School of Management, Zhejiang Shuren University and Institute of Modern Services, Hangzhou 310015, Zhejiang, China.
Xi'an International University, Xi'an 710077, Shaanxi, China.
Comput Intell Neurosci. 2022 Aug 21;2022:7738811. doi: 10.1155/2022/7738811. eCollection 2022.
When studying intelligent community and property management system, network service has an important impact on intelligent community construction and property management service. How to use machine learning and other technologies to improve the network service of intelligent community and integrate it into real estate property management is worthy of further research. This paper introduces the basic model of machine learning and proposes a network data prediction model based on time series. For the time dimension, an improved prediction algorithm model of machine learning is proposed. For mobile data allocation, from the perspective of ensuring the current and future continuity of the spectrum after spectrum allocation, this paper proposes a spectrum allocation algorithm based on the joint measurement of time domain and frequency domain. In addition, the VP-tree algorithm is used to construct the spatial vector relationship of the intelligent community. At the same time, in the time trend and periodicity of the mobile data in the intelligent community network, the attention mechanism is introduced to realize the distribution of mobile data and traffic prediction in the intelligent community by machine learning. This paper analyzes the requirements of the property management system, designs the property management information system including the field subsystem layer, data acquisition layer, and cloud service layer, introduces the property management module and customer service module in detail, and carries out the system test. The test results show that the system runs well. Finally, aiming at the problems existing in the property management industry, this paper puts forward the development strategy of property management.
在研究智能社区和物业管理系统时，网络服务对智能社区建设和物业管理服务有重要影响。如何利用机器学习和其他技术来提高智能社区的网络服务，并将其融入房地产物业管理中，值得进一步研究。本文介绍了机器学习的基本模型，并提出了一种基于时间序列的网络数据预测模型。针对时间维度，提出了一种改进的机器学习预测算法模型。在移动数据分配方面，从确保频谱分配后时域和频域联合测量的频谱连续性的角度出发，提出了一种基于时间和频率的频谱分配算法。此外，使用 VP 树算法构建智能社区的空间向量关系。同时，在智能社区网络中移动数据的时间趋势和周期性方面，引入注意力机制，通过机器学习实现智能社区的移动数据分布和流量预测。本文分析了物业管理系统的需求，设计了包括现场子系统层、数据采集层和云服务层的物业管理信息系统，详细介绍了物业管理模块和客户服务模块，并进行了系统测试。测试结果表明，该系统运行良好。最后，针对物业管理行业存在的问题，提出了物业管理的发展策略。
