Machine translation (MT) is the field of translating from a source language to any other target language. Machine translation is one of the most dominant emerging fields in the world today. Machine translation was born in the early 1940s during the Cold War, when there was a great need to decipher or decrypt the exchange of secretly encoded messages between the English and Russian languages. The technology at that time was called “Science of Cryptography”. MT is the key to the success of any new service. Nowadays many IT industries and other private sectors are converging towards MT technology so as to improve existing product services. This has promoted the development of many new models and has led to the development of an open source Moses statistical machine translation (SMT) system which is deployed in various institutes, research projects etc. There are various approaches to machine translation. Say no to plagiarism. Get a tailor-made essay on "Why Violent Video Games Shouldn't Be Banned"? Get an original essay No author provided Rule-based machine translation (RBSMT) is one of the most basic and initial approaches to machine translation. In RBSMT, we need to develop and maintain rules using different grammatical conventions and lexicons and we need to process the rules [1]. Generally the rules are codified by the Linguistic Expert who has the most experience in this field. The advantage of RBMT is that it is very simple and can be easily extended to handle any situation. There are several RBMT approaches. They are transfer-based RBMT, interlanguage-based RBMT, and dictionary-based RBMT. One of the limitations of RBMT is that we, humans, have to create rules for each analysis and generation step, which is a very cumbersome and sometimes tedious task. We must create and develop rules to adapt to the new evolving environment. Therefore, we used the corpus-based approach due to the failure of rule-based approaches. This may be due to the increasing availability of machine-readable text and the increasing capacity of hardware. There are several approaches that use corpus-based MT. They are mentioned as follows:1.2.3.4.5. Examples: Machine-based translation Statistical machine translation Sentence-based statistical machine translation Tree-based statistical machine translation Neural-based machine translation Example-based machine translation (EBMT) is one of the analogy-based SMT approaches. It also relies on the Bilingual Corpus as its main knowledge base [1]. Given a new test source sentence and corresponding reference sentences, it is translated using examples or analogies from the knowledge base. The translated sentences are stored in the knowledge base. This saves the effort of translating each new test sentence. One of the limitations of this approach is that if there is an unmatched test sentence, it must be regenerated from scratch. It cannot use neighboring sentences or words to predict translation [1, 10]. Statistical machine translation (SMT) is a data-driven or corpus-based approach to machine translation. It used supervised and unsupervised technique of machine learning algorithm to train the translation model. The goal of the SMT system is to produce a target translated sentence from a given source sentence. Among all possible candidate translation sentences for a given input sentence, the SMT decoder tries to find the best possible translation. The approach is called the Noisy channel model of Bayes' Theorem. The argmax operation in this model attempts to search for the best translation from the space of all possible ones.