检索结果-内蒙古大学图书馆

4th International Conference on Advanced Research in computing, ICARC 2024

作者： Dinushan, S. Wijegunasekara, M. Carmel University of Kelaniya Faculty of Computing and Technology Department of Software Engineering Kelaniya Sri Lanka

ISBN: (纸本)9798350384864

The stock market plays a pivotal role in a nation's economy. People tend to invest in the stock market due to its high rate of return. However, predicting the stock market remains a formidable challenge. This research uses machine learning algorithms to predict the ASPI and S&P 20 monthly stock market indexes by considering global economic conditions, macroeconomic variables, and sentiment data. Correlation tests, forward stepwise regression, and p-value tests were carried out to identify the influential variables. Machine learning algorithms including deep learning were utilized for model building, using the influential variables dataset. Shuffling and non-shuffling methods were deployed during dataset splitting. Additionally, included the previous month's stock value to the influential variables dataset and re-conducted the model building to assess the performance. The best-performing model (with the lowest MAE rate) was selected separately for ASPI and S&P 20 prediction. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Towards Mechanistic Interpretability for Autoencoder compression of EEG signals 2

Towards Mechanistic Interpretability for Autoencoder compres...

引用

Joint of the 2nd World Conference on eXplainable Artificial Intelligence Late-Breaking Work, Demos and Doctoral Consortium, xAI-2024:LB/D/DC

作者： Hegedić, Leon Hobor, Luka Marić, Nikola Rogošić, Martin Ante Brcic, Mario University of Zagreb Faculty of Electrical Engineering and Computing Unska ulica 3 Zagreb Croatia

Convolutional Variational Autoencoders (VAEs) have found extensive application in dimensionality reduction, data compressibility, assessment, and signal analysis. However, a comprehensive understanding of their internal mechanisms remains elusive. This study aims to use mechanistic interpretability to elucidate the inner workings of VAEs. By training VAEs on images generated by interpolating EEG signals from the human brain and analyzing the resulting latent space, as well as the signal propagation through the network layers, we aim to construct an explanation of how these specific models internally analyze the generated images. Since we work under hardware constraints, we devised an iterative approach that breaks big task into easier, more manageable steps. © 2024 Copyright for this paper by its authors.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Exploring Pre-scoring Clustering for Short Answer Grading 46

Exploring Pre-scoring Clustering for Short Answer Grading

引用

46th ICT and Electronics Convention, MIPRO 2023

作者： Petricioli, L. Skracic, K. Petrovic, J. Pale, P. University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia

ISBN: (纸本)9789532331042

Automatic short answer grading is a topic that has gained significant popularity recently, especially due to developments in natural language processing. While automated grading in computer supported assessment tasks traditionally imposed significant restrictions on the answer format (e.g., multiple choice questions), automated short answer grading could enable assessment scalability with very few answer format limitations and thereby increase the assessment tasks' validity. Here, 'short answer' refers to a text of up to, approximately, 10 sentences. However, automatic solutions require a lot of pre-graded material. In this paper, several pre-trained machine learning models were utilized to explore pre-scoring clustering for short answer grading of text in Croatian. The aim of this approach is to shorten the process of manual short answer grading by clustering similar answers, facilitating the development of automatic grading solutions. The described approach was evaluated on a dataset containing graduate students' answers in Croatian to six questions related to cyber security topics. The obtained results are promising and show how increases in cluster purity, normalized mutual information, Rand index, and adjusted Rand index measures can be achieved by finetuning a pre-trained model. © 2023 MIPRO Croatian Society.

关键词： Grading

来源：评论

学校读者我要写书评

暂无评论

Uninterruptible Power Supply System for Railway Infrastructure with Power Factor Correction and Immunity to Energy Strike 46

Uninterruptible Power Supply System for Railway Infrastructu...

引用

46th ICT and Electronics Convention, MIPRO 2023

作者： Hecimovic, Ante Patrlj, Leo Sunde, Viktor Ban, Zeljko University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia

ISBN: (纸本)9789532331042

This article deals with an uninterruptible power supply (UPS) for infrastructure equipment used in the railway system. The main feature of this UPS is the high power factor and resistance to overvoltages that occur in the railway network. Power factor correction (PFC) is done by adding a boost DC converter. The resistance to overvoltages is achieved by means of protective elements, i.e. varistors and inductors, as well as an additional overvoltage detection in the power supply network. Based on the output of the mentioned device, the part of the topology intended for power factor correction can be switched off. © 2023 MIPRO Croatian Society.

