Virtual Assistants are playing a vital role in every human’s life. Setting alarms, playing wanted music, face, speech, and handwritten recognition, reminding, responding, diagnosing diseases, etc. were done by virtual assistants. Every individual who is using a virtual assistant share their ideas/views/daily duties to their virtual assistants to train them. Once it got trained, these assistants assist users as accurately as they can. Machine Learning algorithms were fed into a virtual assistant for their efficient performance. These assistants were mainly used in education, medical, agriculture, media, social-networking sites, self-driving cars, banking, fraud detecting, etc. Hence, Virtual assistant’s mimics users by machine learning technologies. The challenges faced by virtual assistants using machine learning methods are summarized in this survey paper and also describes a literature survey of current virtual assistants. Artificial neural networks are used in prediction models by VAs are also discussed. Interaction between humans and machines is one of the tasks of VA. Natural language processing (NLP) is used to solve this task. NLP and its applications used in VAs are discussed. This paper also compares different algorithms and techniques used by current VAs.
Volume 12 | Issue 6
Pages: 1552-1558
DOI: 10.5373/JARDCS/V12I2/S20201352