholder collaboration, a focus on data ethics, bias mitigation, and risk management. We recommend addressing these challenges to ensure successful integration of GAI, providing valuable insights for organizations that navigate digital transformation while balancing innovation with responsible implementation
Synthetic Browsing Histories for 50 Countries Worldwide: Datasets for Research, Development, and Education
No full textAbstract Browsing histories can be a valuable resource for cybersecurity, research, and testing. Individuals are often reluctant to share their browsing histories online, and the use of personal data requires obtaining signed informed consent. Research shows that anonymized histories can lead to re-identification, nullifying the anonymity promised by informed consent. In this work, we present 500 synthetic browsing histories valid for 50 countries worldwide. The synthetic histories are compiled based on real browsing data using a series of transformation criteria, including website content, popularity, locality, and language, ensuring their validity for the respective countries. Each history maintains the order of webpage accesses and covers a one-month period. The motivation for publishing this dataset arises from the community’s call for browsing histories from different countries for research, development, and education. The published synthetic browsing histories can be used for any purpose without legal restrictions
