Title: ExpCoot_DQNN: a novel deep quantum neural network approach for incremental on Amazon reviews

Authors: Konda Adilakshmi; Malladi Srinivas; Anuradha Kodali; V. Srilakshmi

Addresses: Department of CSE, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Andhra Pradesh, India; Department of CSE, Gokaraju Rangaraju Institute of Engineering and Technology (GRIET), Bachupally, Hyderabad – 500090, India ' Department of CSE, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Andhra Pradesh, India ' Department of CSE, Gokaraju Rangaraju Institute of Engineering and Technology (GRIET), Bachupally, Hyderabad – 500090, India ' Department of CSE, Gokaraju Rangaraju Institute of Engineering and Technology (GRIET), Bachupally, Hyderabad – 500090, India

Abstract: This article proposed a novel framework for the sentimental analysis with the Amazon review document using the proposed exponential coot_deep quantum neural network (ExpCoot_DQNN). Here, the DQNN is utilised for sentimental analysis of the document as positive or negative and computes the error. Based on the error, the bounded weight estimation is employed using the dice similarity to re-train the DQNN to perform incremental learning. In this, the weights of the DQNN are adjusted using the proposed ExpCoot to make the learning more effective with reduced information loss, which helps enhance the accuracy of sentiment analysis. The performance of the proposed ExpCoot_DQNN is analysed using F_Measure, recall, and precision and the proposed ExpCoot_DQNN accomplished the F_Measure, Recall, and Precision values of 0.97, 0.99, and 0.98, respectively.

Keywords: hybrid optimisation; deep learning; sentimental analysis; incremental learning; review document.

DOI: 10.1504/IJAHUC.2025.149634

International Journal of Ad Hoc and Ubiquitous Computing, 2025 Vol.50 No.3, pp.170 - 183

Received: 19 Jul 2024
Accepted: 16 Mar 2025
Published online: 07 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article