DGCA3QM: DESIGN OF A DUAL GENETIC ALGORITHM BASED AUTOREGRESSION MODEL FOR CORRELATIVE PREDICTION OF AIR QUALITY METRICS

DGCA3QM: DESIGN OF A DUAL GENETIC ALGORITHM BASED AUTOREGRESSION MODEL FOR CORRELATIVE PREDICTION OF AIR QUALITY METRICS

Blog Article

Prediction of air quality metrics is a multidomain task that involves analysis of different inter-related parameters.These parameters include, type of location, NO2 levels, color touch 7/97 SO2 levels, Respirable Suspended Particulate Matter (RSPM) levels, Fine particulate matter (PM2.5) levels, etc.

Researchers have proposed a wide variety of models to predict air quality via deep learning, bioinspired optimizations, regression analysis, and correlative analysis.But these models are either highly complex, or showcase low efficiency levels, which limits their applicability when used for big data scenarios.Moreover, most of these models do not consider parameter correlation between different metrics, which affects their accuracy levels.

To overcome these issues, this text proposes design of a Dual Genetic Algorithm (DGA) based Auto regression model for Correlative prediction (AC) of Air Quality Metrics.The proposed model initially collects large-scale datasets of multiple Air quality metrics for different areas, and uses a bioinspired optimization model for identification of correlative parameter sets.These sets are processed via an Autoregressive model, which assists in prediction of future air quality metrics via analysis of technical indicators including Simple Moving Average (SMA), Ternary Moving Average (TMA), etc.

The predicted values are further optimized via another bioinspired layer that assists in identification of high correlation value changes, thereby improving prediction performance under large data samples.Due to incorporation of dual bioinspired optimizers with autoregressive correlation, the model is able to improve prediction accuracy by 8.5%, precision by 4.

9%, click here recall by 1.5%, while reducing computational delay by 3.4% when compared with standard air quality analysis models.

These enhancements assist in deploying the model for real-time air quality prediction scenarios.