When we study natural language processing, we must first point out the various difficulties attributed to writing programs that "understand" natural language. We note the ambiguity, imprecision, and informality inherent in natural language and the gap between this ambiguity and the precision and formality of languages ​​used in the development of analysis, design, and programming software. Bridging the gap between informal natural language and formal conceptual models is essential with applications designed to “understand” natural language. Second, we will present some of the applications developed to “understand” natural language over the decades. We'll look at the groundbreaking work done in the 1960s with ELIZA, which used a rudimentary parser to imitate a therapist throughout the 1980s, as well as the great advances IBM made in language translation interfaces until the birth of WATSON, the superhuman. interfacing program that beat two seasoned human players with its massive knowledge bank and SIRI with its voice recognition capabilities. Finally, we will study the different techniques and theories used in natural language processing to create applications that “understand” human language. Namely, theories of parsing, semantics, word sense disambiguation, corpus linguistics and lexicon. Barriers to Understanding Natural Language in Applications In any interaction with a computer, we use natural language to describe our needs and problems. The natural language we use to describe our problems and demands is often complex, vague and ambiguous. Human sentences are complicated and complex. This is especially true when they contain clauses and expressions that describe and relate multiple objects, conditions, and actions. Sen...... middle of paper...... its structure from its constituent elements. We have highlighted that ambiguity in text is a major problem encountered with NLP ambiguity in natural language and that word meaning disambiguation is the theory developed and implemented to counter this major obstacle. Finally, we studied semantics or the study of the meaning of words. We have noted that syntax is as important as understanding natural languages. We noted that any application requires semantic theory to guide understanding and create an effective human interface. By studying the various barriers associated with understanding natural language as well as the applications in the discipline over the decades and finally some of the techniques or methodologies implemented by these applications to successfully understand natural language, we have been able to study in a manner in-depth and comprehensive understanding of natural language..