检索结果-内蒙古大学图书馆

A Study on the Explainability of Thyroid Cancer Prediction:SHAP Values and Association-Rule Based Feature Integration Framework

引用

computers, Materials & Continua 2024年第5期79卷 3111-3138页

作者： Sujithra Sankar S.Sathyalakshmi Department of Computer Applications Hindustan Institute of Technology and ScienceChennaiTamil NaduIndia Department of Computer Engineering Hindustan Institute of Technology and ScienceChennaiTamil NaduIndia

In the era of advanced machine learning techniques,the development of accurate predictive models for complex medical conditions,such as thyroid cancer,has shown remarkable *** predictivemodels for thyroid cancer enhance early detection,improve resource allocation,and reduce ***,the widespread adoption of these models in clinical practice demands predictive performance along with interpretability and *** paper proposes a novel association-rule based feature-integratedmachine learning model which shows better classification and prediction accuracy than present *** study also focuses on the application of SHapley Additive exPlanations(SHAP)values as a powerful tool for explaining thyroid cancer prediction *** the proposed method,the association-rule based feature integration framework identifies frequently occurring attribute combinations in the *** original dataset is used in trainingmachine learning models,and further used in generating SHAP values *** the next phase,the dataset is integrated with the dominant feature sets identified through association-rule based *** new integrated dataset is used in re-training the machine learning *** new SHAP values generated from these models help in validating the contributions of feature sets in predicting *** conventional machine learning models lack interpretability,which can hinder their integration into clinical decision-making *** this study,the SHAP values are introduced along with association-rule based feature integration as a comprehensive framework for understanding the contributions of feature sets inmodelling the *** study discusses the importance of reliable predictive models for early diagnosis of thyroid cancer,and a validation framework of *** proposed model shows an accuracy of 93.48%.Performance metrics such as precision,recall,F1-score,and the area un

关键词： Explainable AI machine learning clinical decision support systems thyroid cancer association-rule based framework SHAP values classification and prediction

来源：评论

学校读者我要写书评

暂无评论

Seizure Prediction on EEG Signals using Feature Augmentation based Multi Model Ensemble

引用

Recent Patents on engineering 2025年第1期19卷 41-49页

作者： Anandaraj, A. Alphonse, P.J.A. Department of Computer Science and Engineering Sri Eshwar College of Engineering Coimbatore India Department of Computer Applications National Institute of Technology Tamilnadu Tiruchirappalli India

Background: Epilepsy is a neurological disorder that leads to seizures. This occurs due to excessive electrical discharge by the brain cells. An effective seizure prediction model can aid in improving the lifestyle of epilepsy patients. After analyzing various patents related to seizure prediction, it is observed that monitoring electroencephalography (EEG) signals of epileptic patients is an important task for the early diagnosis of seizures. Objective: The main objective of this paper is to assist epileptic patients to enhance their way of living by predicting the seizure in advance. Methods: This paper builds a feature augmentation-based multi-model ensemble-based architecture for seizure prediction. The proposed technique is divided into 2 broad categories;feature augmentation and ensemble modeling. The feature augmentation process builds temporal features while the multi-model ensemble has been designed to handle the high complexity levels of the EEG data. The first phase of the multi-model ensemble has been designed with heterogeneous classifier models. The second phase is based on the prediction results obtained from the first phase. Experiments were performed using the seizure prediction dataset from the University Hospital of Bonn. Results: Comparison indicates 98.7% accuracy, with improvement of 5% from the existing model. High prediction levels indicate that the model is highly capable of providing accurate seizure predictions, hence ensuring its applicability in real time. Conclusion: The result of this paper has been compared with existing methods of predicting seizures and it indicated that the proposed model has better enhancement in the accuracy levels. © 2025 Bentham science Publishers.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Parkinson’s disease Severity Detection Based on Fused features with tuned Residual Deep Convolutional Neural Network

引用

Multimedia Tools and applications 2025年第14期84卷 13111-13131页

作者： Vishwakarma, Gagan Nandanwar, Amit Kumar Thakur, Ghanshyam Singh Computer Science & Engineering Indian Institute of Information Technology MP Bhopal India Computer Science & Engineering Maulana Azad National Institute of Technology MP Bhopal India Department of Computer Applications Maulana Azad National Institute of Technology MP Bhopal India

