Opinion mining and sentiment analysis involve the computational processing of opinions, sentiment, and subjectivity in text. This survey discusses techniques and approaches that enable opinion-oriented information-seeking systems. The focus is on methods addressing new challenges in sentiment-aware applications compared to traditional fact-based analysis. Topics include summarization of evaluative text, privacy, manipulation, and economic impact of opinion-oriented information-access services. Available resources, benchmark datasets, and evaluation campaigns are also discussed to facilitate future work. The increasing availability of opinion-rich resources like online reviews and blogs has led to a surge in interest in systems that directly handle opinions. This is driven by the need to understand and act on opinions in various domains, including consumer behavior, political information, and business intelligence. The survey highlights the importance of sentiment analysis in areas such as recommendation systems, question answering, and information extraction. It also discusses the challenges of sentiment analysis, including the subtlety of sentiment expression, the difficulty of identifying opinions, and the impact of context and domain on sentiment interpretation. The survey covers the history and terminology of sentiment analysis and opinion mining, noting that the field has evolved from early projects on beliefs to current applications in social media monitoring and analysis. It discusses the challenges of sentiment classification, including the difficulty of identifying sentiment polarity and the impact of order and context on sentiment interpretation. The survey also addresses the broader implications of opinion-oriented information-access services, including privacy, manipulation, and economic impact. The survey provides an overview of classification and extraction techniques in sentiment analysis, including the use of machine learning for sentiment classification and the challenges of extracting opinions from text. It discusses the importance of modeling discourse structure and sequential information in sentiment analysis. The survey concludes with a discussion of the future directions of opinion mining and sentiment analysis, emphasizing the need for further research and development in this area.Opinion mining and sentiment analysis involve the computational processing of opinions, sentiment, and subjectivity in text. This survey discusses techniques and approaches that enable opinion-oriented information-seeking systems. The focus is on methods addressing new challenges in sentiment-aware applications compared to traditional fact-based analysis. Topics include summarization of evaluative text, privacy, manipulation, and economic impact of opinion-oriented information-access services. Available resources, benchmark datasets, and evaluation campaigns are also discussed to facilitate future work. The increasing availability of opinion-rich resources like online reviews and blogs has led to a surge in interest in systems that directly handle opinions. This is driven by the need to understand and act on opinions in various domains, including consumer behavior, political information, and business intelligence. The survey highlights the importance of sentiment analysis in areas such as recommendation systems, question answering, and information extraction. It also discusses the challenges of sentiment analysis, including the subtlety of sentiment expression, the difficulty of identifying opinions, and the impact of context and domain on sentiment interpretation. The survey covers the history and terminology of sentiment analysis and opinion mining, noting that the field has evolved from early projects on beliefs to current applications in social media monitoring and analysis. It discusses the challenges of sentiment classification, including the difficulty of identifying sentiment polarity and the impact of order and context on sentiment interpretation. The survey also addresses the broader implications of opinion-oriented information-access services, including privacy, manipulation, and economic impact. The survey provides an overview of classification and extraction techniques in sentiment analysis, including the use of machine learning for sentiment classification and the challenges of extracting opinions from text. It discusses the importance of modeling discourse structure and sequential information in sentiment analysis. The survey concludes with a discussion of the future directions of opinion mining and sentiment analysis, emphasizing the need for further research and development in this area.