Arūnas Čiukšys and Rita Butkienė, Department of Information Systems, Kaunas University of Technology, Kaunas, Lithuania
Automatic event extraction (EE) is a crucial tool across various domains, allowing for more efficient analysis and decision-making by extracting domain-specific information from vast amounts of textual data. In the context of under-resourced languages like Lithuanian, the development of EE systems is particularly challenging due to the lack of annotated datasets. This study investigates and evaluates the event extraction capabilities of two large language models (LLMs): OpenAI's GPT and Google Gemini, using few-shot prompting. We propose novel methodologies, including a combined approach and a layered prompting approach, to improve the performance of these models in identifying two specific event types. The models were benchmarked using various performance metrics, such as accuracy, precision, recall, and F1-score, against a manually annotated gold-standard corpus. The results demonstrate that LLMs achieve satisfactory performance in extracting events in Lithuanian, though model accuracy varied depending on the prompting methodology. The findings underscore the potential of LLMs in addressing event extraction challenges for under-resourced languages, while also pointing to opportunities for improvement through enhanced prompt strategies and refined methodologies.
Event Extraction, LLMs, Few-Shot Prompting, Gemini, GPT, Layered Prompting, Combined Prompting
Mohammad Mozammal Huq, Statistics Department, Jahangirnagar University, Bangladesh
The usefulness of several machine learning models and strategies for sentiment analysis is examined in this research study. The gathered data and analysis offer insightful knowledge into the subject of sentiment analysis and its application to a significant number of unlabeled customer reviews and comments on Amazon products. To categorize the sentiment of the reviews, the paper suggests a supervised research model that includes two different feature extractors. Along with a thorough overview of pertinent literature on sentiment analysis utilizing text-based datasets, the core theory of the model, analysis techniques, and performance standards are all the experiments conducted on a small dataset yielded promising results, with an accuracy of over 82 percent achieved by the random forest model. The comparison of different data quantities using cross-validation, varied training-testing ratios, and various feature extraction methods contributed to the robustness of the findings.
Sentiment Analysis, Machine Learning, Text Classification, NLP.
Yashodeep Basnet. University of Northampton , UON, Kathmandu, Nepal
Recommender systems are pivotal in enhancing user experiences across digital platforms by providing personalized content. However, the "cold-start" problem, where systems struggle to generate meaningful recommendations for new users or items due to limited historical data, remains a significant challenge. This paper presents a novel hybrid recommender system that integrates collaborative filtering, content-based filtering, and deep learning techniques to address cold-start scenarios effectively. Utilizing the MovieLens 100K dataset, the proposed system leverages the strengths of each methodology to improve recommendation accuracy and robustness. Experimental results demonstrate that the hybrid model outperforms traditional methods, achieving lower error rates and higher precision, recall, and F1-scores, thereby validating its efficacy in handling data sparsity and enhancing user satisfaction.
Recommender systems, collaborative filtering, content-based filtering, deep learning, cold-start problem
Haseebullah Jumakhan and Lana Weiss, Artificial Intelligence Research Center (AIRC), Ajman University, Ajman, United Arab Emirates
Abstract. Lung cancer remains a leading cause of cancer-related deaths worldwide. This study employs data mining techniques to analyze and predict lung cancer risk based on various patient attributes. Using a dataset of 1000 patients, we explore the relationships between factors such as air pollution exposure, smoking habits, and genetic risk with the likelihood of developing lung cancer. We implement and compare four machine learning models: Multinomial Logistic Regression, Random Forest, Naive Bayes, and a Neural Network. Our findings demonstrate the potential of these models in predicting lung cancer risk, with the Random Forest and Multinomial Logistic Regression models showing particularly high accuracy. This research contributes to the growing body of work on early lung cancer detection and risk assessment, potentially aiding in more timely and effective interventions.
Lung Cancer Risk Prediction, Machine Learning, Data Mining, Model Comparison, SHAP Analysis.
