Natural Language Processing (NLP) emerging as a modern branch of science, has been supported by the state-of-the-art techniques like cloud computing, parallel distributed processing, data mining and artificial neural networks. This paper mainly focuses on the crucial contribution of Adaptive Resonance Theory (ART) algorithm in the process of machine translation followed by machine learning. The corpus of vocabulary and the inference engine with the implementation of ART algorithm improves the learning ratio, resulting in the reduction of errors in the subsequent machine translation (MT) exercises. ART algorithm is a self-organizing network, which uses the apriori knowledge for the machine learning. The corpus of Word Web is being enriched by the cumulative process of addition of new vocabulary list using the neighborhood knowledge and calculation of Minimum Edit Distance. The resonance process of the algorithm adds self-organizing nature of the machine learning.