Emotion classification remains a challenging problem in affective computing. One of the most crucial areas of study in the field of brain wave research is the classification of emotions. Classifying the types of emoti...
详细信息
Emotion classification remains a challenging problem in affective computing. One of the most crucial areas of study in the field of brain wave research is the classification of emotions. Classifying the types of emotions accurately is one of the major issues with the analysis of brainwave emotion. EEG signals used for real-time emotion identification are crucial for affective computing and human-computer interaction. These signals can be produced by the user while engaging in a variety of cognitive, affective, and physical tasks, representing the functionality of the brain. The resulting emotional state produced gives valuable insights on the attitudes and actions of participants in specific situations. The main objective of this research work is to classify the emotions using EEG signals. The process is divided into two steps. The first step is feature extraction and the next step is classification. The feature extraction is performed by using DWT and the selection is done by using L1 norm. The algorithms used to perform signal classification are LSTM, GRU and DNN.
Recently, as the sizes of real tensors have become overwhelmingly large including billions of nonzeros, fast and scalable Tucker decomposition methods have become increasingly important. Tucker decomposition has been ...
详细信息
Recently, as the sizes of real tensors have become overwhelmingly large including billions of nonzeros, fast and scalable Tucker decomposition methods have become increasingly important. Tucker decomposition has been widely used to analyze multidimensional data modeled as tensors. Several GPU-based Tucker decomposition methods have been proposed to enhance the decomposition speed. However, they easily fail to process large-scale tensors owing to the high memory requirements, which are larger than the GPU memory. This paper presents a scalable GPU-based Tucker decomposition method called GTucker, which carefully partitions large-scale tensors into subtensors and processes them with reduced overhead on a single machine. The results of the experiments indicate that GTucker outperforms state-of-the-art methods in terms of scalability and decomposition speed.
Semi-supervised learning, a system dedicated to making networks less dependent on labeled data, has become a popular paradigm due to its strong performance. A common approach is to use pseudo-labels with unlabeled dat...
详细信息
Sorting is a fundamental task in computer programming. Many sorting algorithms have been developed. Sorting algorithms are taught in the programming, data structures, and algorithms courses. The common sorting algorit...
详细信息
ISBN:
(数字)9798350361513
ISBN:
(纸本)9798350372304
Sorting is a fundamental task in computer programming. Many sorting algorithms have been developed. Sorting algorithms are taught in the programming, data structures, and algorithms courses. The common sorting algorithms introduced in these courses are selection sort, insertion sort, bubble sort, merge sort, quick sort, and heap sort. We have created the animations to visually demonstrate how these sorting algorithms work. The animation is developed using HTML5, CSS, and JavaScript. It is platform independent. It can be viewed from a browser on any device. The animations are useful tools for teaching and learning sorting algorithms. This paper presents these animations.
Segmentation of a forest area with high accuracy is a non-trivial task. In this paper, the authors examined the procedure for determining the coordinates of the location of trees and their diameters in order to obtain...
详细信息
Segmentation of a forest area with high accuracy is a non-trivial task. In this paper, the authors examined the procedure for determining the coordinates of the location of trees and their diameters in order to obtain an initial idea of the structure and quantitative filling of the forest area. To conduct the study, LiDAR data were collected from a mixed forest located in Russia with a high density of trees with a total area of about 1 hectare. The proposed procedure provides detection of 97.5% of trees. Several methods are used: covariance analysis and k-d-tree partitioning, hierarchical and non-hierarchical DBSCAN segmentation algorithm, HyperLS circle point approximation method, as well as other algorithms. All the data obtained is planned to be used for further segmentation of the forest area.
In this paper, we present an experimental validation of a photovoltaic/electrolysis system dedicated to supplying an alternating load and producing hydrogen. The system uses new way to produce hydrogen by adapting the...
详细信息
Micromobility IoT devices and Connected Vehicles generate massive mobility data, crucial for time-critical safety-related data analytics. It is challenging to study and understand such data without compromising user p...
详细信息
Micromobility IoT devices and Connected Vehicles generate massive mobility data, crucial for time-critical safety-related data analytics. It is challenging to study and understand such data without compromising user privacy. We propose AFFIRM, a secure privacy-preserving blockchain framework for efficient, scalable and lightweight mobility data generation, validation, storage and retrieval in future Web3 applications. AFFIRM enables nearby devices to self-organize as a fog network and collaboratively train machine learning algorithms locally to securely generate, validate, store and retrieve mobility data via consensus leveraging Information Centric Networking as the underlying architecture. The proposed collaborative learning enables nodes to learn and adapt with respect to parameters related to scalability, timeliness, security, privacy, and resource consumption. We evaluate AFFIRM using mobility data from New York city and results shows it to scalably store mobility data from up to 700 devices with lower delays and overhead.
This paper conducts in-depth research and discussion on data storage security in cloud computing environment. This study collected 1000 dummy data from different organizations as the basic data set for the study. This...
This paper conducts in-depth research and discussion on data storage security in cloud computing environment. This study collected 1000 dummy data from different organizations as the basic data set for the study. This data covers various types of sensitive information, including personally identifiable information, financial data, and medical records. We generate these data using appropriate data generation methods to ensure realistic characteristics and distributions. Based on existing backup technologies and algorithms, we evaluate the impact of different backup strategies on data security. Experimental results show that our data backup and recovery strategy and the application of encryption technology effectively improve the security of data in cloud storage. By selecting backup strategies and recovery mechanisms reasonably, we can ensure data availability and integrity. At the same time, employing a proper encryption scheme can protect data from the risk of unauthorized access and disclosure.
Stochastic control deals with finding an optimal control signal for a dynamical system in a setting with uncertainty, playing a key role in numerous applications. The linear quadratic Gaussian (LQG) is a widely-used s...
详细信息
Stochastic control deals with finding an optimal control signal for a dynamical system in a setting with uncertainty, playing a key role in numerous applications. The linear quadratic Gaussian (LQG) is a widely-used setting, where the system dynamics is represented as a linear Gaussian state-space (SS) model, and the objective function is quadratic. For this setting, the optimal controller is obtained in closed form by the separation principle. However, in practice, the underlying system dynamics often cannot be faithfully captured by a fully known linear Gaussian SS model, limiting its performance. Here, we present LQGNet, a stochastic controller that leverages data to operate under partially known dynamics. LQGNet augments the state tracking module of separation-based control with a dedicated trainable algorithm. The resulting system preserves the operation of classic LQG control while learning to cope with partially known SS models without having to fully identify the dynamics. We empirically show that LQGNet outperforms classic stochastic control by overcoming mismatched SS models.
With the development of big data, the need for realtime dataprocessing becomes more urgent. However, the complexity of the big data business causes the current data components to not be well supported. For this reaso...
With the development of big data, the need for realtime dataprocessing becomes more urgent. However, the complexity of the big data business causes the current data components to not be well supported. For this reason, a complex real-time calculation method based on data center is proposed. Build incremental dimension tables and incremental fact tables based on the original micro-batch scheduling, and analyze the impact of business incremental data changes on statistical results. And through the analysis of business operations, simplify the dataprocessing process, reduce the data Merge operation, and reduce the dataprocessing time. Finally, the feasibility, correctness and real-time nature of the method are verified through the application of real-time calculation of electricity bills.
暂无评论