Machine translation is the process of using computer software to translate text from one natural language to another. The first proposals for machine translation were presented in 1949 and were based on information theory. One of the earliest public demonstrations of machine translation was the Georgetown-IBM experiment held in 1954. Early systems used large bilingual dictionaries and hand-coded rules. Machine translation has advantages like speed and cost but original documents may not be 100% accurate. There are two main approaches - rules-based systems using language rules and dictionaries, and statistical machine translation using statistical models derived from analyzing bilingual text. Popular machine translation applications and services include Google Translate, Bing Translator, and SYSTRAN.