Search or ask a question

How to turn on AMD SVM?

Support vector machine

Facial recognition system

Feature extraction

Particle swarm optimization

Perturbation (astronomy)

Answers from top 7 papers

PDF

Open Access

More filters

Papers (7)	Insight
Open access•Proceedings Article•DOI Facial Action Units intensity estimation by the fusion of features with multi-kernel Support Vector Machine Zuheng Ming, Aurélie Bugeau, Jean-Luc Rouas, Takaaki Shochi - Show less +3 more 04 May 2015 40 Citations	The result of our approach benefiting from taking advantages of the different features adapting to a multi-kernel SVM is shown to outperform the conventional methods based on the mono-type feature with single kernel SVM.
Open access•Book Chapter•DOI PSO-Based Method for SVM Classification on Skewed Data-Sets Jair Cervantes, Farid García-Lamont, Asdrúbal López, Lisbeth Rodriguez, José Sergio Ruiz Castilla, Adrian Trueba - Show less +5 more 20 Aug 2015 26 Citations	Experimental results demonstrate the ability of the proposed method to improve the performance of SVM on imbalanced data-sets.
Open access•Journal Article•DOI Intrusion Detection in KDD99 Dataset using SVM-PSO and Feature Reduction with Information Gain Harshit Saxena, Vineet Richariya - Show less +1 more 18 Jul 2014-International Journal of Computer Applications 53 Citations	The experimental result shows that SVM-PSO acquire high detection rate than regular SVM Method algorithm.
Journal Article•DOI Support Vector Machine (SVM) based prestack AVO inversion and its applications Guangcai Li, Jiachun You, Xuewei Liu - Show less +2 more 01 Sep 2015-Journal of Applied Geophysics 28 Citations	They all demonstrate the feasibility and application of SVM on both synthetic and real data.
Proceedings Article•DOI Detecting, quantifying and accessing impact of news events on Indian stock indices Ishan Verma, Lipika Dey, Hardik Meisheri - Show less +2 more 23 Aug 2017 20 Citations	The proposed system outperforms state of the art linear SVM on data from different stock indices.
Open access•Journal Article•DOI A CSP\AM-BA-SVM Approach for Motor Imagery BCI System Sahar Selim, Manal M. Tantawi, Howida A. Shedeed, Amr Badr - Show less +3 more 31 Aug 2018-IEEE Access 78 Citations	The proposed CSP\AM-BA-SVM transcends the traditional CSP\SVM approach and other existing studies.
Journal Article•DOI Improved neural network for SVM learning Davide Anguita, A. Boni - Show less +1 more 01 Sep 2002-IEEE Transactions on Neural Networks 45 Citations	We show that this formulation contains some unnecessary circuits which, furthermore, can fail to provide the correct value of one of the SVM parameters and suggest how to avoid these drawbacks.

My columns

Related Questions

How do I turn off SVM?15 answers

How do I turn on HDR in pixel 4a?5 answers

How to build SVM model in R?7 answers

How to enable SVM MSI BIOS?4 answers

How to speed up SVM in R?8 answers

What does SVM stand for in machine learning?3 answers

See what other people are reading

What are the current advancements in automatic modulation classification techniques?

Recent advancements in automatic modulation classification (AMC) techniques include the utilization of meta-learning methods for few-shot AMC with distribution bias. Additionally, the integration of Deep Learning (DL) algorithms, such as standalone RNN and a combination of Convolutional Neural Network (CNN) with Long Short-Term Memory (LSTM), has shown promising results in accurately classifying wireless signals. Moreover, the application of deep residual neural networks with masked modeling (DRMM) has been proposed to enhance AMC performance, particularly in scenarios with limited signal samples and low signal-to-noise ratio (SNR). These advancements aim to address challenges in signal classification, improve generalization abilities, and enhance the accuracy of modulation identification in various communication environments.Cancer classification, treatment

