Article Preview
TopIntroduction
With the rapid development of information technology and the advent of the cloud era, the term big data is mentioned and recognized increasingly and people are becoming increasingly aware of the importance of data and trying to explore their hidden value. The term “big data” is usually used to describe the massive data generated in the era of information explosion, which is characterized by the massive information generated by social networks (Sravya, 2020). In brief, a social network is a structure established between people through friends, blood relatives, transactions, interests, links, and other relationships. Social networks are formed based on the relationships between their members, and their purpose is to establish social contact (Zhang et al., 2022). Users share personal information freely through social networks and can also obtain and disseminate other information from other people conveniently and quickly (Bharath et al., 2021; Saraswathy et al., 2021). Due to their influence and the breadth and depth of communication available, people express their emotions and opinions on social networks more and more widely. The massive emotional information flow hidden in social networks seems to be fragmented and disordered, but it may provide huge value when mined through appropriate techniques. This makes research on the emotion classification of texts based on massive data very important (Al, 2021).
Among social networks, microblogs are favored because of their convenient use, rapid dissemination, and strong interaction capabilities. Consequently, they have become one of the mainstream online social networking platforms. However, traditional text emotion classification research is carried out on a single machine. Given the massive data appearing in social networks, traditional emotion analysis algorithm implementations running on single machines cannot accomplish the task of emotion classification quickly, and their time efficiency and limited scalability become a bottleneck. It is therefore necessary to study calculation modes suitable for emotion classification tasks applied to massive data (Samah, 2021). The emergence and development of cloud computing provide a new solution for emotional classification tasks applied to massive data, which allows the shortcomings of traditional single-machine computing approaches to be overcome. The feasibility of emotion classification tasks on massive data is enhanced through distributed emotion classification algorithms and architectures (Mutanov et al., 2021). However, existing cloud computing platforms and methods do not take into account computational efficiency and analysis accuracy, so it is necessary to conduct in-depth research on these aspects.