This research proposes an effective money prediction via a data-driven tool to reduce costs by integrating intelligent data mining techniques. This study plans to improve revenue prediction and find venues for cost st...
详细信息
ISBN:
(数字)9798331512088
ISBN:
(纸本)9798331512095
This research proposes an effective money prediction via a data-driven tool to reduce costs by integrating intelligent data mining techniques. This study plans to improve revenue prediction and find venues for cost stemming by analyzing financial statements for the last 15 years, such as balance sheets and profit and loss reports. data preprocessing, exploratory dataanalysis (EDA), correlation analysis, regression analysis, clustering, and time series analysis are well integrated in finding feasible solutions. Model performance is evaluated on multiple metrics, including R-square and mean absolute error while considering computational complexity and efficiency. The findings illustrate how business houses can use such analytical methods to achieve better decision-making, financial sustainability, and increased efficiency. The work also discusses practical limitations, such as data availability and the model’s fitting to dynamic market conditions. The study highlights the importance of continuously monitoring and refining the model for the prediction in a real-world context.
In developing countries, pavement management systems (PMS) face limitations such as insufficient data, budget constraints, and inadequate analysis tools. To address those limitations, simplified PMS are needed to incr...
详细信息
The aim of the study is to evaluate the methods of autonomous collection and use of local contexts for selecting an answer option by an intelligent system in the dialog interaction process. The following research meth...
详细信息
ISBN:
(数字)9798331531836
ISBN:
(纸本)9798331531843
The aim of the study is to evaluate the methods of autonomous collection and use of local contexts for selecting an answer option by an intelligent system in the dialog interaction process. The following research methods were used: analysis of technical documentation, scientific papers, conference materials and instructional manuals. The current methods of collecting local contexts used in the most successful projects are analyzed. The methods of coding text words for processing by software are analyzed. The methods of measuring the quality of data collected in this way are analyzed. The result of the work is the presentation of the studied methods of collecting, storing and evaluating local contexts. A method for modeling subcategory contexts by a subject area is proposed in order to improve the quality of the decision-making system. A new method of graph modeling of logical connections of natural language data is proposed. A comparison of the effectiveness of various approaches to modeling local contexts is carried out and the corresponding conclusions are made.
This spatiotemporal complexity presents a very big challenge in effectively modeling and predicting patterns across domains that include climate systems, traffic networks, and disaster progression. Currently, the meth...
详细信息
ISBN:
(数字)9798331522667
ISBN:
(纸本)9798331522674
This spatiotemporal complexity presents a very big challenge in effectively modeling and predicting patterns across domains that include climate systems, traffic networks, and disaster progression. Currently, the methods often tend to fail in capturing the time-varying dynamics, optimizing spatial paths, and managing the high dimensionality of such datasets. This leads to suboptimal performance and increased computational overhead. In response to the aforementioned limitations, three new methods of innovative encoding, learning, and optimization are proposed to significantly boost the spatiotemporal analysis of data in this work. First, a dynamic spatiotemporal network called DeepSTNet with adaptive temporal encoding and graph-based spatial attention for capturing region-wise interactions achieved 5-10% higher accuracy and reduced computation by 20-30%. Second, ReinforceSTPath uses a multi-agent reinforcement learning framework to optimize spatial path predictions with temporal feedback, allowing for collaborative learning and achieving a 15-20% reduction in path prediction error while increasing efficiency by 30% compared to deterministic models. Lastly, inspired by the dynamics of an ant colony, BioST-Optimizer introduces a bioinspired swarm intelligence algorithm for feature selection in high-dimensional datasets, reducing dimensions by 30-50% and preserving 98% predictive accuracy while reducing training times by 40%. Collectively, these methodologies advance the state-of-the-art modeling of spatiotemporal data in dynamical alignment of temporal and spatial characteristics, optimization of path prediction, and efficient feature selections. This framework deals with the major shortcomings of previous approaches. Robust, scalable, and efficient solutions with wide applicability and transformative potential for spatiotemporal dataanalysis are provided in process.
Industrial control Systems (ICS) are paramount to the efficient operation of Critical National Infrastructure (CNI) ranging from electricity generation and distribution to manufacturing. However, the growing convergen...
详细信息
Industrial control Systems (ICS) are paramount to the efficient operation of Critical National Infrastructure (CNI) ranging from electricity generation and distribution to manufacturing. However, the growing convergence of ICS with Information Technology (IT) systems renders CNI vulnerable to a range of cyber threats. Graph neural networks are being increasingly used for anomaly detection by adding granularity to the detection process. In this paper, we present a comparative study of graph-based deep learning models for ICS anomaly detection. Through the evaluation of four models using three multivariate industrial datasets, we aim to discern the effectiveness of prediction and reconstruction-based graph models in the ICS domain. We investigate data reduction techniques to minimise features needed to represent the window size and examine the representation of sliding window in terms of feature size for time-series analysis. Additionally, we assess the impact of the length of a context window on anomaly detection performance. Our results show that using feature reduction techniques on a longer context window produces better results while having the computational advantages of a shorter window size. Graph autoencoder is the most resilient to feature size reduction by maintaining similar F1 and AUC-PR score regardless of the number of features used to represent a context window. The results also provide insight to the suitability of graph-based models in this domain and offer recommendations for their optimal usage, paving the way for enhanced security and resilience in ICS.
