TAILIEUCHUNG - Báo cáo khoa học: "The order of prenominal adjectives in natural language generation"

The order of prenominal adjectival modifiers in English is governed by complex and difficult to describe constraints which straddle the boundary between competence and performance. This paper describes and compares a number of statistical and machine learning techniques for ordering sequences of adjectives in the context of a natural language generation system. the large red American car ??the American red large car *car American red the large Some orders are more marked than others, but none are strictly speaking ungrammatical. . | The order of prenominal adjectives in natural language generation Robert Malouf Alfa Informatica Rijksuniversiteit Groningen Postbus 716 9700 AS Groningen The Netherlands malouf@ Abstract The order of prenominal adjectival modifiers in English is governed by complex and difficult to describe constraints which straddle the boundary between competence and performance. This paper describes and compares a number of statistical and machine learning techniques for ordering sequences of adjectives in the context of a natural language generation system. 1 The problem The question of robustness is a perennial problem for parsing systems. In order to be useful a parser must be able to accept a wide range of input types and must be able to gracefully deal with dysfluencies false starts and other ungrammatical input. In natural language generation on the other hand robustness is not an issue in the same way. While a tactical generator must be able to deal with a wide range of semantic inputs it only needs to produce grammatical strings and the grammar writer can select in advance which construction types will be considered grammatical. However it is important that a generator not produce strings which are strictly speaking grammatical but for some reason unusual. This is a particular problem for dialog systems which use the same grammar for both parsing and generation. The looseness required for robust parsing is in direct opposition to the tightness needed for high quality generation. One area where this tension shows itself clearly is in the order of prenominal modifiers in English. In principle prenominal adjectives can depending on context occur in almost any order the large red American car the American red large car car American red the large Some orders are more marked than others but none are strictly speaking ungrammatical. So the grammar should not put any strong constraints on adjective order. For a generation system however it is important that sequences

• Báo cáo khoa học
• The order of prenominal
• adjectives in natural language generation
• Robert Malouf Alfa
• báo cáo khoa học
• mô hình 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 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