Improving the performance of professional blogger's classification

Predicting Influential Blogger’s by a Novel, Hybrid and Optimized Case Based Reasoning Approach With Balanced Random Forest Using Imbalanced Data

“…The main objective of this research is to model, develop, and implement an adaptive framework for the identification of influential bloggers by using blogger features and blog content features which can boost prediction results. It is an effort which extends our previous investigations [19][20][21] in this domain. This work can be differentiated with respect to the prior works by incorporating unstructured data handling along with labeled data in the suggested framework, provided the data is imbalanced.…”

“…However, IB-CBR algorithm is also used for results comparison to investigate it on imbalanced data, although it is not tested and designed for this purpose. The reason behind using this algorithm is its outstanding performance among our former research works [19][20][21]. In this work, standard imbalanced data performance evaluation metrics such as Spec, F-measure, G.M., and ROC area under the curve are used for experimental outcomes evaluation.…”

“…By using a number of ML techniques, they have observed 85% accuracy gain for blogger classification by using RF classifier and Nearest Neighbor classifier along with standard way of cross validation. In this chain of experiments, 2% gain in accuracy is grasped with Artificial Neural Network (ANN) [21]. After investigating already available ML techniques on a number of standard datasets and highlighting a few among them for better blogger classification, a new classifier named Influential Blogger based Case-Based Reasoning (IB-CBR) is offered with adaptive capabilities [19].…”

“…It outperforms previously highlighted classic ensemble method such as RF, Non-ensemble methods such as Nearest Neighbor algorithm and adaptive models such as ANN. However, the aforementioned competitor algorithms were the outcomes of extensive experimental studies which show their outstanding performance over several ML techniques using multiple standard datasets [20,21]. In this work, we shall compare the CBR outcomes with the proposed model in imbalanced dataset domain.…”

An Adaptive Model for Identification of Influential Bloggers Based on Case-Based Reasoning Using Random Forest

“…It was found that Random Forest and Nearest Neighbor achieved 85% accuracy for blogger classification. Besides, 2% gain in accuracy is seen by using Artificial Neural Network [33] coupled with cross validation of results in a standard way. Besides, some of context specific studies are also available in this domain.…”