A Multi-source Data Fusion Analysis Model in the Research on Public Sentiment Index

Abstract

Abstract At present, China’s economy is continuously developing and making progress. The deep integration of society and big data artificial intelligence and other technologies is constantly promoting the construction of new smart cities. The major livelihood service systems have provided great convenience to the people. At the same time, they have accumulated massive livelihood data. Mining the value of data to solve people’s main problems has been widely concerned by the government and the community. However, the data structure of each system is different. The data is noisy and the systems cannot be interconnected, which leads to the one-sided and single results of data analysis, and it is impossible to use these data to obtain effective results and conclusions. This paper proposes a Public Sentiment Index that reflects the intensity of public appeals. A new multi-source data fusion analysis model is proposed by using a multi-source data fusion method. By constructing a public sentiment database and a public sentiment index model, it can truly reflect the changes in people’s demands. This research can provide a reliable and scientific basis for the government to improve decision-making capacity and service quality.