From this need, Natural Language Processing (NLP) developed, that is, the field of research in artificial intelligence that aims to develop models for understanding natural language, that is, the language we use in everyday life. But how do you teach a machine to speak? Language is “taught” to the machine at various levels of granularity: words, relationship between words and their use in a given context, syntactic dependencies and semantic relationships. This approach is integrated by automatic learning algorithms, that is, Machine Learning, and Deep Learning, which try to “imitate” the functioning of the human brain.
Google's first big step in natural language processing came with the introduction of Hummingbird (in 2013) and RankBrain (in 2015). Hummingbird is a Core Update that showed Google's commitment to gaining an increasingly sophisticated understanding of the intent of search queries, in order to provide increasingly relevant results to the user. RankBrain operates under Hummingbird: it is a Deep Learning algorithm that uses mathematical vectors to transform language into entities that a computer can understand, and therefore help the core algorithm to better interpret user queries.
