UNED at WebCLEF 2008: Applying High Restrictive Summarization, Low Restrictive Information Retrieval and Multilingual Techniques Enrique Amig´o, Juan Martinez-Romo, Lourdes Araujo, and V´ ıctor Peinado NLP & IR Group at UNED, ETSI Inform´atica UNED c/ Juan del Rosal, 16. E-28040 Madrid, Spain {enrique, juaner, lurdes, victor}@lsi.uned.es Abstract. This paper describes our participation in the WebCLEF 2008 task, targeted at snippet retrieval from new data. Our system assumes that the task can be tackled as a summarization problem and that the document retrieval and multilinguism treatment steps can be ignored. Our approach assumes also that the redundancy of information in the Web allows the system to be very restrictive when picking information pieces. Our evaluation results suggest that, while the first assumption is feasible, the second one is not always true. 1 Introduction The WebCLEF 2008 task has been defined in a similar way to the previous edition. Systems are asked to return a ranked list of snippets extracted from the 1000 web documents identified using the Google web search engine. Mul- tiple languages are covered by the queries and retrieved documents. This task inherits several aspect from Information Retrieval, Summarization and Question Answering tasks. Our approach, as we will describe, is oriented to summarization strategies. 2 Assumptions Participants are provided with a topic title, a description of the information need, the languages in which the information must be returned, a set of known sources, and a set of queries and their relevant web pages retrieved using Google. The snippets returned by the system must cover the information need without introducing any redundant information already included in the known sources or in other retrieved snippets. Our approach makes the some assumptions that will be tested in the following sections, namely: 1. The terms included in the queries are unambiguous. For instance, “machine translation” (topic 41) refers to systems that translate text from one lan- guage into another.