Abstract | ||
---|---|---|
This paper discusses the challenges of applying named entity linking in a rich, complex domain - specifically, the linking of (1) military units, (2) places and (3) people in the context of interlinked Second World War data. Multiple sub-scenarios are discussed in detail through concrete evaluations, analyzing the problems faced, and the solutions developed. A key contribution of this work is to highlight the heterogeneity of problems and approaches needed even inside a single domain, depending on both the source data as well as the target authority. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1007/978-3-319-59888-8_10 | Lecture Notes in Artificial Intelligence |
Field | DocType | Volume |
World Wide Web,Source data,Optical character recognition,Named entity,World War II,Artificial intelligence,Natural language processing,Geography | Conference | 10318 |
ISSN | Citations | PageRank |
0302-9743 | 3 | 0.48 |
References | Authors | |
12 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Erkki Heino | 1 | 7 | 1.61 |
Minna Tamper | 2 | 5 | 1.55 |
Eetu Mäkelä | 3 | 188 | 23.10 |
Petri Leskinen | 4 | 7 | 1.28 |
Esko Ikkala | 5 | 12 | 9.62 |
Jouni Tuominen | 6 | 178 | 27.45 |
Mikko Koho | 7 | 14 | 9.77 |
Eero Hyvönen | 8 | 843 | 103.43 |