dc.creator | Koniaris M., Anagnostopoulos I., Vassiliou Y. | en |
dc.date.accessioned | 2023-01-31T08:43:55Z | |
dc.date.available | 2023-01-31T08:43:55Z | |
dc.date.issued | 2016 | |
dc.identifier | 10.1007/978-3-319-48740-3_12 | |
dc.identifier.isbn | 9783319487397 | |
dc.identifier.issn | 03029743 | |
dc.identifier.uri | http://hdl.handle.net/11615/75054 | |
dc.description.abstract | The number of freely available legal data sets is increasing at high speed. Citizens can easily access a lot of information about regulations,court orders,statutes,opinions and analytical documents. Such openness brings undeniable benefits in terms of transparency,participation and availability of new services. However,legal information overload poses new challenges,especially in the field of Legal Information Retrieval. Search result diversification has gained attention as a way to increase user satisfaction in web search. We hypothesize that such a strategy will also be beneficial for search on legal data sets. We address diversification of results in legal search by introducing legal domain specific diversification criteria and adopting several state of the art methods from the web search,network analysis and text summarization domains. We evaluate our diversification framework using a real data set from the Common Law domain that we subjectively annotated with relevance judgments for this purpose. Our findings reveal that web search diversification techniques outperform other approaches (e.g. summarizationbased,graph-based methods) in the context of legal diversification,as well as that the diversity criteria we introduce provide distinctively diverse subsets of resulting documents,thus differentiating our proposal in respect to traditional diversification techniques. © Springer International Publishing AG 2016. | en |
dc.language.iso | en | en |
dc.source | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | en |
dc.source.uri | https://www.scopus.com/inward/record.uri?eid=2-s2.0-84996559037&doi=10.1007%2f978-3-319-48740-3_12&partnerID=40&md5=187b08dfd750cab62383a0160884c607 | |
dc.subject | Graphic methods | en |
dc.subject | Information retrieval | en |
dc.subject | Information systems | en |
dc.subject | Natural language processing systems | en |
dc.subject | Systems engineering | en |
dc.subject | Websites | en |
dc.subject | Graph-based methods | en |
dc.subject | Legal information | en |
dc.subject | Legal information retrieval | en |
dc.subject | Multi dimensions | en |
dc.subject | Relevance judgment | en |
dc.subject | State-of-the-art methods | en |
dc.subject | Text summarization | en |
dc.subject | User satisfaction | en |
dc.subject | Laws and legislation | en |
dc.subject | Springer Verlag | en |
dc.title | Multi-dimension diversification in legal information retrieval | en |
dc.type | conferenceItem | en |