Parkinson's disease (PD) diagnosis involves the assessment of a variety of motor and non-motor symptoms. To accurately diagnose PD, it is necessary to differentiate its symptoms from those of other conditions. During the diagnosis process, clinicians must consider gait abnormalities as one of the key symptoms. However, relying solely on clinical judgment for the diagnosis of PD may have limitations. In this work, we have proposed a Multikernel SVM (MK-SVM) based Parkinson's severity analysis method. Initially, an adaptive median filter is used to reduce the noise in the gait image. Then, to extract the features, we propose a dual-layer method that employs stacked autoencoders (SAEs) for local feature extraction and residual deep convolution neural networks (RDCNN) for global feature extraction. The hyper-parameters from the RDCNN are optimized using the Enhanced walrus optimization algorithm (EWO). In the final step, we use MK-SVM to classify PD according to severity. This model is evaluated using the GAIT-IT and GAIT-IST datasets. With the help of Python, we evaluated the proposed model in terms of precision, accuracy, recall, specificity, and F-scores, and the model achieved a maximum accuracy of 97.3%. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Classifying the state of knowledge-based question answering: patterns, progress, and prospects

引用

International Journal of computers and applications 2025年第1期47卷 93-105页

作者： Sharma, Sumit Jain, Sarika Tiwari, Mukesh Kumar lal, Sohan Department of Computer Science and Engineering Quantum University Roorkee Uttarakhand Roorkee India Department of Computer Applications National Institute of Technology Haryana Kurukshetra India

In response to inquiries posed in natural languages, question-answering systems (QASs) produce responses. The capabilities of early QASs are limited because they were designed for certain domains. The current generation of QASs focuses on the types of inquiries that users typically enquire, the features of the data sources that are used, and the formats in which the right responses are produced. Many QASs have been developed since the 1960s when research on the subject first started. There are several cutting-edge frameworks for answering questions that can be used to categorize the QA system according to a number of criteria. The necessity for an extensive survey on QASs comes naturally in order to determine the future focus of research in this field. In this study, QASs are analyzed by the number of criteria used to categorize question-answering (QA) and provide the dimension using a variety of standards to cover all the QAS. We describe the current state of research in each QAS area and make recommendations for its future direction. Based on the dimension criteria outlined in the research methodology, research studies are chosen for additional review and comparison. Constructed some criteria to classify the QASs, we provide the classification and go over the benefits and drawbacks of each class's QAS, as well as the problems with the corresponding study. Results are detailed summarized and compared. As a result of this study, a general architecture of frameworks for QA over knowledge base (KB) is drawn in addition to the comparative analysis of systems, serving as a reference for academics working on related topics. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Chi-Square Target Encoding for Categorical Data Representation: A Real-World Sensor Data Case Study

引用

SN computer science 2025年第3期6卷 1-22页

作者： Anitha, M. Savarimuthu, Nickolas Bhanu, S. Mary Saira Department of Computer Applications National Institute of Technology Trichy India Department of Computer Science and Engineering National Institute of Technology Trichy India

Feature engineering is critical for improving machine learning performance (ML), especially when handling categorical data. Traditional encoding methods, such as one-hot and label encoding, often result in challenges like high dimensionality and loss of category significance. To address these limitations, this study introduces a novel Chi-Square Target Encoding (CSTE) approach, which transforms categorical variables into numerical representations by leveraging the chi-square statistic to evaluate the association between categories and the target variable, preserving information effectively. Unlike conventional techniques, CSTE uses Chi-square test values to generate target-based encoded representations, ensuring reliable transformation without information loss. Comprehensive empirical evaluations were conducted on various datasets for binary classification, showcasing CSTE’s superiority over methods like regularized target encoding, basic target encoding, and fuzzification. A case study on real-world sensor data further validated its efficiency and scalability for large-scale data-driven applications. The proposed CSTE method achieved an average increment rate of 3.9447%, surpassing one-hot and fuzzification (1.0430%) and regularized target encoding (1.8308%). Classification outcomes demonstrated F1-scores exceeding 0.90 and AUC values nearing 0.99 across diverse datasets, highlighting its robustness. Furthermore, the reduced dimensionality significantly enhanced inference time while maintaining high accuracy. The CSTE method offers a robust framework for categorical data transformation, addressing limitations of traditional encoding techniques. It improves the interpretability and efficiency of categorical data representation, boosting ML performance. This innovative approach is well-suited for applications across various domains involving categorical data. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Categorical data encoding Chi- Square statistics Data distribution preservation Feature engineering Machine learning Target encoding

