Citation :

Shital Tayade, Nalini Vaidya, "Predicting and Categorizing Urban Air Pollution: A Comparative Study of Regression and Classification Models with Vedic Mathematical Feature," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 3, pp. 75-88, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I3P106

Abstract

Due to the adverse impacts on both humanity and the environment, severe air pollution issues have drawn attention from all over the world. The air quality in Delhi, the capital of India, has drastically declined. Delhi has the second-leading air pollution rate in the world, according to the 2024 Global Weather Pollution Review of an ecological surveillance organization. This work combines support vector approaches to learning, neural computing models, and Vedic mathematics for investigating Delhi's air quality using an enormous dataset. With a focus on computational performance and data efficiency during processing, we compare the proposed strategy to gradient descent-based optimization techniques like Adagrad, Adadelta, Adam, and RMSProp. The current investigation offers a thorough analytical approach for evaluating urban air quality through classification and regression processes. Based on airborne particle concentration data acquired in Delhi between 2020 and 2023, the solutions suggested utilize regression for accurately predicting Air Pollution Index (AQI) values and algorithmic classification for assessing the degree of air pollution. The modeling approach makes use of results from simulation to evaluate the effects of two Vedic mathematics sutras, Urdhva-Tiryakbhyam and Nikhilam Sutra. We evaluate the performance of models like SVM and ANN using feature transformations based on Vedic mathematics, like Urdhva-Tiryakbhyam. Furthermore, we contrast the Vedic Mathematics multipliers based on the Nikhilam Sutra and Urdhva-Tiryakbhyam with the conventional machine learning coefficient computations in SVM and ANN models, which use a variety of optimization techniques. Common metrics like Mean Squared Error (MSE) and R2 for regression and F1-score, precision, and recall for classification have been employed to evaluate the efficacy of the two approaches. These findings demonstrate how the Vedic Mathematics framework increases computing efficiency by enabling quicker and less expensive coefficient computation. On the other hand, the combined analytical approach offers a more profound understanding of environmental monitoring.

Keywords

SVM, ANN, ADAM, Adagrad, Adadelta, RTMSprop, UrdhvaTiryakbhyam, Nikhilam Sutra.

References

[1] Yichen Wang et al., “Impacts of Natural and Socioeconomic Factors on PM_2.5 from 2014 to 2017,” Journal of Environmental Management, vol. 284, pp. 1-9, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[2] Biao Sun et al., “The Relationship Between Urbanization and Air Pollution Affected by Intercity Factor Mobility: A Case of the Yangtze River Delta Region,” Environmental Impact Assessment Review, vol. 100, pp. 1-31, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[3] Nikhil Vilasrao Deshmukh et al., “Predictive Analytics for Air Quality Classification: A Multi- City Study in India,” International Journal of Environmental Science, vol. 11, no. 4, pp. 1508-1520, 2025.
[Publisher Link]

[4] C. Karkuzhali, and R. Padmapriya, “Multivariate Regressive Gradient Spiral Optimized Deep Belief Learning for Crop Yield Prediction,” International Journal of Environmental Science, vol. 11, no. 14s, pp. 327-345, 2025.
[Publisher Link]

[5] R. Karthi Kumar, and S.P. Vimal, “Comparative Analysis of Vedic Multiplier using Vedic Sutras with Existing Multipliers in Biomedical Application,” Measurement: Sensors, vol. 36, pp. 1-10, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[6] K. Deergha Rao, Ch. Gangadhar, and Praveen K. Korrai, “FPGA Implementation of Complex Multiplier using Minimum Delay Vedic Real Multiplier Architecture,” 2016 IEEE Uttar Pradesh Section International Conference on Electrical, Computer and Electronics Engineering (UPCON), Varanasi, India, pp. 580-584, 2016.
[CrossRef] [Google Scholar] [Publisher Link]

