<p>Nowadays, social networks play a critical role in online social discourse, particularly during major events such as elections, health crises, and wars. Furthermore, individuals have spent significant time on social networks that consume news and express their opinions and viewpoints on various topics, namely, the Russia-Ukraine War. The Ukraine-Russia War is one of the most complicated and unfortunate events of this decade, with many aspects to be considered to have an informed opinion. For this reason, sentiment analysis research can be useful in analyzing sentiments or opinions about the Ukraine-Russia war, such as Twitter (X) posts. In this study, we propose a new deep learning-based sentiment classification model called "Knowledge Graph Convolutional Networks" that predicts and analyzes sentiments concerning the Russia-Ukraine war. There were 500,000 tweets collected in total, however only 410,428 were left for sentiment analysis after cleaning, lemmatization stemming, and deleting duplicate tweets. Results show that "war", "people", "world", "putin", "energy", "gas", "weapon", and "peace" were some of the most frequently occurring words in the tweets. The main aim of this work is to develop a robust system that provides an understanding of public sentiment towards the Russia-Ukraine war on the X platform. Experimental results demonstrate that it is possible to obtain more accurate sentiment classification results by the proposed method. They also have managerial, economic, and research implications.</p>