文本情感分析技术-北京大学计算机科学技术研究所.PDFVIP

文本挖掘技术(2013春) 第十二章： 文本情感分析技术 杨建武 北京大学计算机科学技术研究所 Email:yangjw@ 1 情感计算的概念  情感计算(Affective Computing) 通过计算机技术，自动分析文本、图像或视 音频等对象所包含的情感倾向及其强度 • 例如：正面或负面、喜欢或讨厌、快乐或悲伤、 愤怒和恐惧等 情感计算的分类 • 主观性（Subjectivity ） – 主观性、客观性和中性 • 情感倾向（Orientation ） – 正面（褒义）、负面（贬义）和中性 2 情感计算的应用  81% of Internet users (or 60% of Americans) have done online research on a product at least once;  Among readers of online reviews of restaurants, hotels, and various services (e.g., travel agencies or doctors), between 73% and 87% report that reviews had a significant influence on their purchase;  Consumers report being willing to pay from 20% to 99% more for a 5-star-rated item than a 4-star- rated item (the variance stems from what type of item or service is considered); 3 情感计算的应用  Businesses and organizations: product and service benchmarking. Market intelligence.  Business spends a huge amount of money to find consumer sentiments and opinions. • Consultants, surveys and focused groups, etc  Individuals: interested in other’s opinions when  Purchasing a product or using a service,  Finding opinions on political topics,  Ads placements: Placing ads in the user-generated content  Place an ad when one praises a product.  Place an ad from a competitor if one criticizes a product.  Opinion retrieval/search: providing general search for opinions. 4