virtual personal assistant
645 Views

GC: n

CT: The recent progress in Machine Learning that have made possible the rise of Virtual Personal Assistant, come from a specific approach to Machine Learning: Neural Networks and more specifically Deep Learning Neural Networks. Indeed, these new approach benefit the many different tasks which are at the core of natural language processing.

S: EC – https://goo.gl/Y1qoHv (last access: 12 December 2018).

N: 1. virtual (adj): From late 14c., “influencing by physical virtues or capabilities, effective with respect to inherent natural qualities,” from Medieval Latin virtualis, from Latin virtus “excellence, potency, efficacy,” literally “manliness, manhood”. In the computer sense of “not physically existing but made to appear by software” is attested from 1959.
– personal (n): From late 14c., “pertaining to the self,” from Old French personal (12c., Modern French personnel), from Late Latin personalis “pertaining to a person,” from Latin persona.
– assistant (n): From mid-15c., assistent “one who helps or aids another,” from Latin assistentem (nominative assistens), noun use of present participle of assistere “stand by, attend”. The spelling changed in French then (16c.) in English.

  • Abbreviation in use for “virtual personal assistant”: VPA.

2. Software program meant to interact with an end user in a natural way, to answer questions, follow a conversation and accomplish different tasks.
3. Types of virtual personal assistants:

  • The term, is also commonly used to describe contract workers who work from home doing administrative tasks typically performed by executive assistants or secretaries.
  • Virtual assistants can also be contrasted with another type of consumer-facing AI programming, called smart advisers. Smart adviser programs are subject-oriented, while virtual assistants are task-oriented.
  • Virtual assistant devices and technology are typically cloud-based programs that require internet-connected devices and/or applications to work. Three such applications are Siri on Apple devices, Cortana on Microsoft Devices and Google Assistant on Android devices.

4. For voice-based input, like those from VPA such as Siri, the first need consists of converting speech to actionable data. This “speech-to-text” step, (also called speech recognition), is of paramount importance: if the input is not correctly recognized, all following steps are useless. Indeed, even an error on one word is very likely to result in an inaccurate answer. Once a sequence of spoken words is successfully converted to a text form, numerous and very complex tasks remain:

  • Syntax analysis (or parsing) is used to analyze and identify the structure of the sentence, based on knowledge of grammar.
  • Semantic analysis is used to reach a partial representation of the meaning of the sentence, based on the knowledge of the meaning of words.
  • Pragmatic analysis is used to reach a final representation of the meaning of the sentence, based on information about the context.

5. For the vast majority of applications related to natural language, an answer (oral and/or written) is given back after a query from the user. Question answering deals with information retrieval (using information on the Internet, or in an application) and generating a correct sentence, before the last step of speech synthesis.
6. Cultural Interrelation: One can mention Siri from Apple, and Alexa from Amazon.

S: 1. OED – https://goo.gl/FvmxPo; https://goo.gl/uMQoG2; https://goo.gl/UiG2nx (last access: 12 December 2018). 2. EC – https://goo.gl/Y1qoHv (last access: 12 December 2018). 3. TechT – https://goo.gl/qAv2Dt (last access: 12 December 2018). 4 & 5. EC – https://goo.gl/Y1qoHv (last access: 12 December 2018). 6. Apple – https://goo.gl/RbXacu (last access: 12 December 2018); Amazon – https://goo.gl/7GFtNx (last access: 12 December 2018).

SYN: virtual assistant, personal assistant, digital assistant, intelligent assistant.

S: TERMIUM PLUS – https://goo.gl/iCSqzF (last access: 12 December 2018).

CR: artificial intelligence, cognition , cognitive science, computational intelligence, computer science, deep learning, e-learning, intelligent agent, intelligent system, machine learning, semantic network.