International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 06 | June 2022
p-ISSN: 2395-0072
www.irjet.net
SMART COLLEGE CHATBOT USING NATURAL LANGUAGE UNDERSTANDING Akash Sethiya 1, Dhirsith 2, Raisa Anjum 3 Department of Computer Science Engineering, JAIN UNIVERSITY, Bangalore, Karnataka- India ---------------------------------------------------------------------***--------------------------------------------------------------------Moreover, it may lead to a ‘Lagging Effect’ or low Abstract - Chatbots are not a recent development. They 1,2,3
are simulations which can understand human language, process it, and interact back with humans while performing concrete tasks. A chatbot, for example, can work as a customer service representative. The first chatbot was engendered by Joseph Wiesenbaum in 1966, designated Eliza. It all commenced when Alan Turing published an article designated “Computer Machinery and Intelligence”, and raised an intriguing question, “Can machine cerebrate?”, Since then, we've seen several chatbots that appear to be more naturally conversational and technologically advanced than their predecessors. These advancements have led us to an era where conversations with chatbots have become as mundane and natural as with another human.
evaluation precision, if all the records are treated equipollently, as the invocation contexts of different records are not equipollent. Given these challenges, a novel approach designated Partial-HR (Partial Index Terms—big data, cloud, context-vigilant accommodation evaluation, historical QoS record, weight Historical Records-predicated accommodation evaluation approach) is put forward in this paper. In Partial-HR, each historical QoS record is weighted predicated on its accommodation invocation context. Afterward, only partial paramount records are employed for quality evaluation. Determinately, a group of experiments is deployed to validate the feasibility of our proposal, in terms of evaluation precision and efficiency.
Today, virtually all companies have chatbots to engage their users and accommodate customers by catering to their queries. As per a report by Gartner, Chatbots will be handling 85% of the customer accommodation interactions by the year 2020. Withal, 80% of businesses are expected to have some marginally chatbot automation by 2020 (Outgrow 2018). We virtually will have chatbots everywhere, but this doesn’t compulsorily denote that all will be well-functioning. The challenge here is not to develop a chatbot, but to develop a well-functioning.
2. EXISTING SYSTEM The success of big data applications such as medical data analysis has been aided by cloud computing. The enormous resources provided by cloud platforms might considerably improve the QoS (quality of service) of services that process large data. However, because of unreliable networks or false advertising, service providers' QoS is not always believed. As a result, it becomes necessary to evaluate service quality in a trustworthy manner, based on the services' past QoS data [1]. However, if all of a service's records are recruited for quality review, the evaluation efficiency would be low, and it would be unable to fulfil users' need for a timely answer [2]. Furthermore, if all records are considered similarly, it may result in 'Lagging Effect' or low assessment accuracy, because the invocation contexts of various data are not precisely the same.[2] Considering these issues, this work proposes a unique technique called Partial-HR (Partial Index Terms—big data, cloud, context-aware service evaluation, historical QoS record, weight Historical Records-based service evaluation approach) [3]. Each historical QoS record in Partial-HR is weighted based on the context of its service invocation. Following that, only partial significant records are used for quality evaluation. Finally, a set of tests is carried out to test the practicality of our idea in terms of evaluation accuracy and efficiency [4]. Chatbot is software that uses artificial intelligence to converse with humans. These programmes are used to do duties such as replying swiftly to users, educating them,
Key Words: Chatbot, AI, Automation
1.INTRODUCTION Cloud computing has promoted the prosperity of astronomically immense data applications such as medical data analyses. With the abundant resources provisioned by cloud platforms, the QoS of accommodations that process immensely colossal data could be boosted significantly. However, due to unstable networks or fake advertisements, the QoS published by accommodation providers is not always trusted. Ergo, it becomes an indispensability to evaluate the accommodation quality in a trustable way, predicated on the services’ historical QoS records. However, the evaluation efficiency would be low and cannot meet users’ expeditious replication requisite, if all the records of accommodation are recruited for quality evaluation.
© 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 2050