检索结果-内蒙古大学图书馆

Multi-objective Feature Attribution Explanation for Explainable Machine Learning

ACM Transactions on Evolutionary Learning and Optimization 2024年第1期4卷 1-32页

作者： Wang, Ziming Huang, Changwu Li, Yun Yao, Xin Research Institute of Trustworthy Autonomous Systems Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Departmentof Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China The Advanced Cognitive Technology Lab Huawei Technologies Co. Ltd Shanghai201206 China School of Computer Science University of Birmingham BirminghamB15 2TT United Kingdom

The feature attribution-based explanation (FAE) methods, which indicate how much each input feature contributes to the model's output for a given data point, are one of the most popular categories of explainable machine learning techniques. Although various metrics have been proposed to evaluate the explanation quality, no single metric could capture different aspects of the explanations. Different conclusions might be drawn using different metrics. Moreover, during the processes of generating explanations, existing FAE methods either do not consider any evaluation metric or only consider the faithfulness of the explanation, failing to consider multiple metrics simultaneously. To address this issue, we formulate the problem of creating FAE explainable models as a multi-objective learning problem that considers multiple explanation quality metrics simultaneously. We first reveal conflicts between various explanation quality metrics, including faithfulness, sensitivity, and complexity. Then, we define the considered multi-objective explanation problem and propose a multi-objective feature attribution explanation (MOFAE) framework to address this newly defined problem. Subsequently, we instantiate the framework by simultaneously considering the explanation's faithfulness, sensitivity, and complexity. Experimental results comparing with six state-of-The-Art FAE methods on eight datasets demonstrate that our method can optimize multiple conflicting metrics simultaneously and can provide explanations with higher faithfulness, lower sensitivity, and lower complexity than the compared methods. Moreover, the results have shown that our method has better diversity, i.e., it provides various explanations that achieve different tradeoffs between multiple conflicting explanation quality metrics. Therefore, it can provide tailored explanations to different stakeholders based on their specific requirements. © 2024 Copyright held by the owner/author(s). Publication rights licensed

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Stock and Cryptocurrency Price Prediction using SVM and LSTM Algorithm 4

Stock and Cryptocurrency Price Prediction using SVM and LSTM...

引用

4th IEEE Asian Conference on Innovation in technology, ASIANCON 2024

作者： Deepak Kumar, A. Christopher, Daniel Guru Dikshit, S. Balamurali, A. St. Joseph's Institute of Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350354218

Businesses can endeavour to raise money through multiple avenues. One such would be to realize the value of other business success. Purchasing and selling ownership shares in publicly traded corporations on the stock market. These prices are influenced by various factors such as investor sentiment, economic demography, and company performance. To purchase when low and sell when high or to cash in on company dividends both profit investors. The Stock Market is the backbone of many investment portfolios and impacts the global economy. The concept of data handling and interpreting data has been reshaped by machine learning[13]. This allows for computers to extract large volumes of data, judge, or predict without requiring explicit programming. This abstract covers a broad review of machine learning algorithms and their applications in different fields. A few examples include supervised, unsupervised, and reinforcement learning. In supervised learning, a model is focused on labelled data to train it to predict or categorise new data. This project uses Random Forest, LSTM, and SVM to predict the price of stocks and cryptocurrencies. The prediction graphs can be viewed on the Streamlit website. The project is anticipated to have a 97.6% accuracy Rate. © 2024 IEEE.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Sound Sight-A TTS and STT Integrated Model Using OCR and ASR

Sound Sight-A TTS and STT Integrated Model Using OCR and ASR

引用

2024 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical systems, ICSES 2024

作者： Deepak Kumar, A. Anush Varath, S. Aravindan, R. Balamurali, A. St. Joseph's Institute of Technology Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543617

This paper discusses the feasibility of using an automated pipeline that is text-to-speech and speech-to-text, which will be powered with the usage of Optical Character Recognition (OCR) that may be used for extracting text from images or documents and Automatic Speech Recognition (ASR) which can be used for the benefit of using algorithms and machine learning techniques to give more accurate results. The paper also explores the system architecture, with also a look at the methodology as it implements the technologies. © 2024 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Interference Aware Underlay D2D Multicasting through Effective Use of Under-utilized RBs 2024

Interference Aware Underlay D2D Multicasting through Effecti...

引用

16th International Conference on Contemporary Computing, IC3 2024

作者： Mukherjee, Poulomi De, Tanmay Department of Computer Science & Engineering National Institute of Technology West Bengal Durgapur713209 India

ISBN: (纸本)9798400709722

Conducting simultaneous transmissions of Device to Device (D2D) multicasting alongside cellular network communications in underlay mode has the potential to enhance spectral efficiency. Such multicasting in underlay mode requires appropriate interference mitigation techniques to maintain the quality of network access services. This is due to the mutual interference between the formulated clusters and the ongoing cellular transmissions. Therefore, appropriate resource allocation is an important research concern. This paper addresses an effective strategy of opportunistic spectrum reuse to admit D2D multicasting in underlay cellular networks. The proposed interference-aware D2D multicasting deals with appropriate resource allocation together with cluster formation through efficient cluster-head (CH) selection. The joint problem is mathematically formulated through integer linear programming (ILP). Considering the hardness of the problem an elegant greedy solution has been proposed. The merit of the proposed solution is established through simulations conducted in NS2 showing acceptable service coverage and network throughput. A comparative study with other well-known approaches is also presented. © 2024 Copyright held by the owner/author(s).

