Open AccessPerformance evaluation of health recommendation system based on deep neural network
Author(s) -
Gauri Sood,
Neeraj Raheja
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1131/1/012013
Subject(s) - computer science , deep learning , depiction , jitter , artificial intelligence , artificial neural network , latency (audio) , power consumption , machine learning , layer (electronics) , power (physics) , telecommunications , philosophy , linguistics , physics , chemistry , organic chemistry , quantum mechanics
Deep learning has developed as an innovative zone of machine learning and data mining exploration part. Controlled or unconfirmed methodologies which contain of a number of layers of handling which form a hierarchy are castoff for preparation in deep learning. Every succeeding layer mines an ever more intellectual depiction of the input data and shapes upon the depiction from the preceding layer, usually by calculating a nonlinear alteration of its input. The constraints of these alterations are adjusted by preparation of the prototypical on a dataset. A deep learning prototypical studies better depiction as it is delivered with more volumes of data. Key objective of using deep learning methods in recommender schemes is to lower time complexity and to increase the accurateness of formed expectations. In this paper, performance of planned HRS is evaluated by Arbitration Time, Latency Time, Jitter, Execution Time, Network Bandwidth Consumption, Power Consumption, Training Accuracy and Testing Accuracy.