Ensemble classifier using Fuzzy C5.0 decision Tree for Information Retrieval

Abstract

The exposure of millions of individual remarks or thoughts has increased greatly as the global web boom has spread and the number of pages, review boxes, blogs, websites and other platforms in social networks has risen. There are many subjects on-line views or thoughts such as books, films, consumer goods, motorcycles, politics, etc. This results in a widespread use of opinion interpretation and domain adaptation. It is impossible for modern cross-domain opinion mining to predict the exact result of the occurrence in a cross-domain. Features such as text length and stylistic features should be taken into account. The sentences can be shown in different domains differently of costs and computations for the respective domain are very high. The uncontrolled models are considered for predictions that produce low precision and cannot be adapted to changing environments. Therefore, the ensemble classification is developed to handle data complexity and achieve optimum results using the Fuzzy C5.0 decision tree.