Year: 2022
Type: book-chapter
Abstract: Support vector machine classification algorithm has been deeply studied in the field of intelligent information processing with its solid theoretical foundation and excellent classification performance, and it has been widely used in real life, such as face recognition, text classification, sentiment analysis, and spam filtering, etc. The time complexity of the classic support vector machine algorithm is proportional to the square of the data size, and this poses a serious challenge to the calculability, effectiveness and efficiency of the support vector machine algorithm for massive data. In this paper, a large-scale support vector machine classification algorithm based on the granulation mechanism is proposed from the granulation and fusion of data information. A data granulation mechanism is firstly constructed using binary tree and classification label information, as well as the core sample set is obtained by combining with instance selection method, and then the core sample set and the remaining samples construct multiple weak support vector classifiers; finally draw on the idea of granularity fusion to generate a strong classifier. The experimental results on the standard data set verify the effectiveness and efficiency of the proposed algorithm, and it provides new research ideas and processing methods for processing massive data.
DOI: 10.3233/atde220030
Institution: Shandong Agricultural University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
Awards: G2403720365, G501516934
OpenAlex ID: W4213174868