Ioannou, E.; Nejdl, W.; Niederée, C.; Velegrakis, Y.: Embracing uncertainty in entity linking (2012)
0.02
0.015600072 = product of:
0.046800215 = sum of:
0.046800215 = product of:
0.09360043 = sum of:
0.09360043 = weight(_text_:publishing in 433) [ClassicSimilarity], result of:
0.09360043 = score(doc=433,freq=4.0), product of:
0.24522576 = queryWeight, product of:
4.885643 = idf(docFreq=907, maxDocs=44218)
0.05019314 = queryNorm
0.38169086 = fieldWeight in 433, product of:
2.0 = tf(freq=4.0), with freq of:
4.0 = termFreq=4.0
4.885643 = idf(docFreq=907, maxDocs=44218)
0.0390625 = fieldNorm(doc=433)
0.5 = coord(1/2)
0.33333334 = coord(1/3)
- Abstract
- The modern Web has grown from a publishing place of well-structured data and HTML pages for companies and experienced users into a vivid publishing and data exchange community in which everyone can participate, both as a data consumer and as a data producer. Unavoidably, the data available on the Web became highly heterogeneous, ranging from highly structured and semistructured to highly unstructured user-generated content, reflecting different perspectives and structuring principles. The full potential of such data can only be realized by combining information from multiple sources. For instance, the knowledge that is typically embedded in monolithic applications can be outsourced and, thus, used also in other applications. Numerous systems nowadays are already actively utilizing existing content from various sources such as WordNet or Wikipedia. Some well-known examples of such systems include DBpedia, Freebase, Spock, and DBLife. A major challenge during combining and querying information from multiple heterogeneous sources is entity linkage, i.e., the ability to detect whether two pieces of information correspond to the same real-world object. This chapter introduces a novel approach for addressing the entity linkage problem for heterogeneous, uncertain, and volatile data.