Cancer classification and treatment involve various methods to predict patient outcomes and treatment effectiveness. Classification can be based on targetable events within a tumor, leading to more accurate personalized treatment approaches. Common cancer treatments include surgery, radiation therapy, chemotherapy, immunotherapy, targeted therapy, hormone therapy, and stem cell transplantation. Machine Learning techniques like Artificial Neural Networks (ANN), Decision Trees (DT), Support Vector Machines (SVM), and Bayesian Networks (BNs) have shown high accuracy in classifying cancer datasets, aiding in treatment decisions. Integrated cancer classification models may enhance future treatments by eliminating unclassifiable cancers and providing a more comprehensive understanding of cancer types and their treatments.How does motion capture technology affect the accuracy of body posture monitoring?

Motion capture technology significantly enhances the accuracy of body posture monitoring by utilizing sensor fusion and advanced algorithms. Various studies highlight the importance of motion capture systems in improving posture detection. These systems integrate data from multiple sensors to reduce errors and enhance recognition accuracy. By employing techniques like feature extraction and sensor fusion, motion capture systems can transform human body movements into quantifiable data with high precision. Additionally, the use of wireless inertial motion capture systems distributed across key body limbs ensures synchronized reconstruction of posture and displacement, effectively controlling estimation errors and maintaining consistency with actual movements. Furthermore, automatic attitude capture methods based on sensor fusion have shown high accuracy rates and short time consumption, making them effective for real-time motion posture monitoring.How accurate are machine learning algorithms in predicting mothers with postpartum depression compared to traditional methods?

Machine learning (ML) algorithms have shown promise in predicting postpartum depression (PND) in mothers. Studies have highlighted the importance of leveraging ML models based on Electronic Medical Records (EMRs) to predict depression in early pregnancy, especially in racial/ethnic minority women. Additionally, ML techniques have been utilized to construct models for PND risk prediction, with subjective poor sleep quality and insomnia symptoms being identified as key factors in predicting PND during early pregnancy. Comparatively, traditional methods may not encompass all relevant factors, as seen in the effectiveness of ML algorithms like Support Vector Machine (SVM) in predicting depression, outperforming Logistic Regression and Multinomial Naive Bayes. Therefore, ML algorithms offer a more accurate and comprehensive approach in predicting mothers with postpartum depression compared to traditional methods.How IoT sensors and data analytics are used to detect and locate leaks in real-time in water distribution?

IoT sensors and data analytics play a crucial role in real-time leak detection and localization in water distribution systems. By combining IoT technology with deep learning techniques, such as recurrent neural networks (RNNs) and transformer neural networks, anomalies in flow data can be identified. Additionally, the use of IoT-based solutions enables the development of real-time pipeline leakage alarm systems that can detect leaks, send alerts through the cloud, and facilitate immediate action to prevent water loss. Furthermore, the application of IoT monitoring devices and Artificial Intelligence (AI) technology, including deep learning models like RNN-LSTM, allows for accurate detection of water leaks by analyzing deviations between actual and forecasted sensor values. This integrated approach enhances the efficiency of leak detection and location, aiding in the maintenance and sustainability of water distribution networks.What specific attributes have been found to contribute to the high digital literacy of students?

Attributes contributing to high digital literacy in students include self-initiation, trendiness, usefulness, good attitude, information literacy competence, media literacy, and ICT competence. Self-initiation, trendiness, usefulness, and good attitude were identified as key components of digital literacy in global citizens. Additionally, high-performing students spend more time reading emails, receive high-quality teaching incorporating digital literacy, while low-performing students lack interest in ICT use and are more susceptible to ICT resource abuse, classroom disorder, and discrimination. Moreover, students with medium to high digital literacy skills show improvement in digital literacy after using electronic Invertebrate books for learning. These attributes collectively contribute to enhancing students' digital literacy levels and can guide educators in promoting effective digital literacy education.Why QR code is needed in smart parking?

