Automatic content extraction and visualization of Thai websites for improved information representation

Abstract

Abstract-This paper presents an integrated approach to automatically provide an overview of content on Thai websites based on tag cloud. This approach is intended to address the information overload issue by presenting the overview to users in order that they could assess whether the information meets their needs. The approach has incorporated Web content extraction, Thai word segmentation, and information presentation to generate a tag cloud in Thai language as an overview of the key content in the webpage. From the experimental study, the generated Thai Tag clouds are able to provide an overview of the tags which frequently appear in the title and body of the content. Moreover, the first few lines in the tag cloud offer an improved readability.

Extracting the Main Content of Web Pages Using the First Impression Area

“…Although English and non-English news pages have similar subjects and layouts, the reader modes were not activated. There are several studies tried to resolve this problem by creating a new algorithm or a model for the local language [7]- [9]. However, it is difficult to create new methods for all languages owing to the shortage of developers fluent in each language, especially low-resource languages.…”