Rao Divya, Singh Rohit, Prakashini K, Vijayananda J
Manipal, 576104 India Department of Information and Communication Technology, Manipal Institute of Technology, Manipal Academy of Higher Education.
Manipal, 576104 India Department of Otorhinolaryngology, Kasturba Medical College, Manipal Academy of Higher Education.
Indian J Otolaryngol Head Neck Surg. 2023 Apr 26;75(3):1-7. doi: 10.1007/s12070-023-03813-2.
This study aims to investigate public sentiment on laryngeal cancer via tweets in 2022 using machine learning. We aimed to analyze the public sentiment about laryngeal cancer on Twitter last year. A novel dataset was created for the purpose of this study by scraping all tweets from 1st Jan 2022 that included the hashtags #throatcancer, #laryngealcancer, #supraglotticcancer, #glotticcancer, and #subglotticcancer in their text. After all tweets underwent a fourfold data cleaning process, they were analyzed using natural language processing and sentiment analysis techniques to classify tweets into positive, negative, or neutral categories and to identify common themes and topics related to laryngeal cancer. The study analyzed a corpus of 733 tweets related to laryngeal cancer. The sentiment analysis revealed that 53% of the tweets were neutral, 34% were positive, and 13% were negative. The most common themes identified in the tweets were treatment and therapy, risk factors, symptoms and diagnosis, prevention and awareness, and emotional impact. This study highlights the potential of social media platforms like Twitter as a valuable source of real-time, patient-generated data that can inform healthcare research and practice. Our findings suggest that while Twitter is a popular platform, the limited number of tweets related to laryngeal cancer indicates that a better strategy could be developed for online communication among netizens regarding the awareness of laryngeal cancer.
本研究旨在通过机器学习调查2022年推特上关于喉癌的公众情绪。我们旨在分析去年推特上关于喉癌的公众情绪。为了本研究的目的,通过抓取2022年1月1日起所有文本中包含#喉癌、#咽喉癌、#声门上癌、#声门癌和#声门下癌等标签的推文,创建了一个新的数据集。在所有推文都经过了四倍的数据清理过程后,使用自然语言处理和情感分析技术对其进行分析,以便将推文分类为积极、消极或中性类别,并识别与喉癌相关的常见主题和话题。该研究分析了733条与喉癌相关的推文语料库。情感分析显示,53%的推文为中性,34%为积极,13%为消极。推文中确定的最常见主题是治疗与疗法、风险因素、症状与诊断、预防与认知以及情感影响。本研究凸显了推特等社交媒体平台作为实时、患者生成数据的宝贵来源的潜力,这些数据可为医疗保健研究和实践提供信息。我们的研究结果表明,虽然推特是一个受欢迎的平台,但与喉癌相关的推文数量有限,这表明可以制定更好的策略,用于网民之间关于喉癌认知的在线交流。