This paper presents a method for mining and summarizing customer reviews of products sold online. The goal is to generate feature-based summaries that highlight the product features and the opinions (positive or negative) expressed by customers. The task is divided into three main steps: (1) mining product features that have been commented on by customers; (2) identifying opinion sentences in each review and determining whether each sentence is positive or negative; (3) summarizing the results. The proposed techniques use data mining and natural language processing methods to perform these tasks. The system, called FBS (Feature-Based Summarization), is implemented and tested on reviews of five products. The results show that the techniques are effective in generating summaries that highlight the key features and opinions of the product. The paper also discusses related work in the areas of subjective genre classification, sentiment classification, text summarization, and terminology finding. The authors conclude that their techniques are promising for summarizing customer reviews and believe they can be applied in practical settings. They also note three main limitations of their system: (1) the need for pronoun resolution; (2) the use of adjectives as the only indicators of opinion orientation; and (3) the need to study the strength of opinions. The authors plan to address these issues in future work.This paper presents a method for mining and summarizing customer reviews of products sold online. The goal is to generate feature-based summaries that highlight the product features and the opinions (positive or negative) expressed by customers. The task is divided into three main steps: (1) mining product features that have been commented on by customers; (2) identifying opinion sentences in each review and determining whether each sentence is positive or negative; (3) summarizing the results. The proposed techniques use data mining and natural language processing methods to perform these tasks. The system, called FBS (Feature-Based Summarization), is implemented and tested on reviews of five products. The results show that the techniques are effective in generating summaries that highlight the key features and opinions of the product. The paper also discusses related work in the areas of subjective genre classification, sentiment classification, text summarization, and terminology finding. The authors conclude that their techniques are promising for summarizing customer reviews and believe they can be applied in practical settings. They also note three main limitations of their system: (1) the need for pronoun resolution; (2) the use of adjectives as the only indicators of opinion orientation; and (3) the need to study the strength of opinions. The authors plan to address these issues in future work.