Predicting STC Customers’ Satisfaction Using Twitter

Abstract

The telecom field has changed accordingly with the emergence of new technologies. This is the case with the telecom market in Saudi Arabia, which expanded in 2003 by attracting new investors. As a result, the Saudi telecom market became a viable market [1]. The prevalence of mobile voice service among the population in Saudi Arabia for that, this research aims at mining Arabic tweets to measure customer satisfaction toward Telecom company in Saudi Arabia. This research is a use case for the Saudi Telecom Company (STC) in Saudi Arabia. The contribution of this study will be capitalised as recommendations to the company, based on monitoring in real-time their customers' satisfaction on Twitter and from questionnaire analysis. It is the first work to evaluate customers' satisfaction with telecommunications (telecom) company in Saudi Arabia by using both social media mining and a quantitative method. It has been built a corpus of Arabic tweets, using a Python script searching for real-time tweets that mention Telecom company using the hashtags to monitor the latest sentiments of Telecom customers continuously. The subset is 20,000 tweets that are randomly selected from the dataset, for training the machine-classifier. In addition, we have done the experimented using deep learning network. The results show that the satisfaction for each service ranges between 31.50% and 49.25%. One of the proposed recommendations is using 5G to solve the "Internet Speed" problem, which showed the lowest customer satisfaction, with 31.50%. This paper's main contributions are defining the traceable measurable criteria for customer satisfaction with telecom companies in Saudi Arabia and providing telecom companies' recommendations based on monitoring real-time customers' satisfaction through Twitter.

The impact of key indicators on the overall performance of Saudi Arabian telecommunication companies

“…As a result, Saudi Arabian telecommunication companies are suffering from significant financial losses year after year since their establishment (Kadasah, 2014). Meanwhile, there is a decrease in the overall performance of Saudi Arabia's telecommunication companies (Almuqren & Cristea, 2022;Anaam et al, 2021). However, the use of key performance indicators may contribute to ensuring that organizations are moving in the right direction, assessing the achievement of desired strategies and objectives, and evaluating and controlling the overall business processes (Ishaq Bhatti et al, 2014;Moktadir et al, 2020;Parmenter, 2015).…”

BERT Models for Arabic Text Classification: A Systematic Review

“…28 articles [49], [50], [51], [9], [52], [28], [53], [54], [55], [30], [31], [33], [56], [34], [35], [36], [57], [37], [11], [58], [39], [40], [59], [43], [60], [46], [61], [48] MARBERT 0 9 9 articles [50], [62], [27], [53], [34], [43], [46], [63], [64] ArabicBERT 5 articles [34], [58], [38], [40], […”

“…The new models attempt to address some of the shortcomings of the original models, improving the performance [7] or improving the inference speed [8]. There were also models that were developed to support languages other than English [9,10].…”

Toward a Cognitive-Inspired Hashtag Recommendation for Twitter Data Analysis

“…This method used semantic-based quantitative analysis to uncover important spatiotemporal aspects of COVID-19 stigma to provide timely warnings and risk mitigation. Almuqren and Cristea [26] took aim at the Saudi Telecommunication Company, Riyadh, Saudi Arabia, and monitored their customers' satisfaction via Twitter and questionnaire analysis in real time. They used social media mining and a quantitative method to assess customer satisfaction with a telecommunications provider.…”