AI & NLP…decoding the enigma…
Communication has been the mainstay of all human interaction and Language continues to play a critical role in representing our intelligence. For thousands of years humans have tried to teach language and our ways of communication to animals with limited success so we created machines which could be human, do things that we do, think and communicate like we do. Our progress with computing technology, global connectivity and now Artificial Intelligence is making it a reality.
Computers conventionally understand instructions or commands often through programmed inputs that are accurate and structured. Human speech on the other hand is complex, contextual and ambiguous. In AI the ability of a machine to understand, process and respond to human languages like English is called Natural Language Processing (NLP). With the advancement of Mobile computing, SPEECH as a medium of input is rapidly overtaking TEXT based inputting.
NLP has two facets to its functioning,
• Understanding the language of instruction: Involves representing the given input in natural language into functional components and the process of analyzing different characteristics of the language. This process is called Natural Language Understanding – NLU.
• Generation of Natural Language: It is the process of identification of relevant content, producing meaningful construction of the language like phrases and sentences and representing it into Text or Speech. This is called Natural Language Generation – NLG.
Natural Language is highly contextual and has a very complex form and structure. Derivation of meaning is based on perception and historical reference and therefore very challenging to be condensed into a program. NLG hence is very testing and difficult.
NLP involves multiple facets such as recognizing Sounds/Speech, understanding the Structure and Meaning of words, understanding the Construction of Sentences and phrases, understanding interpretations of words and sentences, and understanding the significance of context and reference in Natural Language.
Early NLP systems were based on complex sets of hand-written instructions but the introduction of machine learning – ML, techniques has revolutionized language processing. Recent developments in deep learning techniques have resulted in significant progress in many natural language tasks, like language modeling, parsing and others.
It would not be surprising that in the future computer programming as we know them, (the C+’s and Java’s) would cease to exist and machines would speak only “human languages”…