[7]  Ke-Lin Du et al., “Exploring Kernel Machines and Support Vector Machines: Principles, Techniques, and Future Directions,” Mathematics, vol. 12, no. 24, pp. 1-58, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[8] Jing Geng et al., “Support Vector Machine Regression (SVR)-based Nonlinear Modeling of Radiometric Transforming Relation for the Coarse-Resolution Data-Referenced Relative Radiometric Normalization (RRN),” Geo-Spatial Information Science, vol. 23, no. 3, pp. 237-247, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[9] Xiaotong Hu, “Support Vector Machine and its Application to Regression and Classification,” MSU Graduate Theses, Missouri State University, pp. 1-57, 2017.
[Google Scholar] [Publisher Link]

[10] Sebastian Ruder, “An Overview of Gradient Descent Optimization Algorithms,” arXiv Preprint, pp. 1-14, 2016.
[CrossRef] [Google Scholar] [Publisher Link]

[11] Diederik P. Kingma, and Jimmy Ba, “Adam: A Method for Stochastic Optimization,” arXiv Preprint, pp. 1-15, 2014.
[CrossRef] [Google Scholar] [Publisher Link]

[12] Mohamed Reyad, Amany M. Sarhan, and M. Arafa, “A Modified Adam Algorithm for Deep Neural Network Optimization,” Neural Computing and Applications, vol. 35, no. 23, pp. 17095-17112, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[13] Tijmen Tieleman, “Lecture 6.5-RmsProp: Divide the Gradient by a Running Average of its Recent Magnitude, COURSERA: Neural Networks for Machine Learning, vol. 4, no. 2, 2012.
[Google Scholar]

[14] Shiliang Sun et al., “A Survey of Optimization Methods from a Machine Learning Perspective,” IEEE Transactions on Cybernetics, vol. 50, no. 8, pp. 3668-3681, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[15] Ahmet Cevahir Cinar, and Narayanan Natarajan, “An Artificial Neural Network Optimized by Grey Wolf Optimizer for Prediction of Hourly Wind Speed in Tamil Nadu, India,” Intelligent Systems with Applications, vol. 16, pp. 1-11, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[16] Alireza Askarzadeh, and Alireza Rezazadeh, “Artificial Neural Network Training using a New Efficient Optimization Algorithm,” Applied Soft Computing, vol. 13, no. 2, pp. 1206-1213, 2013.
[CrossRef] [Google Scholar] [Publisher Link]

[17] Gulfraz Syed et al., “The Role of Artificial Neural Networks in Machine Learning,” Neuro Quantology, vol. 20, no. 13, pp. 3239-3251, 2022.
[Google Scholar]

[18] Faisal Mehmood, Shabir Ahmad, and Taeg Keun Whangbo, “An Efficient Optimization Technique for Training Deep Neural Networks,” Mathematics, vol. 11, no. 6, pp. 1-22, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[19] Matthew D. Zeiler, “ADADELTA: An Adaptive Learning Rate Method,” arXiv Preprint, pp. 1-6, 2012.
[CrossRef] [Google Scholar] [Publisher Link]

[20] Wentao Yang et al., “Prediction of Hourly PM_2.5 using a Space-Time Support Vector Regression Model,” Atmospheric Environment, vol. 181, pp. 12-19, 2018.
[CrossRef] [Google Scholar] [Publisher Link]

[21] Xiang Li et al., “Deep Learning Architecture for Air Quality Predictions,” Environmental Science and Pollution Research International, vol. 23, no. 22, pp. 22408-22417, 2016.
[CrossRef] [Google Scholar] [Publisher Link]

[22] Ujjwal Kumar, and V.K. Jain, “ARIMA Forecasting of Ambient Air Pollutants (O₃, NO, NO₂ and CO),” Stochastic Environmental Research and Risk Assessment, vol. 24, no. 5, pp. 751-760, 2009.
[CrossRef] [Google Scholar] [Publisher Link]

[23] Ying Zhang et al., “A Predictive Data Feature Exploration-based Air Quality Prediction Approach,” IEEE Access, vol. 7, pp. 30732-30743, 2019.
[CrossRef] [Google Scholar] [Publisher Link]

[24] Saverio De Vito et al., “On Field Calibration of an Electronic Nose for Benzene Estimation in an Urban Pollution Monitoring Scenario,” Sensors and Actuators B: Chemical, vol. 42, no. 2, pp. 750-757, 2008.
        [CrossRef] [Google Scholar] [Publisher Link]