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Research on Named Entity Recognition in the Diabetes Field Based on Improved LEBERT 7

Research on Named Entity Recognition in the Diabetes Field B...

引用

7th International Conference on computer Information science and Application technology, CISAT 2024

作者： Li, Zhuli Yao, Feng Huang, Yucheng Li, Dianzhen Wuhan Institute of Technology School of Computer Science & Engineering School of Artificial Intelligence Wuhan China

ISBN: (纸本)9798350375107

The complexity of medical terminologies, variations in professional terms, and the presence of nested entities in the diabetes domain pose challenges for named entity recognition, which has been a focal point of research in intelligent medicine. To address these issues, this study proposes the Lexicon and Syntax Information Enhanced BERT (SLEBERT) model for named entity recognition in the diabetes domain. Building upon the LEBERT model, the SLEBERT model introduces contextual features and syntactic information instances. The SLEBERT model matches the sentence with external medical dictionaries to construct character-word pairs and analyzes syntactic information based on characters. The lexicon-syntax adapter then integrates character information vectors, word information vectors, and syntactic information vectors for feature fusion, which is subsequently incorporated between the initial layers of BERT. Additionally, a diabetes domain corpus dataset is constructed and annotated using the BIOE sequence labeling method. On the self-built DiaNer dataset and the CCKS2019 Chinese electronic medical record dataset, SLEBERT outperforms other comparative models in this study, demonstrating superior entity recognition performance and validating the effectiveness of this approach. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

LSTM based Sentiment Analysis of Financial News

引用

SN computer science 2023年第5期4卷 584页

作者： Sharaff, Aakanksha Chowdhury, Tushin Roy Bhandarkar, Sakshi Department of Computer Science and Engineering National Institute of Technology Raipur Chhattisgarh Raipur India

Sentiment analysis is the contextual mining of textual data representations and information sources that helps us to identify and extract subjective data in a data supply/source and facilitates corporations to recognize the social emotion being expressed by their brand, services or products while having a close look at online conversations. In the context of finance, sentiment could unfold across multiple structures, which include corporate and reviews of analysts, news headlines, and blogs that can be categorized as financial. This paper went on an in-depth analysis of the data source to analyze the way through which sentiments are expressed in textual data. In the financial arena, sentiment can be spread across multiple systems, inclusive of agency and corporate reviews, news headlines, etc. Capturing hidden sentiment/emotions and the contagion procedure, this paper used certain methodologies and is involved in the financial sentiment news analysis that enhances the textual data in news reports for sentiment analysis at the most minute level. The dataset aims to predict the score of the sentiment of the news headlines. As an implementation, the processes of reading and visualizing the dataset to make a comprehensive analysis, and text processing to convert text to lower-case and eliminate punctuation/symbols from words, implementing Doc2Vec and measuring the distance between two vectors, creating the long short-term model (LSTM) model, evaluating the built model to check the performance using metrics like accuracy and loss and finally validating the model by testing it over the test dataset. In this research, the classification of sentiment from financial news by employing a combination of Doc2vec with LSTM, Bidirectional LSTM (Bi-LSTM) and stacked Bi-LSTM has been done. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Financial news LSTM Sentiment analysis Social emotion

来源：评论

学校读者我要写书评

暂无评论

Improving predictive performance in e-learning through hybrid 2-tier feature selection and hyper parameter-optimized 3-tier ensemble modeling

引用

International Journal of Information technology (Singapore) 2024年第8期16卷 5429-5456页

作者： Tirumanadham, N S Koti Mani Kumar S, Thaiyalnayaki M, Sriram Department of Computer Science and Engineering Bharath Institute of Higher Education and Research Department of Information Technology Bharath Institute of Higher Education and Research

The paper presents a new feature selection technique developed in detail here to address improved prediction accuracy not only for the machine-learning algorithm but also for student outcomes related to the learning environment. It is a feature selection methodology, combined with Ridge (L2) regularization and the Boruta optimization clubbed in a two-tier approach named as BR²-2 T, which results in the most effective feature set selection process. To be specific, the present study developed a three-tier ensemble model that consolidates Random Forest with Bayesian Optimization methods Support Vector Machine with random search, and Gradient Boosting with Particle Swam Optimization (PSO) for performing hyperparameter tuning. Some sophisticated techniques applied in this study are the use of Z-score normalization so as to standardize variable distribution, Synthetic Minority Over-sampling Technique (SMOTE) for handling imbalanced dataset and Multiple Imputation by Chained Equations (MICE) for handling messiness among items. The evaluation of the projected approach shows improvement in accuracies, which can attain a maximum of 98.74%. The comparative assessment of the proposed strategy unveiled that predictive modelling of educational outcomes is effectively enhanced more than common methods of topics related to education. The research sharpens student success strategies and educational pedagogy by using powerful algorithms in its direction of decision-making. © Bharati Vidyapeeth's institute of computer Applications and Management 2024.