Most artificial intelligence (AI) applications are designed under the model-centric AI (MCAI) approach, where data scientists aim to optimize the machine learning (ML) models starting with fixed, preprocessed data. Ho...
详细信息
The air purifier is of great significance for purifying air. In this paper, aiming at the use of air purifiers, the data preprocessing of ten representative air purifiers is first carried out. After the consistency te...
详细信息
ISBN:
(纸本)9798400711831
The air purifier is of great significance for purifying air. In this paper, aiming at the use of air purifiers, the data preprocessing of ten representative air purifiers is first carried out. After the consistency test of the selected eight indicators, the entropy weight method is used to weight them, and the TOPSIS method is used to process the data set. A comprehensive multi-index evaluation model for evaluating the advantages and disadvantages of air purifiers is established. Based on this model, the selected ten air purifiers are ranked in terms of comprehensive cost performance. Then, the air purifier model and the living room simulation model were established by using SolidWorks software. Based on the dispersion of fluid mechanics, control equations and equations, the control equation model of gas fluid is constructed. The CFD finite element simulation is used to establish a suitable three-dimensional model of the analysis scene. In this paper, the air purifier and the model and the living room simulation model are meshed and the calculation area is adjusted. The control equation model of the air fluid in the scene is solved by the pressure base solver. The equation and the established material balance model are imported into Fluent software for calculation. Finally, the cloud map of the influence of an air purifier placed at different positions in the room on the concentration of air pollutants at different heights in the room is obtained by fluent-post.
Semiconductor manufacturing requires highly precise defect detection to ensure product quality and yield. This paper presents a deep learning-based defect detection framework using Faster R-CNN to identify and classif...
详细信息
ISBN:
(数字)9798331531850
ISBN:
(纸本)9798331531867
Semiconductor manufacturing requires highly precise defect detection to ensure product quality and yield. This paper presents a deep learning-based defect detection framework using Faster R-CNN to identify and classify defects in Scanning Electron Microscope (SEM) images, specifically within the modeling flow, which is used to simulate and predict semiconductor process behavior based on metrology data. The process begins with manufacturing test patterns through the photolithography process, followed by metrology analysis using a SEM machine. If a defect is present in the SEM image, the associated metrology data may become unreliable, potentially degrading the accuracy of the simulation models built on this data. Using such erroneous data in model creation could lead to inaccurate predictions and weaker *** address this issue, our method filters out defective data, ensuring that only high-quality, defect-free measurements contribute to model development. This leads to more reliable semiconductor process simulations and improved process decisions. Leveraging the proven performance of Faster R-CNN—particularly its effectiveness in detecting small, low-contrast anomalies—the model generalizes across diverse layouts without requiring pattern-specific training. It achieves a mAP of 83% at IoU = 0.5, significantly reducing manual inspection time and improving defect classification accuracy. Our approach enables scalable, automated inspection of SEM data, enhancing the overall efficiency, robustness, and precision of the semiconductor modeling and manufacturing pipeline.
The Wafer Acceptance Test (WAT) is a significant quality control measurement in the semiconductor industry. However, because the WAT process can be time-consuming and expensive, sampling test is commonly employed duri...
详细信息
ISBN:
(数字)9783982674100
ISBN:
(纸本)9798331534646
The Wafer Acceptance Test (WAT) is a significant quality control measurement in the semiconductor industry. However, because the WAT process can be time-consuming and expensive, sampling test is commonly employed during production. This makes root cause tracing impossible when abnormal products have not been tested. Therefore, in our study, we focus on establishing a reliable method to estimate WAT results for non tested shots, including both intra and inter-wafer prediction. Notably, we are the first to combine the use of Chip Probing data with WAT to improve the predictions. Our proposed method first extracts valuable features from Chip Probing test results by using the Automated Machine Learning technique. We then employ Gaussian process Regression to capture the spatiotemporal correlation. Finally, we adopted the linear regression model to ensemble two components and proposed a SMART-WAT model to effectively estimate the wafer acceptance test data. Our method has been tested on a real-world dataset from the semiconductor manufacturing industry. The prediction results of four key WAT parameters indicate that our proposed model outperforms the state-of-the-art methods in both intra and inter-wafer prediction.
暂无评论