关键词： Railroads

来源：评论

学校读者我要写书评

暂无评论

A Transfer Learning Method for Hate Speech Detection 46

A Transfer Learning Method for Hate Speech Detection

引用

46th ICT and Electronics Convention, MIPRO 2023

作者： Smuc, Eva Delac, Goran Silic, Marin Vladimir, Klemo University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia

ISBN: (纸本)9789532331042

In this work we explore the possibilities of using transfer learning techniques to enhance performance of hate speech detection models by relying on similar linguistic problems (e.g. toxic language detection). Multiple algorithms are trained for similar linguistic tasks on larger datasets, and the obtained models are used for getting predictions on the ETHOS dataset, which we chose as the target dataset of our work. The obtained predictions are used as sole or additional features in the subsequently performed experiments. Multiple algorithms are evaluated, including Logistic Regression, SVM, RidgeClassifier, Decision Tree, Random Forest, AdaBoost, GradBoost, Bagging. Furthermore, multiple textual representations are taken into account including Tf-Idf, Bert embeddings and BERT embeddings combined with the aforementioned additional features. Transformer-based models BERT and DistilBERT are introduced and fine-tuned on ETHOS dataset. All the obtained models are evaluated and the resulting performance metrics are compared to results obtained by the authors of the ETHOS dataset. In order to explore the remaining underlying issues, model-agnostic method LIME is used to obtain explanations for incorrect predictions. © 2023 MIPRO Croatian Society.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Image Caption Generation with Python: A Deep Learning-Based Approach 5

Image Caption Generation with Python: A Deep Learning-Based ...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Krishna, Siram Chaitanya Reddy, Painti Nagi Kirubanantham, P. School of Computing SRM Institute of Science and Technology Kattankulathur Department of Computing Technologies Chennai India School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Faculty of Engineering and Technology Department of Computing Technologies Chennai India

ISBN: (纸本)9798331518097

Today, the day-to-day generation of such a large amount of content on social media and other websites makes it impossible to maintain each image manually, so the need for automatically identifying sensitive images and filtering them out becomes inevitable. This project proposes a sensitive image classification application using automatic captions in identifying and filtering inappropriate content. Thus, the information extracted from images by ResNet50 in the caption generation process is allowed to integrate such features with a standard model through Senti-Long Short-Term Memory networks (SLSTM) for better captioning. Deep learning models bring about drastically improved quality of captions along with contextual understanding. The captions are tested rigorously with training loss metrics, BLEU scores, and sentiment analysis through the VADER Sentiment Analyzer. This system efficiently categorizes images involving violence, negativity, or dangerous situations into the class of negative. A Flask-based web application was developed for implementing a scalable and user-friendly interface, ensuring efficient content moderation. This project has illustrated the possibility of integrating deep learning techniques with natural language processing to yield beneficial content moderation, further ensuring safe and controlled online environments. © 2024 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Exploring the influence of motion estimation algorithm selection and its parameters on the quality of HEVC-encoded 4K drone footage 8

Exploring the influence of motion estimation algorithm selec...

引用

8th International Conference on Smart and Sustainable Technologies, SpliTech 2023

作者： Benjak, Jakov Hofman, Daniel University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia

ISBN: (纸本)9789532901283

Real-time video coding on UAV devices presents a challenge where all components of the system must be optimised to get the lowest latency in the video transmission. We focus on the coding and decoding part of the pipeline, where the coding side is located on the UAV and needs to also be energy efficient. Choosing the right number of reference frames which are looked at when searching for a reference block in the motion compensation in High Efficiency Video Coding (HEVC) is crucial for getting the requested video quality and maintaining the latency and power consumption in the requested range. By increasing the number of searched reference frames, the quality increases, but as a drawback, the power consumption and latency also increase. We investigated the impact of the Group of Pictures (GOP) structure and the selection of the Motion Estimation (ME) algorithm on the quality of 4K drone footage encoded using the Kvazaar implementation of the HEVC standard. We vary different encoding parameters such as ME algorithm, ME search step, number of referent frames used in ME, target bitrate, etc. After encoding, three objective quality measures were calculated for all the videos. Since there were over 4000 examined videos, everything was done remotely on a supercomputer BURA. © 2023 University of Split, FESB.

