Toward Web mining of cross-language query translations in digital libraries

Wang, Jenq-Haur; Lu, Wen-Hsiang; Chien, Lee‐Feng

doi:10.1007/s00799-004-0091-y

Cited by 8 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Query translation is a common approach to cross the language barrier and is highly applicable to multilingual DLs. Wang et al describe a query translation system which can be connected to any digital library with monolingual (Chinese or English) content (Wang et al, 2004;Wang et al, 2006). This system mines the web for translations that do not occur in the dictionary (new terms, proper names).…”

Section: Multilinguality In the Digital Librarymentioning

confidence: 99%

Multilinguality in the digital library

Diekema

2012

The Electronic Library

View full text Add to dashboard Cite

Purpose -Together, increasing globalization and the internet created fertile grounds for the establishment of multilingual digital libraries. Providing cross-lingual access to materials is of particular interest to political entities such as the European Union, which currently has 23 official languages, but also to multinational companies and countries that have different languages represented among their citizens. The main objective of this paper is to review the literature on multilingual digital libraries and provide an overview of this area. Design/methodology/approach -Based on a thorough literature search in four different databases, a core set of literature on multilingual digital libraries was retrieved. Literature on various aspects of this topic was reviewed. The paper is organized based on emerging themes directly drawn from the literature. Where warranted additional literature is brought in to provide necessary background information or clarification. Findings -Creating a multilingual digital library is a highly complex undertaking and typically requires a collaborative effort between different organizations and people with different areas of expertise. Enabling users to search across languages requires translation resources to cross the language barrier, which can be challenging depending on the language and resource availability. Additional challenges were found to be in data management (localization and language processing), representation (dealing with different fonts and character codes), development (creating international software, cross-cultural collaboration), and interoperability (system architecture and data sharing). Research in multilingual digital libraries was mostly system based involving experimental systems or system prototypes. Research limitations/implications -Most likely the literature review does not include all possible journal articles on multilingual digital libraries even though the literature searches done to obtain these articles were thorough and deliberate. Journal articles without the descriptors used in this search and those articles not indexed in the four different databases used in the search will not be included here. The review excludes cross-language information retrieval research unless it is directly related to existing multilingual digital libraries, or a connection to digital libraries in general is made in the paper itself. Originality/value -This paper provides the first literature review on the topic of multilingual digital libraries and provides a concise overview of relevant aspects in this area. The number of multilingual digital libraries is growing, as is the interest from the research community in these libraries to apply their research findings from cross-language information retrieval. This review article provides a valuable entry point to the field of multilingual digital libraries for researchers, practitioners, and other interested parties.

show abstract

Section: Multilinguality In the Digital Librarymentioning

confidence: 99%

Multilinguality in the digital library

Diekema

2012

The Electronic Library

View full text Add to dashboard Cite

show abstract

“…To empower browsing of multilingual digital collection, the present techniques commonly used still rely on language resources like multilingual dictionaries or thesauruses in which case a query words or phrases are translated before mining the databases. This technique is integrated with parallel corpora from Web resources such as utilized by a search engine LifeTrans [5].…”

Section: Introductionmentioning

confidence: 99%

Interactive document retrieval from multilingual digital repositories

Mizera–Pietraszko

2009

2009 Second International Conference on the Applications of Digital Information and Web Technologies

View full text Add to dashboard Cite

On the contrary to multilingual information retrieval systems, digital libraries usually do not provide any support for query translation. This paper presents a study aimed at exploring the role of the translation component in a digital library system designed for multilingual access to resources. Thus, we propose a user-oriented approach to one-word query formulation aimed at browsing English and French document collections of digital libraries. In particular, the focus is on comparison of bilingual to monolingual harvesting of digital collections. We explore some multilingual repositories based on OPAC (Online Public Catalogue) systems by utilization of metadata-based search. The evaluation of the digital library systems selected relies on usability criteria. In addition, the search results are grouped into categories according to the language pair phenomena. A diversity of the library contents oriented towards domain specific network communities is also regarded as a relevance factor at the query formulation stage.

show abstract

“…Halimi etc [9] used mixed filtering model to filter BBS data and introduced social network technology to perform aided analysis. Based on reply and quotation relationship between users, Wan Ying [10] constructed users' relationship network and topic social network for further mining. These achievements are all based on the relationship between reply and quotation to construct forum social network.…”

Section: Introductionmentioning

confidence: 99%

Online Music Community System based on SNM

Wang

Marzagalia

2013

JMM

View full text Add to dashboard Cite

Currently, the community structure algorithms proposed by most scholars are contraposing un-weighted network, while there still lacks research on the discovery of weighted network community. In un-weighted networks, the relation between the nodes is simplified as two states, which will neglect too much valuable information. To construct more actual structure of mined community, a novel discovery algorithm for weighted network community is put forward. The social network technology is implemented to a music forum. It can find out the social relation and interactive behavior of users in the music forum, to help the music resource to find suitable consumers. The music community system using social network mining is based on extracted replying and quoting information of users. For the user social network and topic social network of forum, two social network mining methods are proposed: One is applied to the user social network mining, which can find out the use groups who have similar interest in network; The other is applied to the topic social network mining, based on merging idea of weighted network community, and it can find out the similar groups of hot topics. Finally the mining process of network relationship in a certain music forum is described in detail as an example and it has verified the effectiveness of mining result

show abstract

Toward Web mining of cross-language query translations in digital libraries

Cited by 8 publications

References 17 publications

Multilinguality in the digital library

Multilinguality in the digital library

Interactive document retrieval from multilingual digital repositories

Online Music Community System based on SNM

Contact Info

Product

Resources

About