检索结果-内蒙古大学图书馆

Machine Learning Approaches for Mental Illness Detection on Social Media: A Systematic Review of Biases and Methodological Challenges

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Cao, Yuchen Dai, Jianglai Wang, Zhongyan Zhang, Yeyubei Shen, Xiaorui Liu, Yunchong Tian, Yexin Khoury College of Computer Science Northeastern University United States Department of EECS University of California Berkeley United States Center for Data Science New York University United States School of Engineering and Applied Science University of Pennsylvania United States Georgia Institute of Technology College of Computing United States

The global increase in mental illness requires innovative detection methods for early intervention. Social media provides a valuable platform to identify mental illness through user-generated content. This systematic review examines machine learning (ML) models for detecting mental illness, with a particular focus on depression, using social media data. It highlights biases and methodological challenges encountered throughout the ML lifecycle. A search of PubMed, IEEE Xplore, and Google Scholar identified 47 relevant studies published after 2010. The Prediction model Risk Of Bias ASsessment Tool (PROBAST) was utilized to assess methodological quality and risk of bias. The review reveals significant biases affecting model reliability and generalizability. A predominant reliance on Twitter (63.8%) and English-language content (over 90%) limits diversity, with most studies focused on users from the United States and Europe. Non-probability sampling methods (approximately 80%) limit representativeness. Only 23% of studies explicitly addressed linguistic nuances like negations, crucial for accurate sentiment analysis. Inconsistent hyperparameter tuning was observed, with only 27.7% properly tuning models. About 17% did not adequately partition data into training, validation, and test sets, risking overfitting. While 74.5% used appropriate evaluation metrics for imbalanced data, others relied on accuracy without addressing class imbalance, potentially skewing results. Reporting transparency varied, often lacking critical methodological details. These findings highlight the need to diversify data sources, standardize preprocessing protocols, ensure consistent model development practices, address class imbalance, and enhance reporting transparency. By overcoming these challenges, future research can develop more robust and generalizable ML models for depression detection on social media, contributing to improved mental health outcomes globally. © 2024, CC BY.

关键词： Adversarial machine learning

Breast cancer detection and classification: A study on the specification and implementation of multilayer perceptron analog artificial neural networks

学校读者我要写书评

暂无评论

computers in Biology and Medicine 2025年 190卷 110060-110060页

作者： Silas, Koagne Longpa T. Djimeli-Tsajio Alain, B. Thierry, Noulamo Jean-Pierre, Lienou T. Ejuh, Geh Wilson Research Unit of Automation and Applied Computer Science URAIA IUT-FV University of Dschang P.O. Box 134 Bandjoun Cameroon Department of Telecommunication and Network Engineering IUT-FV University of Dschang P.O. Box 134 Bandjoun Cameroon Department of Computer Engineering IUT-FV University of Dschang P.O. Box 134 Bandjoun Cameroon Department of General and Scientific Studies IUT-FV University of Dschang P.O. Box 134 Bandjoun Cameroon Department of Electrical and Electronic Engineering National Higher Polytechnic Institute University of Bamenda P. O. Box 39 Bambili Cameroon

Breast cancer is a leading cause of mortality worldwide. Screening therefore remains the best defense against this disease, highlighting the need for accurate and efficient diagnostic methods. Previous authors addressed this issue by implementing digital hardware with low processing speed and high power consumption constraints. Moreover, a similar approach was developed focusing on implementing an off-chip learning on-chip inference methodology resulting in a high complexity of the proposed systems and complex communication between modules regarding the operations performed by each circuit. This research presents a fully integrated multilayer perceptron analog artificial neural network, classifying breast tumor cells as benign or malignant. A novel analog network architecture and fundamental analog circuit blocks realized by dint of the 90 nm complementary metal-oxide semiconductor technology are developed in this work for breast cancer analysis. The model's reliability findings are confirmed through comprehensive evaluation metrics, achieving sensitivity, specificity, accuracy, F1-Scoree, Intersection-over-Union, and Matthew's correlation coefficient of 0.9668, 0.9869, 0.9800, 0.9708, 0.9433, and 0.9556, respectively. Monte Carlo analysis further validates a 98.00 % accuracy on the Wisconsin Breast Cancer Dataset, with a power consumption of 23.06 mW, using 122 MB of memory while operating at a power supply rail of ±900 mV per analog circuit component and 27 °C computational unit temperature. These findings suggest a significant opportunity for integrating the proposed model into clinical practice, thereby improving patient care in medical diagnostics. © 2025 Elsevier Ltd

关键词： Oxide semiconductors

Hyperspectral Image Denoising Using Non-convex Local Low-rank and Sparse Separation with Spatial-Spectral Total Variation Regularization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Peng, Chong Liu, Yang Chen, Yongyong Wu, Xinxin Cheng, Andrew Kang, Zhao Chen, Chenglizhao Cheng, Qiang College of Computer Science and Technology Qingdao University China The College of Computer Science Harbin Institute of Technology The School of Computer Science and Engineering University of Electronic Science and Technology of China China Department of Computer Science University of Kentucky Department of Statistics and Applied Probability University of California Santa Barbara United States The Institute of Biomedical Informatics University of Kentucky

