چكيده به لاتين
Energy consumption and carbon dioxide emissions are among the major environmental and economic challenges facing the world today. With the increasing demand for energy due to population growth and industrial development, energy consumption from fossil resources such as oil, natural gas, and coal has increased. These fossil resources are the main responsible for the massive CO2 emissions that lead to climate change and global warming. Climate change can have devastating effects such as global temperature increase, melting of polar ice caps, and severe changes in weather patterns. Due to the importance of these issues, it is necessary to examine and analyze energy consumption, which is the subject of this research. In the meantime, data mining, as a powerful tool, can play an important role in reducing energy consumption and CO2 emissions. By using data mining techniques, hidden patterns and trends in big data can be identified and energy efficiency can be improved. These techniques can help identify weaknesses in energy systems, optimize industrial processes, and reduce energy losses. Data mining can also be effective in predicting energy consumption and CO2 emissions, assessing environmental impacts, and developing sustainable policies. By analyzing energy consumption and CO2 emission data, effective solutions can be provided to reduce these two, including optimizing energy consumption in industries, transportation, and buildings, and increasing the use of renewable energy sources. Therefore, data mining as a key tool can help reduce energy consumption and carbon dioxide emissions, protect the environment, and achieve sustainable development. In this research, the CRISP-DM approach has been used to advance the data mining project. With the help of machine learning algorithms such as K-Means and machine learning, data related to Australian commercial buildings has been categorized and classified. The innovation in this research is the use of dynamic boundary. In fact, the dataset itself was used to determine whether the amount of carbon dioxide consumption and emissions was low or high, and the classification logic was obtained from the data itself. Another innovation is the forward and backward labeling of data. Ultimately, the result of this research was the classification of Australian commercial buildings into 5 clusters. One building was placed in a separate cluster and the other two buildings in another unique cluster, with these three buildings having the highest energy consumption and carbon dioxide emissions. The other buildings were also placed in three clusters.