Hifsa Malik and Ruba Qasim, Artificial Intelligence Research Center (AIRC), College of Engineering and Information Technology, Ajman University, Ajman P.O. Box 346, United Arab Emirates
This report represents our project of using machine learning models to forecast the performance of students. The dataset used comprises a wide range of academic, social, and demographic factors of secondary school students in Portugal, largely with regard to their Mathematics performance. Fortunately, we have data models that can be used to analyse and predict student performance. Firstly, the dataset is pre-processed through various data cleaning techniques followed by regression-based models among which are Linear Regression, Random Forest, Gradient Boosting, Support Vector Regressor, and XGBoost for grade prediction. This model will allow schools to have a predictive view of students’ performance at the beginning every year, to set an improvement plan for students.
Linear regression, random forest, feature selection, preprocessing, visualization, modelling, cross validation, outliers, standardization, SMOTE, feature importance, gradient boosting, SVR, XGboost, MAE, MSE,R2.
Michael Abobor and Darsana P. Josyula, Department of Computer Science, Bowie State University, Bowie, USA
Imbalanced datasets present significant challenges in machine learning. The disproportionate distribution of labels in imbalanced multi-label datasets is a result of the low datapoints of the minority class. This leads to biases in model predictions as algorithms tend to favor the majority class, resulting in poor generalization for the minority class. Any effort to balance the inequality within each class can inadvertently create issues across the other classes. This paper introduces the multi-view learning approach that combines pre- trained large language models and embeddings augmented with techniques such as SMOTE, MLeNN, MLSMOTE, MLSOL, and MLTL. This helps address the issue of imbalanced multi-label datasets in classification. This dual input model combines the original tokenized text, and the augmented embeddings extracted from the penultimate layer of the transformer, giving the model the ability to learn from both sources of information. This approach conserves the contextual significance of the input text and makes it possible for training transformers with the augmented embeddings thereby tackling the issue of imbalance multi-class datasets.
Imbalanced datasets, Multi-label, Transformer, Augmented Embeddings, Machine Learning.
Peicheng Yu1, Christopher Wadley2, 1Marks School, 25 Marlboro Rd, Southborough, MA 01772, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768
This paper addresses the inefficiency of traditional staff training for camp employees, which often requires extended in-person sessions before the camp begins. StaffPrep is designed to reduce the time staff spend on-site for training, allowing them to enjoy their summer vacation while learning essential skills remotely [1]. The app has three core components: an AI-driven feature that generates situational questions and answers to quiz staff members; a multiplechoice quiz platform that records scores and reports them to the camp for staff qualification; and an authentication system that enables users to create accounts and track their learning progress. To evaluate the app’s effectiveness, 32 AI-generated questions were assessed for quality and relevance on a scale of 1–10, with results indicating high relevance and quality. Time is valuable, and StaffPrep empowers users to save time while maintaining the integrity and efficiency of the training process.
Remote training, AI-generated quizzes, Staff qualification, Camp employee efficiency.
Mihaela Popa1, 2, Valentina Emilia Balas1, 2, 3, Dana Rad1, 4, 1Doctoral School of Systems Engineering, Petroleum-Gas University of Ploiești, 100680 Ploiești, Romania, 2Faculty of Engineering, Aurel Vlaicu University of Arad, 310032 Arad, Romania, 3The Academy of Romanian Scientists, str. Ilfov nr. 3, sector 5, București, Romania, 4Center of Research Development and Innovation in Psychology, Faculty of Educational Sciences Psychology and Social Work, Aurel Vlaicu University of Arad, 310032 Arad, Romania
Amidst the growing concerns of global climate change, sectors worldwide face increased pressure to adopt sustainable practices and enhance carbon management strategies. Carbon offsetting, wherein organizations counterbalance their greenhouse gas emissions by investing in projects that reduce or eliminate emissions elsewhere, has emerged as a pivotal strategy, especially within the building industry due to its substantial carbon footprint reduction potential. This paper delves into the current state and emerging trends in carbon offsetting within the building sector through a bibliometric analysis of literature from the Web of Science Core Collection. Using VOSviewer, the analysis maps bibliographic data from 87 relevant articles, identifying four thematic clusters from 611 keywords with a minimum cooccurrence threshold of two. The findings reveal key thematic areas, including renewable energy integration, urban planning, and challenges in methodological frameworks, providing actionable insights for policy development and industry practices. The study emphasizes the critical need for robust methodologies in carbon offset projects to ensure genuine environmental benefits, addressing challenges like baseline manipulation. Finally, the research identifies opportunities for future exploration in socioeconomic impacts and advanced modeling tools for carbon management in the built environment.