In this paper, we propose a novel nonconvex approach to robust principal component analysis for HSI denoising, which focuses on simultaneously developing more accurate approximations to both rank and column-wise sparsity for the low-rank and sparse components, respectively. In particular, the new method adopts the log-determinant rank approximation and a novel 2,log norm, to restrict the local low-rank or column-wisely sparse properties for the component matrices, respectively. For the 2,log-regularized shrinkage problem, we develop an efficient, closed-form solution, which is named 2,log-shrinkage operator. The new regularization and the corresponding operator can be generally used in other problems that require column-wise sparsity. Moreover, we impose the spatial-spectral total variation regularization in the log-based nonconvex RPCA model, which enhances the global piece-wise smoothness and spectral consistency from the spatial and spectral views in the recovered HSI. Extensive experiments on both simulated and real HSIs demonstrate the effectiveness of the proposed method in denoising HSIs. Copyright © 2022, The Authors. All rights reserved.

关键词： Spectroscopy

A Systematic Review of Machine Learning Approaches for Detecting Deceptive Activities on Social Media: Methods, Challenges, and Biases

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Yunchong Shen, Xiaorui Zhang, Yeyubei Wang, Zhongyan Tian, Yexin Dai, Jianglai Cao, Yuchen School of Engineering and Applied Science University of Pennsylvania United States Khoury college of computer science Northeastern University United States Center for Data Science New York University United States Georgia Institute of Technology College of Computing United States Department of EECS University of California Berkeley United States

Social media platforms like Twitter, Facebook, and Instagram have facilitated the spread of misinformation, necessitating automated detection systems. This systematic review evaluates 36 studies that apply machine learning (ML) and deep learning (DL) models to detect fake news, spam, and fake accounts on social media. Using the Prediction model Risk Of Bias ASsessment Tool (PROBAST), the review identified key biases across the ML lifecycle: selection bias due to non-representative sampling, inadequate handling of class imbalance, insufficient linguistic preprocessing (e.g., negations), and inconsistent hyperparameter tuning. Although models such as Support Vector Machines (SVM), Random Forests, and Long Short-Term Memory (LSTM) networks showed strong potential, over-reliance on accuracy as an evaluation metric in imbalanced data settings was a common flaw. The review highlights the need for improved data preprocessing (e.g., resampling techniques), consistent hyperparameter tuning, and the use of appropriate metrics like precision, recall, F1 score, and AUROC. Addressing these limitations can lead to more reliable and generalizable ML/DL models for detecting deceptive content, ultimately contributing to the reduction of misinformation on social media. © 2024, CC BY.

关键词： Adversarial machine learning

Polar, Spherical and Orthogonal Space Subdivisions for an Algorithm Acceleration: O(1) Point-in-Polygon/Polyhedron Test

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Skala, Vaclav The University of West Bohemia Faculty of Applied Sciences Department .of Computer Science and Engineering Center of Computer Graphics and Visualization Univerzitni 8 PlzenCZ 30614 Czech Republic

Acceleration of algorithms is becoming a crucial problem, if larger data sets are to be processed. Evaluation of algorithms is mostly done by using computational geometry approach and evaluation of computational complexity. However in today’s engineering problems this approach does not respect that number of processed items is always limited and a significant role plays also speed of read/write operations. One general method how to speed up an algorithm is application of space subdivision technique and usually the orthogonal space subdivision is used. In this paper non-orthogonal subdivisions are described. The proposed approach can significantly improve memory consumption and run-time complexity. The proposed modified space subdivision techniques are demonstrated on two simple problems Point-in-Convex Polygon and Point-in-Convex Polyhedron *** Codes 65D17, 65D18 © 2022, CC BY.

关键词： Computational geometry

Automatic Detection of Breast Cancer in Thermal Images using Optimal Handcrafted Features 31

学校读者我要写书评

暂无评论

Automatic Detection of Breast Cancer in Thermal Images using...

31st International Conference on computer Theory and Applications, ICCTA 2021

作者： Kadry, Seifedine Rajinikanth, Venkatesan Dasari, Susmitha Suvathi, T. Noroff University College Faculty of Applied Computing and Technology Kristiansand Norway St. Joseph's College of Engineering Omr Department of Electronics and Instrumentation Engineering Chennai600119 India St. Joseph's College of Engineering Omr Department of Computer Science and Engineering Chennai600119 India

ISBN: (纸本)9781665478540

Breast Cancer (BC) is one of the harsh illnesses which widely distress the women population and premature recognition is always preferred to reduce its harshness. BC is normally detected in hospitals using a chosen non-invasive medical imaging technique. The proposed work aims to develop and execute the machine learning supported automatic detection of the BC using breast Thermal Images (TI). In TI, examination of thermal silhouette helps to get the necessary information about the normal/cancer infected section. The developed system consist the following phases;(i) Image collection and pre-processing, (ii) Image enhancement and handcrafted feature extraction, (iii) Feature selection with Bat Algorithm, and (iv) Binary classification and validation. In this work, a 5-fold cross validation is employed to classify the test images and the detected accuracy of Fine KNN is better (>92%) compared to the alternatives. The proposed work is tested using benchmark breast TI and the outcome of this research confirms its clinical significance. © 2021 IEEE.

