Intelligent transportation systems in urban areas in developing countries are still being developed. In this process, there are plenteous reports from society that complain about traffic jams, accidents, and other conditions that require immediate handling. For instance, in Surabaya, one of the urban areas in Indonesia, there were 7,975 complaints via @e100ss Twitter's account, the transportation media in Surabaya, in December 2022. This research aims to find contexts from the traffic complaint text to determine the urgency factor from the text. Traffic urgency factors are used to determine the complaint's handling priority. Therefore, the main steps in this research are as follows. The first step is preprocessing and tokenizing the text data. The second step is calculating the term frequency (TF) from the tokenized word. Furthermore, the words with the highest TF values were categorized according to their context. In this context, a model to determine urgency was developed in this research. This research suggests that traffic urgency is moderated by the context of location, context of a group people that being reported, context of the object that mentioned in the complaint text, and context of the condition. This research also found that a specific location has the most frequent reported time.