Improving Heat Index Results Derived From IOT Data through the Application of Detailed Neural Networks

A. Karthik, Dr.G Sunil Kumar

Zenodo (CERN European Organization for Nuclear Research) · 21 November 2022

In one sentence. An LSTM-style deep network forecasts heat index from IoT temperature sensor series and is compared with generalized linear regression; the methods are thinly described.

Abstract

Currently, global warming is the leading source of climate pollution due to the release of CO2 and other pollutants. This combination of factors significantly raises temperatures and has far-reaching effects on the atmosphere. Minimizing the least-efficient activities will have the greatest impact on lowering global temperatures and limiting carbon dioxide emissions. Almost all manufactured items now generate substantial amounts of carbon dioxide as a byproduct due to technological advancements. A forecast is needed to forestall a catastrophe brought on by climate change. Several researchers were investigating the projected temperature range based on agricultural production, and one of them found an increase. A programmed was used to evaluate the algorithms, and it was similar to one used to track agricultural yields or hotel occupancy. Therefore, a deep learning technique is suggested as an option for achieving the goal of heat index forecasting in this context. Data on temperatures is stored in a database kept by the Climate Prediction Center, and those figures came from the Kaggle dataset. You can get into this database right here. There is room in the IoT architecture for time series data collected from temperature sensors. The model with long-term memory and deep learning extrapolates information from limited sources. The effectiveness of the neural network is measured in terms of accuracy, sensitivity, and specificity, and compared to that of a generalized linear regression. All versions of MATLAB R2020b are now running in parallel on Windows 10.

Labels

Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.