<![CDATA[Understanding customer sentiment in real time has become increasingly critical for service-oriented industries, particularly airlines operating within highly competitive and socially sensitive environments. This study proposes an integrated Business Intelligence (BI) framework for sentiment analysis of United Airlines using social media data sourced from Twitter. The framework aims to transform large volumes of unstructured, high-velocity text data into actionable insights that support informed decision-making, customer experience enhancement, and brand reputation management. The proposed architecture incorporates sequential analytical components including data ingestion, preprocessing, natural language processing, machine learning–based sentiment classification, and BI-driven visualization. Modern text analytics techniques such as tokenization, lemmatization, and vectorization are applied to prepare textual content for polarity detection, while supervised learning algorithms are evaluated to classify sentiment into positive, negative, and neutral categories. The study outlines the rationale for adopting a scalable, cloud-compatible architecture that supports both batch and stream processing to accommodate the dynamic nature of social media data. Key implementation challenges—such as handling noisy and ambiguous text, managing evolving linguistic patterns, overcoming API rate limitations, and ensuring data quality—are examined. The paper further discusses best practices to mitigate these challenges, including robust data-cleaning pipelines, periodic model retraining, careful feature engineering, and the incorporation of governance principles for ethical data use. The results demonstrate that integrating sentiment analytics within a BI context enables organizations such as United Airlines to monitor customer perceptions more effectively and respond proactively to emerging issues. The framework provides a practical foundation for organizations seeking to operationalize social media analytics for strategic and operational decision support.]]>
