Search results for: parametric load

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 42

Authors: David J. Foran, Nhan Do, Samuel Ajjarapu, Wenjin Chen, Tahsin Kurc, Joel H. Saltz

Abstract:

The large-scale data and computational requirements of investigators throughout the clinical and research communities demand an informatics infrastructure that supports both existing and new investigative and translational projects in a robust, secure environment. In some subspecialties of medicine and research, the capacity to generate data has outpaced the methods and technology used to aggregate, organize, access, and reliably retrieve this information. Leading health care centers now recognize the utility of establishing an enterprise-wide, clinical data warehouse. The primary benefits that can be realized through such efforts include cost savings, efficient tracking of outcomes, advanced clinical decision support, improved prognostic accuracy, and more reliable clinical trials matching. The overarching objective of the work presented here is the development and implementation of a flexible Intelligent Retrieval and Interrogation System (IRIS) that exploits the combined use of computational imaging, genomics, and data-mining capabilities to facilitate clinical assessments and translational research in oncology. The proposed System includes a multi-modal, Clinical & Research Data Warehouse (CRDW) that is tightly integrated with a suite of computational and machine-learning tools to provide insight into the underlying tumor characteristics that are not be apparent by human inspection alone. A key distinguishing feature of the System is a configurable Extract, Transform and Load (ETL) interface that enables it to adapt to different clinical and research data environments. This project is motivated by the growing emphasis on establishing Learning Health Systems in which cyclical hypothesis generation and evidence evaluation become integral to improving the quality of patient care. To facilitate iterative prototyping and optimization of the algorithms and workflows for the System, the team has already implemented a fully functional Warehouse that can reliably aggregate information originating from multiple data sources including EHR’s, Clinical Trial Management Systems, Tumor Registries, Biospecimen Repositories, Radiology PAC systems, Digital Pathology archives, Unstructured Clinical Documents, and Next Generation Sequencing services. The System enables physicians to systematically mine and review the molecular, genomic, image-based, and correlated clinical information about patient tumors individually or as part of large cohorts to identify patterns that may influence treatment decisions and outcomes. The CRDW core system has facilitated peer-reviewed publications and funded projects, including an NIH-sponsored collaboration to enhance the cancer registries in Georgia, Kentucky, New Jersey, and New York, with machine-learning based classifications and quantitative pathomics, feature sets. The CRDW has also resulted in a collaboration with the Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC) at the U.S. Department of Veterans Affairs to develop algorithms and workflows to automate the analysis of lung adenocarcinoma. Those studies showed that combining computational nuclear signatures with traditional WHO criteria through the use of deep convolutional neural networks (CNNs) led to improved discrimination among tumor growth patterns. The team has also leveraged the Warehouse to support studies to investigate the potential of utilizing a combination of genomic and computational imaging signatures to characterize prostate cancer. The results of those studies show that integrating image biomarkers with genomic pathway scores is more strongly correlated with disease recurrence than using standard clinical markers.

Keywords: clinical data warehouse, decision support, data-mining, intelligent databases, machine-learning.

Procedia PDF Downloads 126