QR codes are essential in smart parking systems for various reasons. They are utilized for assigning parking spaces in real-time, authorizing vehicles, and facilitating spot selection based on availability. QR codes integrated with microcontroller technology and website-based applications help monitor parking space availability in real-time, optimizing parking area utilization. Additionally, QR codes play a crucial role in enhancing security within parking systems by providing a means of vehicle identification and access control. They enable features like car plate recognition and serve as a backup in case primary identification methods fail, ensuring a comprehensive security system for users. Overall, QR codes streamline parking operations, enhance user experience, and contribute to efficient parking management in smart parking solutions.What percentage of instructors in Higher Education use Artificial intellegence in the classroom?

Approximately 17% of instructors in Higher Education utilize Artificial Intelligence (AI) in the classroom. This percentage is derived from a systematic review spanning from 2016 to 2022, which analyzed 138 articles on AI in Higher Education. The review highlighted that while 72% of the studies focused on students, only 17% were centered around instructors, indicating a lower integration of AI for instructional purposes. The study also identified various ways AI was used in Higher Education, including as an AI Assistant, in Intelligent Tutoring Systems, and for managing student learning. The findings suggest a significant gap in AI integration for instructors compared to students, emphasizing the need for further exploration and adoption of AI tools in teaching practices.Why data security is important in financial institutes?

Data security is crucial in financial institutions due to various reasons. Firstly, the safe storage of financial data is essential to prevent potential attacks and breaches. Such breaches can lead to significant financial losses, collapse, and loss of customer trust. Moreover, inadequate data security measures can result in information leakage, potentially caused by internal factors like employee behavior and lack of training. Implementing robust data protection mechanisms is vital to safeguard confidential information from competitors, corrupt officials, and criminals. Therefore, ensuring data security in financial institutes is imperative to maintain trust, prevent financial damage, and protect sensitive information from various threats, ultimately safeguarding the institution's stability and reputation.How to identify promising import items from the perspective of government?

To identify promising import items from the government's perspective, various factors need consideration. Firstly, optimizing government functions and economic management capabilities are crucial for evaluating market development and enhancing economic capacity. Secondly, analyzing the impact of government procurement policies, such as domestic content requirements, on import behavior can provide insights into import sourcing decisions. Additionally, understanding the relationship between government expenditures and imports through empirical analysis can help in assessing the direct effect of changes in government expenditure on import demand. Moreover, prioritizing import health standards based on economic considerations can aid in identifying potential import opportunities while managing available resources effectively. Lastly, adopting a 'government perspective' analytic framework can help evaluate investments in medical technology by considering changes in morbidity and mortality that influence government accounts, expenditures, and tax revenues.What other measures are required along with technological interventions to combat farm theft?

To effectively combat farm theft, a multifaceted approach that combines technological interventions with community engagement, legal frameworks, and educational initiatives is required. Technological interventions, such as the use of CCTV, wireless sensor networks, IoT-based livestock tracking, and face recognition systems, have shown promise in enhancing farm security by monitoring and alerting farm owners of unauthorized entries and livestock movements. These technologies can deter theft and facilitate the tracking of stolen goods or livestock. However, their effectiveness can be significantly amplified when integrated with additional measures. Community engagement and education play a crucial role in preventing farm theft. Initiatives that involve regular patrols, proactive operations, and community education about farm crime prevention have been widely used and are perceived as effective by some police respondents. Moreover, the implementation of crime prevention tactics that increase the effort and risk for offenders, as well as reduce the rewards from farm theft, can be more successful with the active participation of the farming community. Legal frameworks and enforcement are also critical. The Agricultural Crime, Technology, Information, and Operations Network (ACTION) initiative highlights the importance of increasing guardianship measures among farmers and the role of arrest and prosecution in reducing victimization. This suggests that strengthening the legal and enforcement mechanisms to support technological and community-based interventions is essential. Furthermore, addressing the underlying issues such as fear of crime, which can significantly impact farmers' production decisions and overall agricultural development, requires policy interventions that improve the rule of law and provide a sense of security to farmers. This holistic approach, combining technological advancements with community engagement, legal support, and educational efforts, can create a more secure and resilient agricultural sector.