carbon offsetting, building industry, bibliometric analysis, sustainability, rooftop photovoltaics, urban planning
Camus Hu1, Yu Sun2, 1Fairmont Preparatory Academy, 2200 W Sequoia Ave, Anaheim, CA 92801, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768
Wristwell presents a novel approach to environmental monitoring, offering real-time insights into air, water, and noise pollution levels [1]. This paper discusses the development, mplementation, and potential of Wristwell in addressing critical environmental challenges [2]. We explore the systems architecture, sensor integration, data processing algorithms, and user interface design. Additionally, we examine the challenges and limitations faced by Wristwell, including sensor accuracy, network connectivity, and scalability issues. We propose strategies to mitigate these challenges, such as sensor calibration, offline capabilities, and user experience enhancements. Through these efforts, Wristwell aims to provide accurate, reliable, and actionable pollution data to support informed decisionmaking and promote environmental sustainability.
Environmental Monitoring Systems, Real-Time Pollution Insights, Sensor Integration and Data Processing.
Daniel Alexuta1,2, Valentina Emilia Balas1,2,3*, Marius Mircea Balas2, 1Doctoral School of Systems Engineering, Petroleum-Gas University of Ploiești, 100680 Ploiești, Romania, 2Faculty of Engineering, Aurel Vlaicu University of Arad, 310130 Arad, Romania, 3The Academy of Romanian Scientists, str. Ilfov nr. 3, sector 5, București, Romania
Aquaponics represents the integrated production of fish and hydroponic crops using recirculation of a nutrient solution in such a way that fish excretions are used as fertilizer for plants. There is a great interest in aquaponics in the EU and around the world due to increased interest in sustainable agriculture. Such systems can operate in any area and can be controlled remotely via mobile applications. In this article, we will introduce the different types of aquaponic systems: media-based aquaponic systems, nutrient film techniques, and deep-water culture. We will also present various developments of aquaponic systems in Romania. Additionally, we will provide a review of recent literature in the field of aquaponics, including different statistics. Many studies have been conducted on such aquaponic systems, reviewing the general concept, components, types of parameters, and factors influencing the productivity and efficiency of such systems. We will discuss the water parameters in aquaponics that are important for both plants and fish. Maintaining water quality parameters is crucial to provide sufficient nutrients for growing fish and plants, and monitoring these parameters is essential for the health of the aquaponic system. The contribution of this paper is the provision of detailed and indepth discussion on aquaponics.
Aquaponics, Monitoring, Control, Sustainability, Urban agriculture, Greenhouses.
Gabrielle Marie Alfonso and Sherwin Romano, School of Information Technology, Mapua University, Makati, Philippines
As the utilization of online services becomes increasingly prevalent in the lives of Filipino individuals, the security of personal information assumes paramount importance [1]. Passwords serve as the primary line of defense for user accounts, making their strength and effectiveness crucial [1]. This research paper aims to investigate the strategies of password security adopted by Filipino individuals and their engagement with online services as an integral part of their daily lives [1]. By examining the unique context of the Filipino population, including cultural factors and socioeconomic considerations, this study seeks to provide valuable insights into the password practices, challenges, and opportunities specific to this demographic [1, 2]. The research findings will enhance the overall cybersecurity landscape and help shape tailored recommendations for the Filipino population, promoting safer and more secure online experiences [1].
Password Strategies, Password Security, Secure Passwords, Cybersecurity, Data Breaches, Confidential Information Protection, Digital Transformation.
Rasim Alguliyev and Ramiz Shikhaliyev, Ministry of Science and Education Republic of Azerbaijan, Institute of Information Technology, Baku, Azerbaijan
Cybersecurity monitoring is essential for safeguarding computer networks. However, the increasing scale, complexity, and data volume of modern networks present significant challenges for traditional monitoring methods. To address these challenges, we propose a deep learning-based method for network security monitoring. Our method integrates convolutional neural networks (CNNs) with long short-term memory (LSTM) models. Trained on the CICIDS2017 dataset, the proposed model achieved a classification accuracy of 96.76% and an error rate of 9.34%, showcasing its effectiveness in managing complex and voluminous network data.
