Scholarlink

Here different type of machine learning algorithm and techniques to enhance the prediction and identification of effect air quality on human health and environment. For doing this, we are using supervised learning algorithm (in this, prediction is done on the basis of labelled data) And unsupervised learning(in this, prediction is done on the basis of hidden structure within the dataset without any labelled outcomes). In this, we are applying different methods and finding which is the best for specific type of analysis. In this study, we are generally testing the data on the linear, Naive Base, K-nearest neighbors (KNN), Support vector machine (SVM), and Decision tree for doing the prediction based on the dataset and we are using K-mean and DBSCANE algorithm for clustering method to finding hidden pattern in the dataset. Before applying different method or we can say different machine learning algorithm, we first have to prepare the dataset. There are many steps we have to follow for preparation of dataset to apply the algorithm. If we don't prepare the dataset, then it is hard to apply the algorithms. The different method of preparation of data are Data Collection, Data Cleaning, Handling Categorical Data, Data normalization, and Scaling, Feature Engineering, Feature Selection, and Data Splitting. After applying algorithm, our result will tell that certain methods such as Supervised learning algorithm are very efficient for the task to predict the binary outcomes, and by saying binary outcome, we mean that air quality is above or below a certain threshold. On other part, unsupervised clustering method provide data which will tell the type of pollution and their distribution in different areas. Through this research, we know the strength and limitation of each algorithm. This study provides an insights and also we know what we can do to improve these machine learning models, how we optimize these models for future work, we are exploring different advanced machine learning algorithms, like Neural Networking and Ensemble method. This will be even more useful for predicting the air quality and pollution and having great impact on public health and environment [15, 19].

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning Techniques Applied to Predictive Modelling and Clustering for Air Quality and Health Diagnostics

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • J. Sarada,
  • G. S. Pradeep Ghantasala,
  • Vikram Neerugatti,
  • T. V. Ramana,
  • R. Rajesh Sharma,
  • Akey Sungheetha

摘要

Here different type of machine learning algorithm and techniques to enhance the prediction and identification of effect air quality on human health and environment. For doing this, we are using supervised learning algorithm (in this, prediction is done on the basis of labelled data) And unsupervised learning(in this, prediction is done on the basis of hidden structure within the dataset without any labelled outcomes). In this, we are applying different methods and finding which is the best for specific type of analysis. In this study, we are generally testing the data on the linear, Naive Base, K-nearest neighbors (KNN), Support vector machine (SVM), and Decision tree for doing the prediction based on the dataset and we are using K-mean and DBSCANE algorithm for clustering method to finding hidden pattern in the dataset. Before applying different method or we can say different machine learning algorithm, we first have to prepare the dataset. There are many steps we have to follow for preparation of dataset to apply the algorithm. If we don't prepare the dataset, then it is hard to apply the algorithms. The different method of preparation of data are Data Collection, Data Cleaning, Handling Categorical Data, Data normalization, and Scaling, Feature Engineering, Feature Selection, and Data Splitting. After applying algorithm, our result will tell that certain methods such as Supervised learning algorithm are very efficient for the task to predict the binary outcomes, and by saying binary outcome, we mean that air quality is above or below a certain threshold. On other part, unsupervised clustering method provide data which will tell the type of pollution and their distribution in different areas. Through this research, we know the strength and limitation of each algorithm. This study provides an insights and also we know what we can do to improve these machine learning models, how we optimize these models for future work, we are exploring different advanced machine learning algorithms, like Neural Networking and Ensemble method. This will be even more useful for predicting the air quality and pollution and having great impact on public health and environment [15, 19].