来源：评论

学校读者我要写书评

暂无评论

Aggregated Relative Similarity (ARS): a novel similarity measure for improved personalised learning recommendation using hybrid filtering approach

引用

Multimedia Tools and applications 2025年第18期84卷 18733-18780页

作者： Pal, Saurabh Dutta Pramanik, Pijush Kanti Choudhury, Prasenjit Department of Computer Science and Engineering National Institute of Technology Durgapur West Bengal Durgapur India School of Computer Applications and Technology Galgotias University Uttar Pradesh Greater Noida India Department of Computer Science and Engineering School of Engineering Brainware University Kolkata India

To improve the effectiveness of online learning, the learning materials recommendation is required to be personalised to the learner material recommendations must be personalized to learners. The existing approaches are ineffective in recommending learning materials according to the learner’s learning situation, the learner’s internal learning characteristics, and fitting to the learner’s preferences and learning suitability. In this work, we propose a hybrid recommendation approach cthat combines the advantages of knowledge-based and collaborative filtering approaches. The knowledge-based filtering approach enables us to find the most suitable learning materials that fit the learner’s situation and learning state by appropriately mapping the learners’ contexts with the metadata of the learning materials. We use collaborative filtering to refine learning material selection by considering learner-learner similarity. We propose a novel similarity measure, aggregated relative similarity (ARS), to determine similar learners based on implicit learner characteristics such as education, knowledge, cognitive ability, and learning style. The experimental results show that the proposed ARS algorithm outperforms popular similarity measures such as Jaccard similarity, Sorensen-Dice coefficient, and overlap coefficient. The overall recommendation performance of the proposed hybrid approach attained mean absolute error (MAE) and root mean square error (RMSE) of 0.681 and 0.9198, respectively. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

An Intelligent System to Predict Diabetic Retinal Diseases Based on Diabetic Attributes

引用

SN computer science 2024年第6期5卷 755页

作者： Eswari, M. Shanmuga Balamurali, S. Ramkumar, K. Department of Computer Applications Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Rajalakshmi Institute of Technology

The most common illness among individuals and the general population in the medical field is diabetes. This is coupled with a careful diabetic retinal that has no signal. The historical record offers unrecoverable insight. To avoid a hallucinatory dispute at the time, early detection of retinal illness is necessary. Positively, the results of diabetic patients have been able to foretell the threat of diabetic-related eyeailments such as Diabetic Retinopathy, Glaucoma, and Diabetic Macular Edema.A diabetic retinal disease prediction exemplary have been created by via system learning rules, such as narrow neural network, bilayered neural network, wide neural network and trilayer neural network, in close proximity to the onset of the disease. The trilayer neural network with various activation functions accuracy based results servedas the foundation for the final model that evaluates real time data to predict the likelihood of *** learner representation accuracy of 0.95and the evaluation phase accuracy of 0.94 are *** ideal diabetic features-based diabetic retinal system outcome occupies a crucial place in the medical decision support system to make prompt assessments in the field of E-healthcare system to connect ophthalmology field to eradicate insulin based eye problems at the time of regular assessment. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Boosting computer-aided Design Diabetic Mellitus Ensemble Machine Learner Retinal eye Disease

来源：评论

学校读者我要写书评

暂无评论

A hybrid encryption for secure data deduplication the cloud

引用

International Journal of Cloud Computing 2023年第2-4期12卷 295-307页

作者： Ebinazer, Silambarasan Elkana Savarimuthu, Nickolas Bhanu, S. Mary Saira Department of Computer Science and Engineering Cyber Security IIIT Kottayam India Department of Computer Applications National Institute of Technology Tiruchirappalli India Department of Computer Science and Engineering National Institute of Technology Tiruchirappalli India

Cloud computing (CC) is a cost-effective platform for users to store their data on the internet rather than investing in additional devices for storage. Data deduplication (DD) defines a process of eliminating redundant data in the cloud by leaving a single copy. This paper provides a hybrid encryption scheme using genetic algorithm (GA) and attribute-based encryption (ABE) to ensure secure data transmission in the cloud and preventing the data deduplication by using message locked encryption (MLE). The hybrid encryption scheme provides a secure data transmission in the cloud. From the simulation output, the time requirement is reduced up to 94% and consumes 43% bandwidth compared with existing works. The proposed method tends to be quite effective in reducing storage and bandwidth costs. Copyright © 2023 Inderscience Enterprises Ltd.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

