A Machine Learning Approach for Air Quality Prediction: Model Regularization and Optimization

Abstract

Abstract:In this paper, we tackle air quality forecasting by using machine learning approaches to predict the hourly concentration of air pollutants (e.g., ozone, particle matter (PM 2.5 ) and sulfur dioxide). Machine learning, as one of the most popular techniques, is able to efficiently train a model on big data by using large-scale optimization algorithms. Although there exist some works applying machine learning to air quality prediction, most of the prior studies are restricted to several-year data and simply train standard regression models (linear or nonlinear) to predict the hourly air pollution concentration. In this work, we propose refined models to predict the hourly air pollution concentration on the basis of meteorological data of previous days by formulating the prediction over 24 h as a multi-task learning (MTL) problem. This enables us to select a good model with different regularization techniques. We propose a useful regularization by enforcing the prediction models of consecutive hours to be close to each other and compare it with several typical regularizations for MTL, including standard Frobenius norm regularization, nuclear norm regularization, and 2,1 -norm regularization. Our experiments have showed that the proposed parameter-reducing formulations and consecutive-hour-related regularizations achieve better performance than existing standard regression models and existing regularizations.

Using deep learning to extend the range of air pollution monitoring and forecasting

“…Finally, there is a long history of the use of machine learning in pollution monitoring [36, 37, 38, 39, e.g.]. Recently, Zhu et al [39] considered a coarse (0.25 degree resolution) grid of mainland China, with more than two years of air quality measurement and meteorological data, without any further insights, such as pollution sources, surface roughness, the reaction model, the multi-resolution aspects, or similar. Qi et al [38], considered a joint model for feature extraction, interpolation, and prediction while employing the information pertaining to the unlabelled spatio-temporal data to improve the performance of the predictions.…”

Air Quality Prediction in Visakhapatnam with LSTM based Recurrent Neural Networks

“…E Kalapanidas et al have proposed a naï ve approach using machine learning techniques for air quality prediction [32]. Athanasiadis et al have performed elaborate studies on air quality data in real time through machine learning approach [31]. Dixian et al have developed an approach for optimizing and regularization of machine learning models for air quality prediction [32].…”

Intelligent Air Pollution Prediction System using Internet of Things (Iot)

“…Following KF calculation finishes the forecast of the centralization of the impurity next to the following minute, RPi will send the expected results and observing information to the cloud, which stores the information in the database [2] [12] and criticisms results to the customer. Clients be able to see the present air quality and the most recent 24-H AQF (Air Quality file) pattern during the program's entrance, when appeared in the below image.…”

“…Next to human-made exercises, a cataclysmic event, for example, volcanic emission may bring about the defilement of air. Globalization is a noteworthy explanation behind defilement [2]. As a rule air poisons can be:…”

“…Presently multi daylight hours are better if each activity is finished utilizing latest innovation so as to fulfill the interest of a person, Organization, Enterprise, and so on. Web of Things (WIoT) is single of the fundamental correspondence improvements during the most recent decade [2] [4], from end to end this idea, it is conceivable to associate innumerable Low fueled keen implanted items to one another and to the Internet. The unavoidable nearness around us of different remote advancements, for example, Radio Frequency Identification (RFID) labels, sensors, actuators, and cell phones comprise the foundation of the IoT idea, these articles be able to send along with get information self-governing, along these lines opening new skylines for home, wellbeing, and modern applications.…”