Multi-label metric learning, as an extension of metric learning to multi-label scenarios, aims to learn better similarity metrics for objects with rich semantics. Existing multi-label metric learning approaches employ...
详细信息
Lexical simplification (LS) method based on pretrained language models is a straightforward yet powerful approach for generating potential substitutes for a complex word through analysis of its contextual surroundings...
详细信息
Lexical simplification (LS) method based on pretrained language models is a straightforward yet powerful approach for generating potential substitutes for a complex word through analysis of its contextual surroundings. Nonetheless, these methods necessitate distinct pretrained models tailored to diverse languages, often overlooking the imperative task of preserving a sentence’s meaning. In this paper, we propose a novel multilingual LS method via zero-shot paraphrasing (LSPG), as paraphrases provide diversity in word selection while preserving the sentence’s meaning. We regard paraphrasing as a zero-shot translation task within multilingual neural machine translation that supports hundreds of languages. Once the input sentence is channeled into the paraphrasing, we embark on the generation of the substitutes. This endeavor is underpinned by a pioneering decoding strategy that concentrates exclusively on the lexical modifications of the complex word. To utilize the strong capabilities of large language models (LLM), we further introduce a novel approach PromLS that incorporates the results of LSPG to generate heuristic-enhanced context, enabling the LLM to generate diverse candidate substitutions. Experimental results demonstrate that LSPG surpasses BERT-based methods and zero-shot GPT3-based methods significantly in English, Spanish, and Portuguese. We also demonstrate a substantial improvement achieved by PromLS compared to the previous state-of-the-art LLM approach. LS approaches usually assume that complex words and their replacements are individual terms, concentrating on word-for-word substitutions. To tackle the more challenging task of multi-word lexical simplification, including phrase-to-phrase replacements, we extend LSPG and PromLS into MultiLSPG and MultiPromLS. MultiLSPG identifies multi-word expressions matched with their corresponding word counts in specific positions, while MultiPromLS, akin to PromLS, utilizes these candidates to generate a heuristi
Big data (BD) has emerged as a transformative force, offering unprecedented opportunities for organizations to extract valuable insights and drive informed decision-making. This research paper presents a comprehensive...
详细信息
Chatbot-based tools are becoming pervasive in multiple domains from commercial websites to rehabilitation applications. Only recently, an eleven-item satisfaction inventory was developed (the ChatBot Usability Scale, ...
详细信息
The Internet of Things (IoT) is one of the most intriguing technological revolutions of the last few years due to its rapid expansion. The advancement of IoT applications is dependent on standard and real-time communi...
详细信息
ISBN:
(数字)9798350389128
ISBN:
(纸本)9798350389135
The Internet of Things (IoT) is one of the most intriguing technological revolutions of the last few years due to its rapid expansion. The advancement of IoT applications is dependent on standard and real-time communication protocols. The IoT protocols are designed to enable reliable messaging. However, the selection of a standard and efficient messaging protocol is a difficult and intimidating task for any organization. One of the main challenging issues facing the loT is wireless sen-sor networks (WSNs). The devices that make up these networks have low levels of memory, processing power, and energy. Certain restricted protocols must be in place to function in WSNs due to their constrained nature. In the context of a business application for monitoring, this paper shows the experimental performance of two protocols: Message Queue Telemetry Transport (MQTT) and Constrained Application Protocol (CoAP).
In the contemporary dynamic environment, integrating technology is crucial for effective solutions. This paper addresses the pressing need for accurate monitoring and management of natural resources, with a specific f...
详细信息
In the contemporary dynamic environment, integrating technology is crucial for effective solutions. This paper addresses the pressing need for accurate monitoring and management of natural resources, with a specific focus on forests, utilizing satellite imagery. The primary goal is to develop an image analytics system utilizing satellite imagery for automated tree counting. traditional methods, such as on-site evaluations and manual surveys, are expensive, time-consuming, and error-prone. The dataset comprises 91 lower orbit satellite images, each annotated with bounding boxes to ensure precise spatial information. Three algorithms were evaluated for tree counting performance using annotated images and bounding box labels. Metrics such as Mean Squared Error (MSE), Intersection over Union (IOU), recall, precision, and F1 score were utilized for performance assessment. An ensemble network, combining a faster Recurrent Convolutional Neural Network (RCNN) with object detection algorithms, demonstrated superior performance. The model achieved an accuracy of 94.8% and a precision of 0.854 in density regression. The proposed strategy outperformed previous approaches with an average density regression accuracy of 91.58%. The model's precision of 0.854 and accuracy of 94.8% enhance visualization and yield more precise tree counting data. These advancements underscore the superiority of our satellite imagery-based automated tree counting model for sustainable resource management.
The neurodegenerative disease known as Alzheimer's disease (AD), which affects people worldwide, is complicated to treat and expensive. Conventional approaches, including manual Support Vector Machine (SVM) based ...