关键词： Motion estimation

来源：评论

学校读者我要写书评

暂无评论

Detection of Tea Box Orientations in Retail Shelves Images 13

Detection of Tea Box Orientations in Retail Shelves Images

引用

13th International Symposium on Image and Signal Processing and Analysis, ISPA 2023

作者： Filipovic, Branimir Sikic, Franko Kalafatic, Zoran Loncaric, Sven Subasic, Marko University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia

ISBN: (纸本)9798350315363

This paper presents a new deep learning-based method for the detection of tea box orientations in shelf images. Due to the lack of data, we had to generate synthetic datasets using the Blender rendering framework. This is done by first hand-crafting a 3D model of a retail shelf which is then stocked with a selection of various tea boxes. These 3D models of the tea boxes are obtained by the process of 3D scanning and uploading the scans to the Blender rendering software. Next, the behavior of customers is simulated using Markov chains, and therefore some tea boxes are removed from the shelf. While stocking the tea boxes they are randomly oriented for the sake of dataset variance maximization. Finally, 10000 virtual retail scenes are rendered and annotated in three different annotation sets. We train the YOLOv5 object detection model on the synthetic datasets and achieve an impressive result of up to 84.2% mean average precision. This experimental result shows high promise for future research in this field of object pose estimation in dense scene images. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Setting Standards for Personal Health Data in the Age of 5G and 6G Networks

Journal of ICT Standardization

引用

Journal of ICT Standardization 2024年第1期12卷 47-70页

作者： Koren, Ana Prasad, Ramjee University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia CTIF GLOBAL CAPSULE Department of Business Development and Technology Aarhus University Herning Denmark

Electronic health records (EHRs) play a vital role in simplifying thorough and effective patient treatment, promoting smooth exchange of information between medical professionals, and enhancing the process of making clinical decisions. With the increasing adoption of sensor-embedded smart wearables and home automation devices, new opportunities arise for innovative solutions in various sectors, such as eHealth. In the age of 5G and 6G, the potential of utilizing user-collected health data becomes vast, promising significant improvements in people’s health and well-being. Realizing continuous healthcare access takes a step closer to reality by equipping EHRs to effectively store and interpret data collected by these sensors. This would result in personalized medical services that adhere to standardized practices. This paper presents a comprehensive review of contemporary advancements in the realm of standardization methods aimed at managing personal health data. The study delves into an extensive analysis of state-of-the-art solutions that have emerged to address the intricate challenges associated with the harmonization and uniformity of personal health information. By systematically examining these cutting-edge approaches, the review elucidates the diverse strategies employed to establish a cohesive framework for organizing, storing, and exchanging personal health data. Furthermore, the review critically evaluates the effectiveness and limitations of each solution in terms of promoting interoperability, safeguarding data privacy, and facilitating seamless data sharing among healthcare stakeholders. Furthermore, this paper then presents an approach to standardize the data by establishing semantic constraints for healthcare data types and proposing a validation procedure to ensure compliance with relevant standards and regulations. © 2024 River Publishers.

关键词： eHealth

来源：评论

学校读者我要写书评

暂无评论

Leveraging Robotics to Enhance Accessibility and Engagement in Mathematics Education for Vision-Impaired Students 5

Leveraging Robotics to Enhance Accessibility and Engagement ...

引用

5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications, HORA 2023

作者： Al Omoush, Muhammad H. Mehigan, Tracey Dublin City University Faculty of Engineering and Computing Dublin Ireland

ISBN: (纸本)9798350337525

There is potential to use robotics in education to revolutionize teaching and learning in mathematics. This is particularly relevant for vision-impaired students, who face different challenges when accessing mathematical content. Educational robotics can potentially enhance accessibility, motivation, and engagement in mathematics for students through enjoyable and novel interactions. Students commonly experience positive interactions with educational robots during learning activities, which influences their learning motivation. Recent studies show that students with disabilities face issues related to classroom participation, lack of collaborative learning, reduced social engagement, and potential for isolation. Digital-based learning technologies have transformed how vision-impaired students engage with and learn mathematics. Leveraging robotics in mathematics teaching and learning through personalised guidelines offers considerable benefits for vision-impaired students, including enhanced engagement, multimodal learning opportunities, and improved collaboration and communication skills, which enhances the opportunities for inclusive classroom experiences. This paper outlines the role of educational robotics in inclusive education. It examines the challenges and benefits of using educational robotics in mathematics for vision-impaired students. The importance of human-robot interaction (HRI) in steering the design and functionality of educational robots and their potential use within the classroom to facilitate learning is also highlighted. © 2023 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：