Analyzing the Role of Twitter as Electronic Word of Mouth for Health-care Needs in Saudi Arabia

Analyzing the Role of Twitter as Electronic Word of Mouth for Health-care Needs in Saudi Arabia by Manahil Abdulmoin Alqulaity (2018)

Twitter is one of the most popular microblogging platforms that has a large number of users. It provides easy access for different types of users to publish posts called tweets. Nowadays, Twitter has been increasingly used for information sharing and opinions exchanging. Thus, the trend of online communication has been improved as a result of its efficiency and usability which indeed empower the engagement of online users in order to communicate their experiences. Twitter is widely used in Saudi Arabia. In fact, Saudi Arabia ranks in the top countries for active Twitter internet users. Among many other domains, people share their experiences on Twitter about different healthcare services. The main focus of this research is to analyze the usage of Twitter as electronic Word of Mouth (eWoM) about healthcare services in Saudi Arabia. From this research point of view, eWOM for healthcare are rarely explored scientifically yet, specifically for Saudi Arabia. In this research, we investigated Twitter microblogging as a form of electronic word of mouth for sharing patients’ opinions about healthcare services in Saudi Arabia. People may have a variety of sentiments about the provided healthcare services and to understand the issues related to healthcare through social media, we need to understand their sentiment about healthcare service on social media. Therefore, one of the most important research problems is to understand the people’s sentiments for specific healthcare services but due to the large volume of the available online data, it would be time-consuming to manually check them. Further, we approached this problem by using supervised machine learning techniques, where a sample is manually labeled to train a model which can handle a large volume of data in less time. Thus, this research would help healthcare service providers to improve their services. During the research, we analyzed the overall sentiment trends of healthcare services in Saudi Arabia, the structure of these microblog postings, the types of expressions, the movement in positive, negative and neutral sentiment, and the patterns of engagement between healthcare providers and their patients by applying sentiment analysis models on a corpus of collected tweets posted on Twitter. The overall process has been performed by using methods and techniques from the fields of Data Science, Natural Language Processing, and Social Network Analysis. The collected data for the research was a streaming data collected from a set of manually verified healthcare accounts in Saudi Arabia where specified techniques used for data collection. Then, different sentiment analysis models have been tested to classify a set of labeled data with different functions to reach higher accuracy. Furthermore, patients’ sentiment polarity has been measured by using the best model resulted from the previous step. Finally, Then, we plan to study how different types of interactions are carried out by various healthcare providers.