Harth, A.; Hogan, A.; Umbrich, J.; Kinsella, S.; Polleres, A.; Decker, S.: Searching and browsing linked data with SWSE* (2012)
0.01
0.01321952 = product of:
0.0660976 = sum of:
0.0660976 = weight(_text_:books in 410) [ClassicSimilarity], result of:
0.0660976 = score(doc=410,freq=2.0), product of:
0.24756333 = queryWeight, product of:
4.8330836 = idf(docFreq=956, maxDocs=44218)
0.051222645 = queryNorm
0.2669927 = fieldWeight in 410, product of:
1.4142135 = tf(freq=2.0), with freq of:
2.0 = termFreq=2.0
4.8330836 = idf(docFreq=956, maxDocs=44218)
0.0390625 = fieldNorm(doc=410)
0.2 = coord(1/5)
- Abstract
- Web search engines such as Google, Yahoo! MSN/Bing, and Ask are far from the consummate Web search solution: they do not typically produce direct answers to queries but instead typically recommend a selection of related documents from the Web. We note that in more recent years, search engines have begun to provide direct answers to prose queries matching certain common templates-for example, "population of china" or "12 euro in dollars"-but again, such functionality is limited to a small subset of popular user queries. Furthermore, search engines now provide individual and focused search interfaces over images, videos, locations, news articles, books, research papers, blogs, and real-time social media-although these tools are inarguably powerful, they are limited to their respective domains. In the general case, search engines are not suitable for complex information gathering tasks requiring aggregation from multiple indexed documents: for such tasks, users must manually aggregate tidbits of pertinent information from various pages. In effect, such limitations are predicated on the lack of machine-interpretable structure in HTML-documents, which is often limited to generic markup tags mainly concerned with document renderign and linking. Most of the real content is contained in prose text which is inherently difficult for machines to interpret.
