The Informative Vector Selection in Active Learning using Divisive Analysis
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2017, Vol 8, Issue 10
Abstract
Traditional supervised machine learning techniques require training on large volumes of data to acquire efficiency and accuracy. As opposed to traditional systems Active Learning systems minimizes the size of training data significantly because the selection of the data is done based on a strong mathematical model. This helps in achieving the same accuracy levels of the results as baseline techniques but with a considerably small training dataset. In this paper, the active learning approach has been implemented with a modification into the traditional system of active learning with version space algorithm. The version space concept is replaced with the divisive analysis (DIANA) algorithm and the core idea is to pre-cluster the instances before distributing them into training and testing data. The results obtained by our system have justified our reasoning that pre-clustering instead of the traditional version space algorithm can bring a good impact on the accuracy of the overall system’s classification. Two types of data have been tested, the binary class and multi-class. The proposed system worked well on the multi-class but in case of binary, the version space algorithm results were more accurate.
Authors and Affiliations
Zareen Sharf, Maryam Razzak
Keywords
Related Articles
- EP ID EP260750
- DOI 10.14569/IJACSA.2017.081009
- Views 89
- Downloads 0
Ranking Documents Based on the Semantic Relations Using Analytical Hierarchy Process
With the rapid growth of the World Wide Web comes the need for a fast and accurate way to reach the information required. Search engines play an important role in retrieving the required information for users. Ranking al...
Method for Uncertainty Evaluation of Vicarious Calibration of Spaceborne Visible to Near Infrared Radiometers
A method for uncertainty evaluation of vicarious calibration for solar reflection channels (visible to near infrared) of spaceborne radiometers is proposed. Reflectance based at sensor radiance estimation method for sola...
Experimentation for Modular Robot Simulation by Python Coding to Establish Multiple Configurations
Most of the Modular Self-reconfigurable (MSR) robots are being developed in order to have the capability of achieving different locomotion gaits. It is an approach of robotic system which involving a group of identical r...
Workshop Session Recordings on Green Volunteering Activities of Students in a Disadvantaged Area According to the Good-Hearted Vocation Teacher to Support Itinerant Junk Buyers
This project was aimed to provide workshop session recordings on green volunteering activities of students in one disadvantaged area under the bridge of zone 1, Pracha-Utit Road 76, Toong-kru District, Bangkok where the...
Electronic Health as a Component of G2C Services
This paper explores electronic health as a segment of electronic government. International practice in electronic health field and electronic health strategies adopted in Europe are analysed. Current practices in deliver...