Ranking of Classifiers based on Dataset Characteristics using Active Meta Learning

Abstract

Classification is a machine learning technique which is used to categorize the different input patterns into different classes. To select the best classifier for a given dataset is one of the critical issues in Classification. Using cross-validation approach, it is possible to apply candidate algorithms on a given dataset and best classifier is selected by considering various evaluation measures of Classification. But computational cost is significant. Meta Learning automates this process by acquiring knowledge in form of Meta-features and performance information of candidate algorithm on datasets and creates a Meta Knowledge Base. Once Meta Knowledge Base is generated, system uses k-Nearest Neighbor as a Meta Learner that identifies the most similar datasets to new dataset. But generation of Meta Example is a costly process due to a large number of candidate algorithms and datasets with different characteristics involved. So Active Learning is incorporated into Meta Learning System that reduces generation of Meta example and at the same time maintaining performance of candidate algorithms. Once the training phase is completed based on Active Meta Learning approach, ranking is provided based on Success Rate Ratio (SRR) method that considers accuracy as a performance evaluation measure.

Heterogeneous Ensemble Combination Search Using Genetic Algorithm for Class Imbalanced Data Classification

“…Therefore, by combining the outputs of many classifiers, the ensemble of classifiers strategically enhances the power of the committee, as an aggregated method, to achieve better prediction accuracy than any of the individual classifiers could alone [ 2 ]. Moreover, the generalisation capability of each classifier also differs depending on the data characteristics and relations (such as dimensionality, class distributions, noise ratio and so on) [ 3 ]. In recent years, many difficult “real-world” datasets have been characterised as imbalanced [ 4 ], where sample distributions among classes are skewed.…”

A survey on prediction approaches for epidemic disease outbreaks based on social media data

“…Numerous analysts are utilizing statistical and data mining procedures for the analysis of coronary illness. Bhatt et.al [33] utilized Naive Bayes a basic information mining system to show better outcome and precision. None of the system predicts heart disorders subject to risk factors, for instance, age, family heritage, hypertension, diabetes, alcohol confirmation, tobacco smoking, heftiness or physical inaction, raised cholesterol, etc.…”

Recent advances in scaling‐down sampling methods in machine learning

“…Uncertainty sampling is the most commonly used and extensively researched active learning framework and has been shown to work successfully in a variety of applications including text classification, robotics, computer vision, meta‐learning for algorithm recommendation, image sequence recognition, data stream regression, and several others. In the uncertainty sampling framework, the learning algorithm is typically provided with a sample of the labelled instances and an initial model is learned.…”