TAILIEUCHUNG - Báo cáo khoa học: "A Rose is a Roos is a Ruusu: Querying Translations for Web Image Search"

We query Web Image search engines with words (., spring) but need images that correspond to particular senses of the word (., flexible coil). Querying with polysemous words often yields unsatisfactory results from engines such as Google Images. We build an image search engine, I DIOM, which improves the quality of returned images by focusing search on the desired sense. Our algorithm, instead of searching for the original query, searches for multiple, automatically chosen translations of the sense in several languages. . | A Rose is a Roos is a Ruusu Querying Translations for Web Image Search Janara Christensen Mausam Oren Etzioni Turing Center Dept. of Computer Science and Engineering University of Washington Seattle WA 98105 USA janara mausam etzioni @ Abstract We query Web Image search engines with words . spring but need images that correspond to particular senses of the word . flexible coil . Querying with polysemous words often yields unsatisfactory results from engines such as Google Images. We build an image search engine Idiom which improves the quality of returned images by focusing search on the desired sense. Our algorithm instead of searching for the original query searches for multiple automatically chosen translations of the sense in several languages. Experimental results show that IDIOM outperforms Google Images and other competing algorithms returning 22 more relevant images. 1 Introduction One out of five Web searches is an image search Basu 2009 . A large subset of these searches is subjective in nature where the user is looking for different images for a single concept Linsley 2009 . However it is a common user experience that the images returned are not relevant to the intended concept. Typical reasons include 1 existence of homographs other words that share the same spelling possibly in another language and 2 polysemy several meanings of the query word which get merged in the results. For example the English word spring has several senses - 1 the season 2 the water body 3 spring coil and 4 to jump. Ten out of the first fifteen Google images for spring relate to the season sense three to water body one to coil and none to the jumping sense. Simple modifications to query do not always work. Searching for spring water results in many images of bottles of spring water and searching for spring jump returns only three images out of fifteen of someone jumping. Polysemous words are common in English. It is estimated that average polysemy of .

• Báo cáo khoa học
• A Rose is a Roos is a Ruusu
• Querying Translations
• Web Image Search
• báo cáo khoa học
• báo cáo ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Bài giảng Xử lý ngôn ngữ tự nhiên
• Ngôn ngữ tự nhiên
• Kỹ thuật lập trình
• Xử lý nhập nhằng ngữ nghĩa
• Xử lý ngữ nghĩa
• Xử lý ngôn ngữ
• Natural Language Processing
• Ngôn ngữ lập trình
• Nhập nhằng cấu trúc
• Tri thức về ngôn ngữ
• Mô hình ngôn ngữ
• Mô hình n gram
• Dịch máy
• Phương pháp dịch máy
• Hiểu ngôn ngữ
• Phân tích ngữ nghĩa
• Phân loại tin tự động
• Biểu diễn vị từ
• Thuộc tính về sự kiện
• Thuộc tính về sự kiện
• Mô hình ngôn ngữ Google Book N grams
• Mô hình ngôn ngữ KenLM
• Phương pháp làm mịn
• Tách từ tiếng Việt
• Gán nhãn từ loại
• Phân tích cú pháp
• Nghĩa từ vựng
• Phân giải nhập nhằng từ
• Phân loại văn bản
• Lexical chain
• Phản hồi thông tin
• Bài toán PTCP
• Cấu trúc ngữ pháp
• Phân tích cú pháp xác suất
• CKY kết hợp xác suất
• Văn phạm phi ngữ cảnh xác suất
• Hình thái học