In this study, we apply the energy model to improve accuracy and performance in object detection. The research uses popular datasets such as PASCAL VOC 2012 to train and evaluate the model. The results show that by integrating energy models instead of relying solely on regression methods to predict the bounding box coordinates, as in traditional approaches, this energy-based framework combines visual, geometric, and spatial factors to optimize object detection more comprehensively and accurately.
In this study, the authors proposed a new method for constructing an developing an effective model aimed at assessing and predicting fluctuations in the foreign exchange (Forex) market using the Ensemble learning method. Accurately forecasting the trends in Forex trading offers investors more suitable options for their trading decisions. The model is equipped with the capability to automatically process trading parameters from online information sources, allowing it to predict market trends with high accuracy. It integrates time-related features and intelligently optimizes parameters for training, which contributes to its clear stability and high performance. Specifically, the model achieved impressive performance metrics, with an R-squared value of 0.999 and a root mean squared error (RMSE) of 0.00357. This proposal not only offers an effective method for predicting fluctuations in the Forex market but also highlights significant potential for future developments in assessment and prediction models within the currency field.
In the Fourth Industrial Revolution Era with the fast-growing of multiple types of massive data, digitalized content as well as text-based information, which is increased through social media like Youtube, Facebook,… Many scientists worldwide are deeply interested in data analysis. Many researchers have applied various text classification techniques to improve the Naive Bayes algorithm, particularly due to its tendency for attribute dependence. Resolving this interdependence is key to enhancing the model’s accuracy. The Multinomial Naïve Bayes model is widely used in text classification, where each document is represented by a feature vector of length d, corresponding to the number of words in the dictionary. The manual process of selecting articles is to filter and extract them into lists of documents. But as relevant information is getting more complicated day by day and bulky, together with it is the increasing of fake news, wrong information. That’s why the ability to be able to precisely filter out these online articles is highly important. To ensure whether the readers receive the reliable information or not. That is why we need to develop a precise news selecting system and also get our hands on the predict and filter system to sort out online articles using the strong and well-known Naïve Bayes Algorithm to do the work since it is simple, fast and accurate.
Segmentation of overlapping cells in biological images remains a challenging task due to the complexity introduced by occlusion and overlap, which often leads to decreased accuracy in automated medical diagnostics systems. This research focuses on improving the segmentation of overlapping cells using an advanced deep learning approach. The model is trained on the ISBI2014 dataset of cervical cytology cells, a collection of images with noisy and overlapping cells. Experimental results demonstrate that the proposed model achieves higher accuracy, even with limited labeled data. By calculating the nucleus-to-cytoplasm (N/C) ratio from the segmentation results, the proposed method determines whether a cell is normal or exhibits pathological characteristics. The proposed method provides an efficient and scalable solution for the classification and segmentation of overlapping cells, offering new opportunities for advancements in automated image analysis and contributing to the development of more accurate and reliable systems for cell-based medical diagnosis.
This paper presents a machine learning-based methodology for the classification and real-time statistical analysis of vehicular traffic to facilitate the assessment of traffic flow on important routes. We propose a system architecture model to integrate monitoring devices, data processing capabilities, and user interaction through interfaces, including a 3D WebGIS system utilized for the simulation of traffic system. Moreover, this study conducts experimental scenarios to evaluate the system's performance in the context of traffic management and monitoring in Da Nang.The results of this research not only provide valuable insights for planners in monitoring and coordinating traffic flow but also contribute to the construction of a Digital Twin system, thereby supporting the future development of smart city.