A Novel Eccentric Intrusion Detection Model Based on Recurrent Neural Networks with Leveraging LSTM

引用

computers, Materials & Continua 2024年第3期78卷 3089-3127页

作者： Navaneetha Krishnan Muthunambu Senthil Prabakaran Balasubramanian Prabhu Kavin Kishore Senthil Siruvangur Kavitha Chinnadurai Jehad Ali Department of Computer Applications Karpagam College of EngineeringCoimbatore641032India Department of Computer Science and Engineering Karpagam College of EngineeringCoimbatore641032India Department of Data Science and Business Systems College of Engineering and TechnologySRM Institute of Science and TechnologyKattankulathur603203India Department of Computer Science and Engineering Sathyabama Institute of Science and TechnologyJeppiaar NagarRajiv Gandhi SalaiChennai600119India Department of AI Convergence Network Ajou UniversitySuwon16499Korea

The extensive utilization of the Internet in everyday life can be attributed to the substantial accessibility of online services and the growing significance of the data transmitted via the ***,this development has expanded the potential targets that hackers might *** adequate safeguards,data transmitted on the internet is significantly more susceptible to unauthorized access,theft,or *** identification of unauthorised access attempts is a critical component of cybersecurity as it aids in the detection and prevention of malicious *** research paper introduces a novel intrusion detection framework that utilizes Recurrent Neural Networks(RNN)integrated with Long Short-Term Memory(LSTM)*** proposed model can identify various types of cyberattacks,including conventional and distinctive *** networks,a specific kind of feedforward neural networks,possess an intrinsic memory *** Neural Networks(RNNs)incorporating Long Short-Term Memory(LSTM)mechanisms have demonstrated greater capabilities in retaining and utilizing data dependencies over extended *** such as data types,training duration,accuracy,number of false positives,and number of false negatives are among the parameters employed to assess the effectiveness of these models in identifying both common and unusual *** are utilised in conjunction with LSTM to support human analysts in identifying possible intrusion events,hence enhancing their decision-making capabilities.A potential solution to address the limitations of Shallow learning is the introduction of the Eccentric Intrusion Detection *** model utilises Recurrent Neural Networks,specifically exploiting LSTM *** proposed model achieves detection accuracy(99.5%),generalisation(99%),and false-positive rate(0.72%),the parameters findings reveal that it is superior to state-of-the-art techniques.

关键词： Cybersecurity intrusion detection machine learning leveraging long short-term memory(LLSTM) CICIDS2019 dataset innovative cyberattacks

来源：评论

学校读者我要写书评

暂无评论

Learner’s intention analysis to mitigate the cold start problem in personalized learning recommendation systems

引用

Multimedia Tools and applications 2024年 1-55页

作者： Pal, Saurabh Pramanik, Pijush Kanti Dutta Choudhury, Prasenjit Department of Computer Science and Engineering School of Engineering Brainware University Kolkata India School of Computer Applications and Technology Galgotias University Uttar Pradesh Greater Noida India Department of Computer Science & amp Engineering National Institute of Technology Durgapur West Bengal India

Understanding the learner’s requirements and status is important for recommending relevant and appropriate learning materials to the learner in personalized learning. For this purpose, the learning recommendation system must include the learner’s contextual and behavioral information. However, in the case of cold start problems, the system has no prior information about a new user. Moreover, the system cannot acquire the required information instantly because the automated and instantaneous acquisition of this information is not trivial. This hinders the personalized recommendation for a new learner. Engaging the learner in a questionnaire session may allow the learning system to acquire most of the required information. However, this makes the learner uninterested and disengaged with the system. To address this, we propose a recommendation system for personalized learning that mitigates the cold start problem by acquiring important information by analyzing the learner’s questions submitted to the learning system. By analyzing the meta-information of a question, we identified the learner’s exact requirements, domain and topic knowledge, and expected knowledge gain from the learning material. Ontology, BiLSTM, and context mapping are used for learner modelling and topic classification, and identifying suitable learning materials, respectively. Furthermore, MCDM techniques are used for ranking the learning materials. Our experimental assessment and validation suggest that the information acquired based on our proposed approach is sufficient for understanding a new learner’s needs and suitability and, accordingly, for recommending suitable personalized learning materials. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Recommender systems

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

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

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：