TAILIEUCHUNG - Báo cáo khoa học: "Text Segmentation with LDA-Based Fisher Kernel"

In this paper we propose a domainindependent text segmentation method, which consists of three components. Latent Dirichlet allocation (LDA) is employed to compute words semantic distribution, and we measure semantic similarity by the Fisher kernel. Finally global best segmentation is achieved by dynamic programming. Experiments on Chinese data sets with the technique show it can be effective. Introducing latent semantic information, our algorithm is robust on irregular-sized segments. | Text Segmentation with LDA-Based Fisher Kernel Qi Sun Runxin Li Dingsheng Luo and Xihong Wu Speech and Hearing Research Center and Key Laboratory of Machine Perception Ministry of Education Peking University 100871 Beijing China sunq lirx dsluo wxh @ Abstract In this paper we propose a domainindependent text segmentation method which consists of three components. Latent Dirichlet allocation LDA is employed to compute words semantic distribution and we measure semantic similarity by the Fisher kernel. Finally global best segmentation is achieved by dynamic programming. Experiments on Chinese data sets with the technique show it can be effective. Introducing latent semantic information our algorithm is robust on irregular-sized segments. 1 Introduction The aim of text segmentation is to partition a document into a set of segments each of which is coherent about a specific topic. This task is inspired by problems in information retrieval summarization and language modeling in which the ability to provide access to smaller coherent segments in a document is desired. A lot of research has been done on text segmentation. Some of them utilize linguistic criteria Beeferman et al. 1999 Mochizuki et al. 1998 while others use statistical similarity measures to uncover lexical cohesion. Lexical cohesion methods believe a coherent topic segment contains parts with similar vocabularies. For example the TextTiling algorithm introduced by Hearst 1994 assumes that the local minima of the word similarity curve are the points of low lexical cohesion and thus the natural boundary candidates. Reynar 1998 has proposed a method called dotplotting depending on the distribution of word repetitions to find tight regions of topic similarity graphically. One of the problems with those works is that they treat terms uncorrelated assigning them orthogonal directions in the feature space. But in reality words are correlated and sometimes even synonymous so that texts with very few

• Báo cáo khoa học
• Text Segmentation
• LDA Based Fisher Kernel
• Qi Sun
• Runxin Li
• Dingsheng Luo and Xihong Wu
• báo cáo khoa học
• báo cáo ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Text Segmentation by Language Using Minimum Description Length
• Hiroshi Yamaguchi
• Kumiko
• Tanaka Ishii
• ngôn ngữ tự nhiên
• Unsupervised Segmentation
• Chinese Text
• Use of Branching Entropy
• mô hình ngôn ngữ
• A Statistical Model
• Domain Independent
• Reiteration and Collocation
• Amanda C
• Jobbins
• Multiple Surface Linguistic Cues
• Long Papers
• Optimal Multi Paragraph Text Segmentation
• Dynamic Programming
• Oskari Heinonen
• Cohesion and Collocation
• Using Context Vectors
• BASED TEXT SEGMENTATION
• SIMILARITY BETWEEN WORDS
• Hideki Kozima
• MULTI PARAGRAPH SEGMENTATION
• EXPOSITORY TEXT
• Marti A
• Hearst
• Linear Text Segmentation
• Dynamic Programming Algorithm
• Fragkou Pavlina
• scientific reports
• model language
• process natural language
• giáo dục
• đào tạo
• cao đẳng
• đại học
• Linuxs solution
• LXR can help
• Objdump can disassemble
• Use a text editor
• Role of segmentation
• A race cure case study
• International Journal of Computer Science & Communication Networks
• Cluster based hierarchical approach
• The recognition of bengali handwritten character
• Bengali handwritten character
• Bengali text segmentation