Searching XML documents via XML fragments

摘要: Most of the work on XML query and search has stemmed from publishing database communities, mostly for needs business applications. Recently, Information Retrieval community began investigating issue to answer information discovery needs. Following this trend, we present here an approach where can be expressed in approximate manner as pieces documents or "XML fragments" same nature that are being searched. We extension vector space model searching collections via fragments ranking results by relevance. describe how have extended a full-text engine comply with model. The value proposed method is demonstrated relative high precision our system, which was among top performers recent INEX workshop. Our indicate certain queries more appropriate than others Specifically, relatively specific contexts but vague best situated reap benefit Finally show one may not fit all types it could worthwhile use different solutions