Sosyal medyanın iletişimdeki gücü, Covid-19’un beraberinde getirdiği yeni dünya ile daha önemli hale gelmiştir. Bu araştırmanın temel amacı, pandemi döneminde Türkiye Cumhuriyeti Sağlık Bakanı tarafından paylaşılan tweetleri çalışan motivasyonu bağlamında incelemektir. Bu kapsamda Python kütüphaneleri kullanılarak Gizli Dirichlet Ayırımı (LDA) temelli olasılıksal bir model oluşturulması alt bir amaç olarak belirlenmiştir. Analizler 3371 tweet ile başlamış ve konu bağlamında 482 tweet ile sürdürülmüştür. Ulaşılan sonuçlar, sağlık çalışanlarıyla ilgili olarak paylaşılan tweetlerin ilgili konunun oranına göre sırasıyla; “bilgilendirme”, “aşı ile mücadele ve örneklik”, “şiddet”, “fedakârlık ve güven” ve “haklar ve özel günler” şeklinde beş konuda kümelendiğini göstermektedir. Paylaşım anlamında son iki sırada olan “fedakârlık ve güven” ve “haklar ve özel günler” konuları, en çok beğeni ve retweet edilme ortalamasına sahip ikinci ve üçüncü konu olarak görülmektedir. Bu kapsamda açıklanan sonuçlar, modelin yerel konuşma dilinde yayımlanmış ilgili metinlerin yayımcısının görüşlerini özetleyen kelimelere indirgenebileceğini ve izleyiciler üzerinde oluşturduğu tepkinin, duygusal olarak görülmesini mümkün kıldığını; bununla birlikte paylaşılan mesajların motivasyonel anlamda okuyuculara verdiği mesajı tam olarak yansıtamadığını göstermiştir.
The power of social media in communication has become more important with the new world brought about by Covid-19. The main purpose of this research is to examine the tweets shared by the Minister of Health of the Republic of Türkiye during the pandemic period in the context of employee motivation. For this, creating a probabilistic model based on Latent Dirichlet Allocation (LDA) using Python libraries has been determined as a sub-goal. The analyzes started with 3371 tweets, and continued with 482 tweets in the context of the subject. The results show that the tweets shared about healthcare professionals are clustered in five topics. According to the ratio of the relevant topics, these are; “information”, “struggle against vaccination and exemplary”, “violence”, “self sacrifice and trust” and “rights and special days”. “Self sacrifice and trust" and "rights and special days" are in the last two places in sharing rates. Nevertheless, these two topics are the second and third topics with the most likes and retweets on average. According to the results, the model can be reduced to words that summarize the views of the possessor of the relevant texts published in the local spoken language and make it possible to see the reaction it creates on the audience/reader emotionally. However, the model cannot fully reflect the motivational message that the shared messages give to the readers.
Covid-19 Employee Motivation Natural Language Processing Latent Dirichlet Allocation Topic Modelling
Primary Language | Turkish |
---|---|
Subjects | Behaviour-Personality Assessment in Psychology |
Journal Section | Research Articles |
Authors | |
Publication Date | July 30, 2022 |
Submission Date | July 12, 2022 |
Published in Issue | Year 2022 Volume: 4 Issue: 2 |