Computer Networks, Computer Network Cybersecurity Monitoring, Deep Learning Model, CNN-LSTM Model, Network Traffic Classification
A.U. Kalizhanova1,2, A.U. Utegenova1,2, S.Zh. Daruish1,2, S.L. Tikhomirov1,2, M.A. Vorogushina1,2 and Z.M. Rakhimzhanova1,2, 1NAO "Almaty University of Power Engineering and Telecommunications named after Gumarbek Daukeev," Kazakhstan , 2Institute of Information and Computational Technologies, CS MES RK, Kazakhstan
The study is dedicated to the development of a geoinformation system for the environmental passportization of rocket booster drop zones. It examines the principles of designing a modular system architecture, which includes basic data processing modules and functional monitoring modules. The implementation features of modules for data loading, unloading, and exchange, enabling the integration of various information sources, are analyzed. The functional capabilities of monitoring modules designed to handle data on natural and climatic conditions, pollution sources, and contamination levels in the territories are described. The results of the system implementation are presented, demonstrating the effectiveness of the proposed solutions in geospatial data management and environmental monitoring organization. The developed system lays the foundation for a comprehensive assessment of the environmental condition of territories affected by rocket and space activities.
Geoinformation System, Environmental Passportization, Modular Architecture, Environmental Monitoring, Adaptive Data Representation.
Lin Zhang, Qing Li, Yang Wang, and Jingmei Zhao, Southwestern University of Finance and Economics Chengdu, Sichuan , China
Large Language Models (LLMs) have significant reasoning capabilities and can act as agents interacting with the real world. However, they are often segmented and, unlike humans, lack integrated systems for validating their thoughts and actions. This limitation often leads LLMs to encounter “local optima” in task performance. To mitigate this problem, we propose a replanning mechanism for LLM-based agents that dynamically incorporates feedback from actions and exploits implicit information not initially available in the reasoning framework. This approach effectively bridges the gap between the cognitive and action phases of LLMs. Experimental results on real world ticket booking platforms such as Ctrip.com and Booking.com show that our method exhibits greater robustness in following clear instructions, successfully completing more steps, and achieving a higher success rate in practical applications, especially in complex tasks requiring interactive reasoning and action.
LLM, Agents, Replanning, Reaction, Logical path reasoning.
Anwur Alenezi1, Abouelyazed Kuliab2 and Yousef Alabaiadly3, 1Water Research Centre, Kuwait Institute for Scientific Research, Safat 13109, Kuwait, 2New & Renewable Energy Authority, Ministry of Electricity & Renewable Energy, Egypt, 3Studies and Research Department, Ministry of Electricity and Water, Ministries Zone 12010, Kuwait
This paper investigates the feasibility of recovering energy from exhaust air of air conditioning central plants experimentally. A mini vertical axis small wind turbine is connected directly with the exhaust air of condenser fan. The exhaust air energy recovery unit includes an air rotor-blades with generator. The electricity produced from the recovery unit is based on the value of fan speed and its air flow rate, CFM. The exhaust air from different central types of A/C systems is measured experimentally. As example, the speed of air exhausted in a central package unit with capacity of 5 ton refrigeration (TR) is reached to 8-15 m/s. The wind speed levels play significant effect on the performance of wind turbine. a small vertical wind turbine was installed on the exhaust of air flowing from the condenser fan of central package 10-ton refrigeration (TR). The proposed exhaust air energy recovery unit is produced to 35-40% of the total energy consumption in a building having an A/C plant.
Energy recovery, wind turbine, exhaust air, HVAC systems.
