الگوریتم مجموع تجمعی فضایی با تجزیه و تحلیل کلان داده
Abstract
1- Introduction
2- Related work
3- Proposed framework
4- Proposed algorithm for seasonal climate change detection
5- Comparison of various change detection algorithms
6- Result and discussion
7- Conclusion
References
Abstract
Big data plays a vital role in the prediction of diseases that occur due to climate change. For such predictions, scalable data storage platforms and efficient change detection algorithms are required to monitor the climate change. However, traditional data storage techniques and algorithms are not applicable to process the huge amount of climate data. This paper presents a scalable data processing framework with a novel change detection algorithm. The large volume of climate data is stored on Hadoop Distributed File System (HDFS) and MapReduce algorithm is applied to calculate the seasonal average of climate parameters. Spatial autocorrelation based climate change detection algorithm is proposed in this paper to monitor the changes in the seasonal climate. The proposed climate change detection algorithm is compared with various existing approaches such as pruned exact linear time method, binary segmentation method, and segment neighborhood method.
Introduction
“Big Data” is defined by volume, velocity, and variety of data. Big data is very complex to process by traditional data processing techniques and tools. Nowadays, data generation sources like telescopes, satellite, sensor networks, social networks, wearable devices, mobile devices, streaming machines and high throughput instruments are continuously generating a large volume of data. Recently, big data analytics has been applied in various domains, such as healthcare, business process, scientific research, natural resource management, share marketing, social networking, community administration and climate modeling. Climate data is observed from various advanced sensor technologies and is used to represent the seasonal changes. Weather data collected from different climate laboratory and advanced computing technologies are used to give valuable information to the world. Meteorological data is most often used to predict the weather and other climate-related phenomena. In addition, climate data is also used for various purposes that lead to a significant development in weather forecasting, rocket launching, and public health. However, climate data collected from various sources are used to identify the seasonal changes. In general, the climate laboratories generate data in unstructured format. Statistical techniques or machine learning algorithms are used to get meaningful information from the raw data. For example, statistical techniques are used to identify the number of precipitation days for a specific region. In this regard, the World Climate Data Monitoring Programme (WCDMP) is developed by WMO’s World Climate Programme (WCP) that focuses on management and collection of large climate data observed from the global climate system [1].