详细信息
ISBN:
(数字)9798350379990
ISBN:
(纸本)9798350391558
The neurodegenerative disease known as Alzheimer's disease (AD), which affects people worldwide, is complicated to treat and expensive. Conventional approaches, including manual Support Vector Machine (SVM) based methods with separate feature extraction and preprocessing, lack automation and pose limitations. This project recognizes the pressing need for efficient and automated solutions. Building on existing work that employed SVM and manual processing for Alzheimer's diagnosis, this work proposes a novel approach. Utilizing brain MRI scans and the VGG16 architecture, our automated system achieves high accuracy in predicting both labeled and unlabeled data. The integration of VGG16 ensures a streamlined process, eliminating manual intervention in feature extraction and preprocessing. This automation enhances accessibility, reduces costs, and offers a user-friendly interface for users. In contrast to traditional methods, our proposed system allows users to input MRI scans, facilitating prompt and accurate predictions. Upon analysis, the proposed system provides doctor recommendations and suggests prescriptions. The VGG16-based approach not only ensures superior diagnostic accuracy but also streamlines the entire process, emphasizing the importance of early detection for improved outcomes in Alzheimer's disease management.
The escalating costs of healthcare globally necessitate the development of accurate prediction models to address the financial strain on individuals, families, businesses, and governments. This research employs linear...
详细信息
ISBN:
(数字)9798350348637
ISBN:
(纸本)9798350348644
The escalating costs of healthcare globally necessitate the development of accurate prediction models to address the financial strain on individuals, families, businesses, and governments. This research employs linear regression models to forecast healthcare expenditures, analyzing the intricate relationship between various factors and costs. A comprehensive literature survey explores topics such as the impact of environmental pollution on healthcare spending, prediction of surgical expenses, budget impact analysis for cancer treatment, and the integration of big data in healthcare infrastructure investments. The methodology involves gathering a diverse dataset encompassing patient demographics, medical history, treatment methods, and associated expenditures. Exploratory data analysis employs Pearson's correlation coefficient to identify trends and correlations between variables. Two linear regression models are proposed and compared, with the second model excluding non-significant variables for enhanced accuracy and interpretability. Results indicate that the second model, with refined feature selection, outperforms the first in terms of mean squared error and R-squared values. A comparative analysis of different studies underscores the diverse findings in healthcare economics, ranging from the impact of CO2 emissions on health to predicting surgery costs and budgetary savings in cancer treatment. The research contributes to the growing body of knowledge in healthcare economics, providing decision-makers with a valuable tool for precise expenditure estimation. The emphasis on linear regression models aids stakeholders in understanding factors influencing healthcare costs, guiding resource allocation, policy creation, and financial planning for more sustainable and effective healthcare systems.
Integrating Artificial Intelligence (AI) with the Internet of Things (IoT) has recently become more advantageous for various complex fields like innovative health, intrusion detection in various devices, and continuou...
详细信息
ISBN:
(数字)9798331518097
ISBN:
(纸本)9798331518103
Integrating Artificial Intelligence (AI) with the Internet of Things (IoT) has recently become more advantageous for various complex fields like innovative health, intrusion detection in various devices, and continuous monitoring of crops yielding in various farming lands. From the health point of view, it is essential to monitor multiple patients' health according to their abnormal health conditions. The proposed Integrated Health Monitoring System (IHMS) is developed in this work by combining AI-based models such as Recurrent Neural Networks (RNNs) and advanced IoT sensors. The IHMS is mainly used in health care applications such as heart care analysis and other abnormal health conditions like blood pressure, levels of oxygen, and glucose readings, and it transmits this information to AI-powered systems. The IHMS mainly analyzes abnormal patterns, detects anomalies, and provides timely analytical analysis of patient data. Finally, the proposed approach IHMS analyzes several datasets and shows the performance in terms of classification.
Open Government data (OGD) refers to the provision of data produced by the government to the general public, in a format that is readily readable and can be used by machines with ease. It can also promote transparency...
Open Government data (OGD) refers to the provision of data produced by the government to the general public, in a format that is readily readable and can be used by machines with ease. It can also promote transparency, improve decision-making, enhance accountability, create economic opportunities, and encourage civic engagement. The OGD can help citizens understand the government and its legitimacy and transparency. Thus, when the government shares its data with people, it helps to create trust by being transparent, accountable, and promoting innovative solutions that benefit everyone. However, each published dataset has no indication of its quality assessment at all; thus, making it difficult for citizens to assess the reliability of the data from the OGD. Therefore, a data quality assessment for OGD should be developed. This will help create effective datasets which in turn help users understand the data. This study proposes QUALYST, a system that assesses Thailand's OGD dataset and validates it for analytic and visualization purposes. The study focuses on designing the data storage and implementing the assessment system. Furthermore, the proposed data quality dimensions, the developed data pipeline, and the assessment process are elaborated. Finally, the prototype system is demonstrated using Thailand's OGD datasets with examples in a visualized format.
暂无评论