Software Impacts最新文献

Over conventional statistical models, machine learning mechanisms are establishing themselves as a potential area for modeling and forecasting complex time series. Because it can integrate several forecasting methodologies’ capabilities, hybrid time series models are fundamental in data science. Here, we present a Python script that builds a combined architecture of the ARIMA-LSTM model with random forest technique to generate a high-accuracy prediction. This script is a step-by-step process to create a statistical and then machine learning model through statistical assumption.

The article presents an open-access code, written in CUDA® and C++ programming language, applicable for generating computational models of nanostructured surface coatings deposited by electrodeposition. The code uses the Schrödinger equation, energy potentials, and electrochemistry as a theoretical basis to determine the deposition and electrodeposition energies, allowing the prediction of the formation and growth of these coatings. Likewise, the parameter variation enabled within the code provides for determining the main electrodeposition parameters (voltage, current, concentration, and residence time) for experimental depositions. The code can be easily implemented for any metallic coating-substrate arrangement where the filler material is nanomaterials.

The paper introduces ACS-IoT, an Anomaly Classification System for IoT networks, structured as a two-tiered framework. In the first, it employs a decision tree classifier for anomaly detection. In the second, a CNN-BiLSTM model is utilized for more profound analysis and classification of anomaly types. To address data imbalance, SMOTE is used, and feature selection is enhanced with PSO. The approach showcases strong practical applicability in real-world industrial settings, achieving an accuracy of 88%, precision of 89%, recall of 88%, and F1-score of 88% for multi-class classification, surpassing other machine learning approaches by at least 6% in all metrics.

SPARC is an accurate, efficient, and scalable real-space electronic structure code for performing ab initio Kohn–Sham density functional theory calculations. Version 2.0.0 of the software provides increased efficiency, and includes spin–orbit coupling, dispersion interactions, and advanced semilocal as well as hybrid exchange–correlation functionals, where it outperforms state-of-the-art planewave codes by an order of magnitude and more, with increasing advantages as the number of processors is increased. These new features further expand the range of physical applications amenable to first principles investigation.

This paper presents a scalable framework for modeling floc evolution and flocculation kinetics in water treatment. Unlike the existing methods that subjects Non-intrusive Dynamic Image Analysis (NiDIA) data to complex mathematical concepts, the proposed software devised a scaling concept for NiDIA data and designed an effective algorithm with the capability to predict varying floc lengths and the underlying kinetics under a broad flocculation conditions ($\text{G}f$ and $\text{T}f$). Technically, the designed machine-intelligence framework (MI-NiDIA) involves data preprocessing, automatic parameter selection, validation and prediction of floc length evolution with metrics. For instance, MI-NiDIA-MLP recorded $R^2$ of 0.95–1.0 for varying floc length at $\text{G}f60s^{-1}$.

Raster Forge is a Python library and graphical user interface for raster data manipulation and analysis. The tool is focused on remote sensing applications, particularly in wildfire management. It allows users to import, visualize, and process raster layers for tasks such as image compositing or topographical analysis. For wildfire management, it generates fuel maps using predefined models. Its impact extends from disaster management to hydrological modeling, agriculture, and environmental monitoring. Raster Forge can be a valuable asset for geoscientists and researchers who rely on raster data analysis, enhancing geospatial data processing and visualization across various disciplines.

This paper presents MR-LEAP (Mixed-Reality Learning Environment for Aspirational Programmers), a framework developed for learning programming through Mixed Reality and gamification mechanics. MR-LEAP’s architecture is designed to facilitate the understanding of basic programming concepts while allowing the gradual incorporation of more complex concepts. The framework provides a simple visual level editor. MR-LEAP is supported by the Mixed Reality Toolkit framework to promote portability to new Mixed Reality devices. Our goal is to facilitate programming education using Mixed Reality technology. MR-LEAP has already been used in both research and educational.

alPCA is a software coded in R and designed to automatically combine predictions from a collection of individual forecasting methods that integrate it. It employs three categories of weights derived from the PCA scores, and decision rules to determine the optimal combination of these methods. alPCA serves as an automated component within the artificial intelligence toolkit for monthly time series processing with the objective of obtaining the best forecast.

Most websites and applications are hosted on a public or private cloud. In-house deployments also require dealing with system resources. Researchers have started considering resources utilized by application workloads to estimate and reduce application running costs. RAW-HF (Resource Availability & Workload aware Hybrid Framework) framework tries to analyze two types of resource utilization; (1) System Resource Utilization and (2) Resource Utilized by each Query task. The RAW-HF code tries to provide a lightweight solution to monitor & analyze the system and DBMS process resource utilization. It filters the required data in real time to find available resources and allocate query-specific resources based on their complexity by utilizing less than 2% CPU resources.

DNNs are widely used for complex tasks like image and signal processing, and they are in increasing demand for implementation on Internet of Things (IoT) devices. For these devices, optimizing DNN models is a necessary task. Generally, standard optimization approaches require specialists to manually fine-tune hyper-parameters to find a good trade-off between efficiency and accuracy. In this paper, we propose OptDNN, a software that employs innovative and automatic approaches to determine optimal hyper-parameters for pruning, clustering, and quantization. The models optimized by OptDNN have a smaller memory footprint, faster inference time, and a similar accuracy to the original models.