Home // AICT 2024, The Twentieth Advanced International Conference on Telecommunications // View article

Implementation of LSTM Neural Networks for Predicting Competition in Telecommunications Markets

Authors:
Diego Armando Giral-Ramirez
Luis Fernando Pedraza-Martínez
Cesar Augusto Hernández-Suarez

Keywords: deep learning; LSTM; market operations; neural networks; competition prediction; styling.

Abstract:
The telecommunications services market faces significant challenges in an increasingly flexible and customer-adaptable environment. Research has highlighted that the monopolization of spectrum by an operator reduces competition and negatively impacts users and the overall dynamics of the sector. This article addresses the importance of competition analysis and its prediction in telecommunications markets. A Long-Short Term Memory (LSTM) network is implemented to forecast the number of users, the amount of revenue, and the amount of traffic for fifteen network operators. The ability of LSTMs to handle temporal sequences, long-term dependencies, adaptability to changes, and management of complex data makes them an excellent strategy for predicting and forecasting the telecommunications market. As identified in the literature review, diverse works involve LSTM and telecommunications. However, many questions remain in the area of prediction. Various strategies can be proposed, and permanent work must focus on providing cognitive engines to address more challenges. MATLAB is used for the design and subsequent implementation, with a root mean square error index of 0.0776; the results demonstrate the accuracy of the implemented strategy.

Pages: 17 to 22

Copyright: Copyright (c) IARIA, 2024

Publication date: April 14, 2024

Published in: conference

ISSN: 2308-4030

ISBN: 978-1-68558-145-9

Location: Venice, Italy

Dates: from April 14, 2024 to April 18, 2024