关键词： Boruta optimization Ensemble model Gradient boosting L2 regularization Multiple imputation by chained equations (MICE) Random forest Support Vector Machine Synthetic minority over-sampling technique (SMOTE)

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Approach for Parkinson' s disease Prediction through Quantum Computing Techniques 2

Machine Learning Approach for Parkinson' s disease Predictio...

引用

2nd IEEE International Conference on Advances in Information technology, ICAIT 2024

作者： Arpitha, K. Lohith, H.R. Nemmadi, H.K. Sujan, K. Yashwanth, C.L. Dept. of Computer Science Engineering BGS Institute of Technology Adichunchanagiri University Mandya India

ISBN: (数字)9798350383867

ISBN: (纸本)9798350383867

One of the most serious illnesses that may impact the neurological system in humans is dementia, of which Parkinson' s disease is a major subtype. Patients with Parkinson' s disease have significant behavioural changes. The suggested system uses quantum computing and machine learning to build a speech signal-based method for Parkinson' s disease prediction. Four main machine learning methods are used in this prediction: Artificial Neural Networks, KNN, Decision Trees, and Naïve Bayes. Furthermore, different ensemble learning models are used to combine numerous models and improve the accuracy. An industry standard dataset from the UCI Machine Learning Repository is used to assess the model' s performance. Ensemble models have proven to outperform extremely efficient techniques like neural networks in terms of accuracy. A multi-layer perceptron, a very significant machine learning optimization technique, is incorporated into the suggested system. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Hybrid Optimisation with Black Hole Algorithm for Improving Network Lifespan

引用

Intelligent Automation & Soft Computing 2023年第2期35卷 1873-1887页

作者： S.Siamala Devi Chandrakala Kuruba Yunyoung Nam Mohamed Abouhawwash Department of Computer Science and Engineering Sri Krishna College of TechnologyCoimbatore641042India Department of Computer Science and Engineering Vignan Nirula Institute of Technology and ScienceGuntur522009India Department of Computer Science and Engineering Soonchunhyang UniversityAsan31538Korea Department of Mathematics Faculty of ScienceMansoura UniversityMansoura35516Egypt Department of Computational Mathematics Scienceand Engineering(CMSE)Michigan State UniversityEast LansingMI48824USA

Wireless sensor networks(WSNs)are projected to have a wide range of applications in the *** fundamental problem with WSN is that it has aﬁnite *** a network is a common strategy for increasing the life-time of WSNs and,as a result,allowing for faster data *** cluster-ing algorithm’s goal is to select the best cluster head(CH).In the existing system,Hybrid grey wolf sunﬂower optimization algorithm(HGWSFO)and optimal clus-ter head selection method is *** does not provide better competence and out-put in the ***,the proposed Hybrid Grey Wolf Ant Colony Optimisation(HGWACO)algorithm is used for reducing the energy utilization and enhances the lifespan of the *** hole method is used for selecting the cluster heads(CHs).The ant colony optimization(ACO)technique is used toﬁnd the route among origin CH and *** open cache of nodes,trans-mission power,and proximity are used to improve the CH *** grey wolf optimisation(GWO)technique is the most recent and well-known optimiser module which deals with grey wolves’hunting activity(GWs).These GWs have the ability to track down and encircle *** GWO method was inspired by this hunting *** proposed HGWACO improves the duration of the net-work,minimizes the power consumption,also it works with the large-scale *** HGWACO method achieves 25.64%of residual energy,25.64%of alive nodes,40.65%of dead nodes also it enhances the lifetime of the network.

关键词： Energy efﬁciency power consumption lifespan of the network black hole method ant colony optimisation routing and cluster heads(CHs)

来源：评论

学校读者我要写书评

暂无评论

Exploration and Evaluation of Congestion Control Algorithms for Data Center Networks

引用

SN computer science 2023年第5期4卷 1-15页

作者： Nandhini, C. Gupta, Govind P. Department of Computer Science and Engineering National Institute of Technology Tiruchirappalli India Department of Information Technology National Institute of Technology Raipur India

In recent years, Data Center Networks (DCN) have become a very popular platform for hosting various online services and applications, such as e-commerce, social networking, large-scale computing, and web searching due to their cost-effective and efficient service provisioning. DCN, online services, and applications typically require minimal latency in any information exchange. Moreover, compared with Internet traffic, the nature of traffic in DCN is bursty, delay, and throughput sensitive. Due to this reason, state-of-the-art TCP congestion control algorithms perform poorly and suffer from the problems, such as TCP Incast, TCP Outcast, Pseudo-Congestion Effect, Buffer pressure, and Queue build-up. For improving the performance of DCN, in recent years, various congestion control algorithms have been proposed for DCN. This paper summarizes the reason why the state-of-the-art TCP congestion control algorithm performs poorly and presents an overview of the recently proposed congestion control algorithms for DCN, followed by a comparative summary of their performance. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Congestion control Data center Data center networks TCP

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：