|
| Abstract: |
| In recent times, an abrupt upswing has emerged within the data mining domain, particularly within the sphere of sentiment analysis. Encompassing diverse dimensions such as sentiment extraction, subjectivity categorization, opinion summarization, and spam detection, sentiment analysis, also acknowledged as Opinion Mining (OM), undertakes a mathematical exploration into the perspectives, emotions, evaluations, and conduct of individuals toward entities, encompassing products, services, individuals, and events. The advent of web technology tools empowers users to liberally articulate their opinions across varied online platforms, culminating in the generation of a substantial corpus of invaluable yet unstructured data. This exposition scrutinizes the importance of harnessing this data, delving into the intricate process of refining and metamorphosing it for subsequent operations like classification and aspect-oriented sentiment analysis. The crux of the discourse centers on a thorough scrutiny of four data mining algorithms deployed for prognostication and categorization, providing insights into their efficacy within the domain of sentiment analysis. This investigation transcends into pragmatic applications, challenges encountered in the field, and potential trajectories ahead, culminating in a nuanced comprehension of the dynamic panorama of sentiment analysis and its ramifications. |
| Key words: data mining OM classification |
| DOI：10.11916/j.issn.1005-9113.2024006 |
| Clc Number:TP391 |
| Fund: |
