A well-designed platform for analyzing and visualizing big data is necessary with huge growth of data in various domains. The current study presented and implemented an Extract-Transform-Load (ETL)-based platform for analysis and visualization of big data, which includes few computing nodes and storage nodes. Nevertheless, analyzing and visualizing continuous spatial-temporal result for big data is always a challenge in predictive work, for instance, air quality forecasting. Therefore, the current study extends a previous work and presents a platform for analysis and visualization of continuous spatial-temporal monitoring and forecasting of the air quality in cloud-enabled big data platform. This platform consists of Spark Hadoop-based machine learning environment for big data analysis, Tensorflow-based Deep Learning framework for forecasting, and a well-known map for demonstration and visualization of the result. To achieve the goal, Inverse Distance Weighted (IDW) was exploited at first to estimate the concentration of particulate matter in space without air quality monitors. Further, cloud-based Non SQL DB was also exploited to store the pre-calculated air quality image and utilize sliding window-based mechanism to easily slide the time bar so as to watch the result. This work also provides an instant overview of air quality in Taiwan with which one can get to know the air quality in all the parts of Taiwan immediately. Few evaluations were conducted to understand the performance of the platform which inferred that the platform, presented in this paper, is efficient and provides continuous spatial-temporal result to the needy.