Albert van Niekerk, Department of Applied Research, AvNFoundationRsa, RSA
Intellectual property (IP) management is evolving with artificial intelligence (AI) integration, offering enhanced efficiency, accuracy, and strategic decision-making. This paper presents a framework for AI-driven IP portfolio development, positioning prompt engineering as both an intellectual asset and a tool for optimising key processes such as identification, valuation, and monetisation. The framework aligns with international standards, including ISO 56005, ISO 10668, and ISO 31000, ensuring compliance and governance. A structured validation process using weighted scoring and statistical methods enhances the reliability of AI-generated insights. Case studies highlight the framework's speed, scalability, and cost-efficiency benefits while addressing data quality, bias, and regulatory compliance challenges. The paper concludes with recommendations for businesses and policymakers to adopt AI-driven IP strategies and suggests future research directions, contributing to the growing discourse on AI in IP management.
Artificial intelligence (AI), Intellectual property (IP) management, Prompt engineering,IP portfolio development, ISO standards compliance, IP valuation and monetisation, AI-driven decision-making, Risk management, Ethical AI considerations, and Innovation strategy.
Xia Li, Allen Kim, The Department of Software Engineering and Game Design and Development, Kennesaw State University, Marietta, USA
To improve the work efficiency and code quality of modern software development, users always reuse Application Programming Interfaces (APIs) provided by third-party libraries and frameworks rather than implementing from scratch. However, due to time constraints in software development, API developers often refrain from providing detailed explanations or usage instructions for APIs, resulting in confusion for users. It is important to categorize API reviews into different groups for easily usage. In this paper, we conduct a comprehensive study to evaluate the effectiveness of prompt-based API review classification based on various pre-trained models such as BERT, RoBERTa, BERTOverflow. Our experimental results show that prompts with complete context can achieve best effectiveness and the model RoBERTa outperforms other two models due to the size of training corpus. We also utilize the widely-used fine-tuning approach LoRA to evaluate that the training overhead can be significantly reduced (e.g., 50% reduction) without the loss of the effectiveness of classification.
Software engineering, API review classification, pre-trained models, fine-tuning.
T D C Pushpakumara1, And Fazeela Jameel Ahsan2, 1Department of Civil Engineering, University of Moratuwa, Sri Lanka, 2Department of Marketing, University of Colombo, Sri Lanka
This systematic literature review explores the transformative role of artificial intelligence (AI) chatbots in promoting sustainable tourism, particularly in the ecotourism sector. AI chatbots are pivotal in enhancing operational efficiency, fostering environmental responsibility, and improving tourist engagement. The study identifies their contributions to sustainability by optimizing resource use, reducing environmental impact, and educating tourists about local cultural and ecological practices. Despite these benefits, significant challenges such as data privacy concerns, infrastructural limitations, and cultural biases hinder widespread adoption. The findings emphasize the need for robust digital infrastructure, ethical frameworks, and culturally adaptive chatbot designs to overcome these barriers. By aligning technological innovation with sustainability goals, AI chatbots can significantly advance sustainable tourism practices. Future research should prioritize empirical analyses and inclusive strategies to maximize the potential of AI chatbots in fostering long-term sustainable development in ecotourism.
AI, Chatbots, Ecotourism, Sustainability, Management, innovation.
Sona Daison, Department of Computer Science and Engineering Karunya Institute of Technology and Sciences, India
Theworld-wide populationwithasthma experiences ongoingmedical issues because urgent emergency conditions often result in necessary hospital admissions which impacts their general health quality.The control of asthma becomes harder because asthma triggers suddenly emerge from the interaction between environmental factors and personal health conditions. The Asthma Wellness Care Platform resolves asthma care difficulties by integrating AI technology for predicting physician records and individualized treatment and continuous data monitoring systems. The application processes real- time data by merging Air Quality Index information with sleeppattern and stress measurement data submitted by users through multiple machine learning models including KNN, Random Forest and XGBoost and Logistic Regression to forecast asthma attacks. Using this platform lets users access breath exercise tools while also providing them with an asthma journal record system to enhance asthma management.The chatbot responds immediately to userneedsatthe same time emergency alerts immediately contact both emergencyresponders and healthcare providers in criticalsituations. The platform provides secure respon- sible services for data sharing authentication that allows users to enhance asthma management while decreasing hospital visits and boosting medical care efficiency through its novel features.
asthma prediction, machine learning, smart healthcare, real-time monitoring, artificial intelligence, personalized management.
Yuxun Chen, Jianlang Deng, Zili Wang, Mingrui Li, Zexuan Pan, Computer Science and Engineering Faculty, South China University of Technology, Guangzhou, China
With the development of intelligent manufacturing and industrial automation, circuit board quality inspection, as a crucial part of industrial production, urgently needs efficient and precise target detection models. This project aims to design and optimize a target detection model based on deep learning methods that can quickly and accurately identify defective circuit boards. By introducing the RepVGG structure to improve the Yolov5 backbone network and integrating multiple attention mechanisms (such as CBAM, SE, SCA, etc.), this research significantly enhances the detection performance. Experimental results show that the improved Yolov5 + RepVGG + SE model achieved an accuracy rate of 89% on the defective circuit board dataset provided by the Beijing University Intelligent Robot Open Laboratory, which is higher than other combinations.
Intelligent Manufacturing, Object Detection, RepVGG, Attention Mechanism, Defect Detection.
Vasiliy Znamenskiy, Rafael Niyazov, Joel Hernandez, Borough of Manhattan Community College, The City University of New York, USA
This study investigates the educational potential of generative artificial intelligence (GenAI) platforms based on large language models (LLMs), such as ChatGPT, Claude, and Gemini, as tools for student-centered learning. Recognizing the current limitations of GenAI—particularly its propensity for generating inaccurate or misleading information—the paper proposes a novel instructional strategy: an interdisciplinary laboratory designed to foster critical evaluation of GenAI-generated outputs. In this pedagogical model, students engage with GenAI systems by posing questions or solving problems drawn from topics they have already studied and understand. Equipped with correct answers, students are positioned to assess the accuracy, and relevance of AI-generated responses across multiple modalities, including text, images, and video. Students design such difficulty prompts and tasks which help compare the intellectual performance of various GenAI. This approach was implemented for a specially designed lab session within a general astronomy course for non-science majors. Multiple student groups completed the lab, demonstrating high levels of engagement, initiative, and critical thinking. Findings suggest that such activities not only deepen students’ comprehension of scientific content of learning courses but also cultivate essential skills in digital literacy and critical interaction with AI technologies.
