Semi-Automated Intelligent Systems

Chinam, Srihari Akash and Reddy, Subba N V and Prema, KV (2017) Semi-Automated Intelligent Systems. In: Internation conference on Advances in Computing ,Communication and Informatics, 13/09/2017, MIT, Manipal. India.

[img] PDF
842.pdf - Published Version
Restricted to Registered users only

Download (808kB) | Request a copy


An intelligent system uses machine learning algorithms to provide outputs to every input provided. The introduction of emotions in intelligent systems is required to create systems that are more similar to human beings and thus more reliable. In this paper, the idea of introducing the emotion ‘uncertainty’ in Intelligent Systems is proposed. A Semi-Automated Intelligent System is introduced in this paper, which combines a Naïve Bayesian Classifier, a Random Forest Classifier and a Multi Layer Perceptron using a Multi Model Strategy to introduce uncertainty. When used individually the classifiers had errors in the range of 6-7% but when combined as the Semi-Automated Intelligent System, the false predictions were kept under 0.2%. The paper also discusses several preprocessing techniques that were applied on the text documents to ensure effective analysis of data

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Intelligent Systems, Naïve Bayes, Multi Layer Perceptron, Random Forest, Text Classification
Subjects: Engineering > MIT Manipal > Computer Science and Engineering
Depositing User: MIT Library
Date Deposited: 06 Oct 2017 08:43
Last Modified: 25 Nov 2017 08:56

Actions (login required)

View Item View Item