Merita Bakiji, Faculty of Contemporary Sciences and Technologies , South East European University , Tetovo, North Macedonia
As a result of the great boom experienced by global business, rapid technological developments, IT Outsourcing came as a result of organizations attempts to reduce operational costs and increase efficiency through external expertise.Through this study, it is intended to explore the current models of IT Outsourcing, detailing their sustainability and suitability in different market environments.This goal is attempted to be achieved by relying on a comprehensive summary of existing literature, articles and existing studies on IT Outsourcing, industry reports, consultancy reports, technological trends and their impacts on the market.The study also analyzes the IT Outsourcing industry map in the Republic of North Macedonia revealing the IT Outsourcing market and trends.By synthesizing existing research and data, this paper presents a valuable resource for decision makers in IT outsourcing, by providing practical recommendations that can serve organizations that are constantly trying to adapt to with rapidly changing market conditions.
IT Outsourcing, Artificial Intelligence, Market Trends, North Macedonia.
Sikha S. Bagui1, Dustin Mink2, Subhash C. Bagui3, Jadarius Hill1, Farooq Mahmud1 and Michael Plain13, 1Department of Computer Science, University of West Florida, Pensacola, Florida, USA, 2Department of Cybersecurity, University of West Florida, Pensacola, Florida, USA, 3Department of Mathematics and Statistics, University of West Florida, Pensacola, Florida, USA
This study uses motif-based graph databases to classify tactics in the MITRE ATT&CK framework. Machine Learning classification models capable of detecting Reconnaissance network attack tactics, labelled as per the MITRE ATT&CK framework, are created for the newly created UWF-ZeekData22 dataset. The work analyzes Zeek Connection logs. Feature selection is performed using graph motifs. Results show that model performance can be increased using various network graph motifs. Upon completion of this work, it was concluded that the most important feature for predicting Reconnaissance network attacks within the Zeek Connection Logs dataset was the “From” feature, which represents the network address from where the connection is originating. It was also determined that, irrespective of which motif was used to train the model, the Decision Tree algorithm performed best.
Assessing the Accuracy of Variational Quantum Eigensolver and Quantum Phase Estimation for Molecular Hydrogen
Xiaofei Zhao1 and Hua Wang2, 1School of Information Engineering, Lanzhou Petrochemical University of Vocational Technology,, 2School of Automation and Electrical Engineering, Lanzhou Jiaotong University Digital Business Department, State Grid Gansu Electric Power Company
Quantum Computing, Quantum Chemistry, Variational Quantum Eigensolver (VQE), Quantum Phase Estimation (QPE), Hydrogen Molecule, Ground State Energy, Ansatz, Electron Correlation,Trotterization, Quantum Simulation.
Graph Databases, Motifs, Reconnaissance, Machine Learning, Cybersecurity.
Alfrin Saldanha and Hyoil Han, School of Information Technology, Illinois State University, Normal, USA
The rapid advancement of Artificial Intelligence has driven the adoption of machine learning technologies across diverse domains, with recommender systems playing a pivotal role in delivering personalized suggestions. However, as user-centric applications become increasingly sophisticated, providing recommendations without clear explanations is no longer adequate. Explainable recommendation systems bridge this gap by enhancing transparency, user understanding, and trust through interpretable and contextually relevant explanations. These systems strive to balance high recommendation accuracy with the clarity of their explanations. This paper examines state-of-the-art models and methodologies in explainable recommendation systems, focusing on their computational underpinnings, evaluation metrics, and practical outcomes. We analyze the strengths and limitations of existing approaches and discuss opportunities for integrating innovative techniques and emerging technologies. Our study aims to advance the development of more effective, explainable recommendation systems adaptable to diverse application domains, aligning with the interdisciplinary focus of computational science.
Explainable Recommender Systems, Artificial Intelligence, Knowledge Mining, Virus, Machine Learning.
N Chitha, O R Mnyaka; J T Thabethe, N Ntsele, S Nomatshila; W Chitha, NV Khosa, and R Tshabalala, Department of Public Health, Faculty of Medicine and Health Science, University of Walter Sisulu, Mthatha, South Africa
Information is crucial tool for nurses, and how they acquire and use it is key to their performance. Professional nurses need information that is accessible, good quality, up-to- date, manageable, and relevant, as well as information services that assist in finding that information. This study assessed the most reliable information sources nurses use to make clinical decisions at the point of care. A quantitative cross-sectional survey was conducted in four referral hospitals across four South African provinces, Mpumalanga, Limpopo, Eastern, and Northern Cape, between May and July 2022. The hospitals were identified using simple random sampling. Stratified random sampling was utilised to select nurses within the hospitals. Data were entered into Microsoft Excel and analysed using STATA version 17 and SPSS version 26. Nurses mostly relied on nursing colleagues (86.8%, 362/417) or doctors (78.4%, 309/394) for information whilst they sometimes consulted protocols or guidelines (63.5%, 247/389).
