scispace - formally typeset
Search or ask a question
Journal ArticleDOI

An automated ontology learning for benchmarking classifier models through gain-based relative-non-redundant feature selection: a case-study with erythemato-squamous disease

TL;DR: The Gain based Relative-Non-Redundant Attribute selection approach for diagnosis of ESD yielded 98.1% classification accuracy with Adaboost algorithm that executed J48 as the base classifier and an optimal feature set comprising of 19 selected features.
Abstract: Erythemato-squamous disease (ESD) is one of the complex diseases in the dermatology field, the diagnosis of which is challenging, due to common morphological features and often leads to inconsistent results. Besides, diagnosis has been done on the basis of inculcated visible symptoms pertinent with the expertise of the physician. Hence, ontology construction for prediction of erythemato-squamous disease through data mining techniques was believed to yield a clear representation of the relationships between the disease, symptoms and course of treatment. However, the classification accuracy required to be high in order to obtain a precise ontology. This required identifying the correct set of optimal features required to predict ESD. This paper proposes the Gain based Relative-Non-Redundant Attribute selection approach for diagnosis of ESD. This methodology yielded 98.1% classification accuracy with Adaboost algorithm that executed J48 as the base classifier. The feature selection approach revealed an optimal feature set comprising of 19 selected features.
Citations
More filters
Journal ArticleDOI
TL;DR: An extensive knowledge-based approach using a reasoning mechanism based on competency questions for individual approaches to determine their ontology learning method profiles is proposed, aiming to provide a more efficient OL solution from text.

7 citations

Book ChapterDOI
01 Jan 2023
TL;DR: In this paper , the authors have experimented with a set of various standard machine learning algorithms ensemble through the stacking process by gathering the data from web source through the usage of different tools that could directly understand the logic from the pages and retrieve the relevant information to the user.
Abstract: ESD is a kind of complicated skin disease which has familiar morphological properties that makes diagnosis of this disease complex, and the outcomes were usually inconsistent. Generally, the medication for this problem is depended on indoctrinated observed symptoms related to those results of a physician. Here, an ideal construction for ESD was developed to secure trustworthiness, flexibility, dealing with computational time, labor, and expertise to reduce the manual error. This research paper has been experimented with a set of various standard machine learning algorithms ensemble through the stacking process by gathering the data from web source through the usage of different tools that could directly understand the logic from the pages and retrieve the relevant information to the user. The researchers have claimed that they have used various classification algorithms over the collected data and stated that the stacked ensemble framework of DT classifier and KNN has attained a greater performance of 95.8%. The developed model is said to be an automated model that could actively retrieve data and perform various modules on it. In existing approaches, either researcher implemented as single classification algorithms by analyzing the data or few implemented as ensemble mechanisms without performing the elimination of irrelevant data. The stacking algorithm with tuned parameters not only helps in improvement but also helps in taking good decisions at the time of diagnosis because these are data-dependent algorithms.