Cognitive (Internet of) Things by Arvind Sathi

Author:Arvind Sathi
Language: eng
Format: epub
Publisher: Palgrave Macmillan US, New York

In all cases, most probably the user needs to reset password. In the first case, the question could be interpreted in other ways as well. May be the user cannot locate the login screen, or it may also be possible that he has forgotten his user ID. As the Cognitive Thing analyzes the speech utterances and formulates the sentence, how would it decide if the password needs to be reset, and respond with “Would you like me to reset your password?”

Natural Language Processing (NLP) is the area of artificial intelligence that examines human language and the different techniques to analyze, systematically process, and understand language. Natural language has an inner structure that is not explicit, and part of the task of automatic processing is to organize the language.

The process of analyzing language can vary greatly from language to language, as language incorporates the world vision and common sense of their speakers. Some languages share common roots and evolved influencing each other (for example, Spanish and Portuguese), while others can be independent of each other (consider Arabic, Russian, and Japanese).

Text analytics or text mining refers to the tasks involved in analyzing unstructured text and extracting or generating structured data, and performing some level of interpretation on that data.

There are many ways of collecting intent from a sentence, and each approach is pertinent to certain use cases. In the simplest approach, a key word search can match words used in a sentence to probable intent. For example, a number of traditional Interactive Voice Response (IVR) units graduate from a multi-layered menu to free-form questions and key words to guess the intent. The response can be in the form of a confirmation or disambiguation to match the most probable intent. Once the match is made the Cognitive Thing can respond by executing the stated action. The second approach uses statistical methods in matching sentences against a library of sentences. Given a large training set, the library provides many ways in which the same intent can be expressed. As long as the sentence ingested matches one of the sentences in the library, the Cognitive Thing can confirm, disambiguate, and respond. The comprehensive approach involves parsing the sentence into its components as tuples, such as subject–verb–object and understanding the intent by matching these tuples to known intents. This approach facilitates contextual interpretation, as a pronoun in a subsequent statement may refer to a noun in the previous statement. The sentence can be ambiguous and may be parsed in different ways depending on the context of the sentence.8

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.