Combining Machine Learning and Natural Language Processing for Language-Specific, Multi-Lingual, and Cross-Lingual Text Summarization

Abstract

The recent advances in multimedia and web-based applications have eased the accessibility to large collections of textual documents. To automate the process of document analysis, the research community has put relevant efforts into extracting short summaries of the document content. However, most of the early proposed summarization methods were tailored to English-written textual corpora or to collections of documents all written in the same language. More recently, the joint efforts of the machine learning and the natural language processing communities have produced more portable and flexible solutions, which can be applied to documents written in different languages. This chapter first overviews the most relevant language-specific summarization algorithms. Then, it presents the most recent advances in multi- and cross-lingual text summarization. The chapter classifies the presented methodology, highlights the main pros and cons, and discusses the perspectives of the extension of the current research towards cross-lingual summarization systems.