关键词： Image enhancement

Detection of Diabetic Macular Edema in Retinal OCT Images with Firefly-Algorithm Selected Features 31

学校读者我要写书评

暂无评论

Detection of Diabetic Macular Edema in Retinal OCT Images wi...

31st International Conference on computer Theory and Applications, ICCTA 2021

作者： Kadry, Seifedine Rajinikanth, Venkatesan Jerry, Jiny Mariam Gifta, R. Carolin Noroff University College Faculty of Applied Computing and Technology Kristiansand Norway St. Joseph's College of Engineering Omr Department of Electronics and Instrumentation Engineering Chennai600119 India St. Joseph's College of Engineering Omr Department of Computer Science and Engineering Chennai600119 India

ISBN: (纸本)9781665478540

In modern hospitals, computer algorithms are widely implemented to diagnose the disease in patients using the medical data collected using a recommended protocol. Recently, the diabetes is human group is gradually rising and the early stage can be predicted by screening the eye. Diabetic-Macular-Edema (DME) is one of the harsh diseases and the untreated disease will initiate various eye abnormalities including the vision loss. In clinics, the retinal image recorded with a chosen imaging system is considered to inspect the disease. In this work, Optical Coherence Tomography (OCT) images are considered for the assessment of the DME in the OCT images. The proposed scheme consist the following stages;(i) Image collection and resizing, (ii) Retinal section extraction using joint thresholding and segmentation, (iii) Feature extraction and dominant feature selection with Firefly-Algorithm, (iv) Binary classification and validation. In this work, 500 numbers of Normal/DME class images with pixel dimension 256 × 256 ×1 pixels are considered and a binary classification with a 5-fold cross validation is implemented. The proposed technique helped to get a detection accuracy of 95% with K-Nearest Neighbors (KNN) classifier. © 2021 IEEE.

关键词： Optical tomography

Global-Local Active Contour for Segmenting Images with Simultaneous Inhomogeneity and Texture

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Shao, Pei-Chiang Department of Applied Mathematics Chung Yuan Christian University Zhongli District Taoyuan City320314 Taiwan Undergraduate Program in College of Electrical Engineering and Computer Science Chung Yuan Christian University Zhongli District Taoyuan City320314 Taiwan

Severe intensity inhomogeneity (InH) and complex real-world textures cause great difficulties and become two important issues in image segmentation and object extraction applications. Plenty of methods are proposed to tackle these issues, however, they usually perform poorly on images possess both InH and texture. In this paper, we propose a region-based active contour model for segmenting images with simultaneous inhomogeneity and texture. A new bias field possesses both global and local smoothness properties is proposed to handle the InH issue, and a pixel inhomogeneity factor (PIF) is employed to jointly deal with InH and textured images. We also employ the total variation (TV) regularization to further smooth and denoise the transformed PIF image, making the segmentation process to have the nice properties of global convergence and independence of contour initialization. A large number of numerical experiments and comparisons are conducted to demonstrate the effective performance of the newly proposed method. © 2024, The Authors. All rights reserved.

关键词： Textures

Credit card score prediction using machine learning models: A new dataset

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Arram, Anas Ayob, Masri Albadr, Musatafa Abbas Abbood Sulaiman, Alaa Albashish, Dheeb Department of Computer Science Birzeit University Birzeit Palestine Universiti Kebangsaan Malaysia Selangor Bangi43600 Malaysia Department of Oil and Gas Engineering Basrah University for Oil and Gas Basrah61004 Iraq Selangor Bangi43600 Malaysia Computer Science Department Prince Abdullah bin Ghazi Faculty of Information and Communication Technology Al-Balqa Applied University Salt Jordan

The use of credit cards has recently increased, creating an essential need for credit card assessment methods to minimize potential risks. This study investigates the utilization of machine learning (ML) models for credit card default prediction system. The main goal here is to investigate the best-performing ML model for new proposed credit card scoring dataset. This new dataset includes credit card transaction histories and customer profiles, has been proposed and tested using a variety of machine learning algorithms, including logistic regression, decision trees, random forests, multi layer perceptron (MLP) neural network, XGBoost, and LightGBM. To prepare the data for machine learning models, we employ different data preprocessing techniques such as feature extraction, handling missing values, managing outliers, and applying data balancing methods. Experimental results demonstrate that MLP outperforms logistic regression, decision trees, random forests, LightGBM, and XGBoost in terms of predictive performance in true positive rate, achieving an impressive area under the curve (AUC) of 86.7% and an accuracy rate of 91.6%, with a recall rate exceeding 80%. These results indicate the superiority of MLP in predicting the default customers and assessing the potential risks. Furthermore, they help banks and other financial institutions in predicting loan defaults at an earlier stage. © 2023, CC BY.

关键词： Multilayer neural networks