AAI_2025_Capstone_Chronicles_Combined

‭ResolveAI‬

‭This word cloud for medium-priority tickets (Fig. 5) highlights repeated references to‬ ‭“analytics tools”, “digital strategies”, “brand growth”, and “remains unresolved” suggesting a‬ ‭balance between marketing/data initiatives and unresolved issues. Additionally, the phrases‬ ‭“security measures” and “unauthorized access” show ongoing attention to system safety,‬ ‭although they appear alongside more business-oriented topics.‬

‭This word cloud for low-priority tickets (Fig. 6) prominently features “tools”, “details”,‬ ‭and “brand growth” suggesting a focus on general inquiries, routine updates, and broader‬ ‭business strategies rather than urgent troubleshooting. Mentions of “guidance”, “address”, and‬ ‭“information” reflect a need for clarification or support, though not at an emergency level.‬ ‭The variables in this dataset are closely tied to the project’s goal of automating ticket‬ ‭prioritization. Text fields (subject, body, answer) provide rich contextual information, while‬ ‭categorical variables (type, queue, priority, language) offer clear, discrete signals related to ticket‬ ‭urgency and handling.‬ ‭Exploratory analysis reveals that ticket type and queue exhibit strong associations with‬ ‭priority making them valuable predictors. Moreover, patterns in text length and recurring‬ ‭keywords across priorities indicate that the text data is also informative. Although there are‬ ‭correlations among some categorical features (such as type and priority), these relationships can‬

‭10‬

58