ABSTRACT: Antigenic drift in epitopic part of a virus, especially for Hepatitis C Virus is a well-known fact. However, this problem can be overcome due to the fact that the epitopes are dispersed amongst a few proteins that are already filtered out by researchers. Minor changes or variation in the sequential structure of this protein group results in amendment of their epitope structures which ultimately renders any vaccine or drug ineffective against the target organism. Therefore the problem is reduced to first revisiting the identification steps of altered sequences of these 10 proteins which is quite achievable experimentally and secondly identification of epitopic part out of these altered peptide sequences. Hidden Markov models (HMMs) have been comprehensively deployed in analysis of bio-molecular sequences. The work presented in this paper deals with the recognition step of epitopes through probabilistic machine learning model Viterbi of HMM and achieved significantly high efficiency towards this direction. As a consequence, a considerably high precision was obtained for T-cell based linear epitope recognition.