Reliable Information, Information Sources, Nurses Information, Health Information.
N Chitha, O R Mnyaka; J T Thabethe, N Ntsele, S Nomatshila; W Chitha, NV Khosa, and R Tshabalala, Department of Public Health, Faculty of Medicine and Health Science, University of Walter Sisulu, Mthatha, South Africa
This paper gives complete guidelines for authors submitting papers for the AIRCC Journals. In this paper, we study Fermats equation, with positive integers such that . Consider the set of hypothetical solutions of equation (1) and . Let be a prime, we establish the following results: - . This completes the direct proof of Abels conjecture. - . This completes the direct proof of the second case of even exponent FLT. - if is a non-prime odd integer. - If then . This provides simultaneous Diophantine evidence for the first case of FLT and the second case . We analyse each of the evidence from the previous results and propose a ranking in order of increasing difficulty to establish them.
Fermat Last Theorem, First case, Secund case, Abel Conjecture, Fermat equation, Kimou main divisors Theorem. Odd non-prime exponent.
Kimou Kouadio Prosper and Kouassi Vincent Kouakou, Alan Verdegraal,Mountainair, New Mexico, USA
In Mathematics, there is no general procedure for any Natural Number n to determine whether n-1 or n+1 is a prime or composite simply by examining n itself. Factorization of n fails to produce meaningful information regarding the primality of n-1 and n+1. The research discussed in this paper shows how representing a number n as a distinct set of sequences, heuristically derived from a circle with n points, demonstrates the primality of not only n but of n-1 and n+1; i.e., the Natural Number n “knows” whether its immediate neighbors n-1 and n+1 are either prime or composite. This method, although simple to comprehend, has significant implications for the Theory of Numbers.
Number Theory, Primality Checks, Prime Numbers, Composite Numbers.
Nikitha Merilena Jonnada, Information Technology (Information Security Emphasis), University of the Cumberlands, Williamsburg, Kentucky, USA
In this paper, the author discusses how the techniques and methods of the workforce are being replaced by Artificial Intelligence (AI), which could be both advantageous and disadvantageous. The concepts of AI and Machine Learning (ML) could be helpful to many students, artists, engineers, marketing professionals, and job professionals when used with caution.
Artificial Intelligence, Machine Learning, Security, Hackers, Cloud.
Imen Chebbi, Department of Computer Engineering, FSEG Sfax
In order to help athletes avoid injuries, prevention strategies are increasingly incorporating contemporary techniques like machine learning, which allow for an assessment of injury risk. This essay’s goal is to assess the injury risk for 250 athletes. The players self-reported their physical and psychological health every morning and evening using a bespoke application, which served as the risk indicators for the day. The output data matched the injuries reported by the athletes. A CNN model that predicted the chance of an injury was trained and optimized using the evaluated characteristics. Our model’s performance score has an accuracy of 99.63. It is challenging to quantify the risk of harm due to the disparity between the number of injuries and observations. The prediction model indicated that positive emotional and physical aspects were the most important.
CNN, Injury, Detection, Behavioral.
Liza Simran Fernandes, and Hyoil Han, School of Information Technology, Illinois State University, Normal, USA
Entity linking is the task of identifying and assigning precise meanings to entities within textual content using machine learning and natural language processing techniques. Much like word sense disambiguation relies on dictionaries to clarify word meanings, entity linking utilizes knowledge bases to resolve ambiguity and establish contextual relevance for entities. This reliance underscores the importance of publicly accessible knowledge bases, such as YAGO, DBpedia, and Wikipedia, which are widely used for this purpose. In this paper, we conduct a comprehensive investigation into existing entity-linking techniques, examining their strategies and the role of knowledge bases in enhancing their effectiveness.
Entity Linking, Machine Learning,Natural Language Processing, Knowledge Bases.
Copyright © ISTECH 2024