Knowledge Graphs as Context Models: Improving the Detection of Cross-Language Plagiarism with Paraphrasing

Abstract

Abstract. Cross-language plagiarism detection attempts to identify and extract automatically plagiarism among documents in different languages. Plagiarized fragments can be translated verbatim copies or may alter their structure to hide the copying, which is known as paraphrasing and is more difficult to detect. In order to improve the paraphrasing detection, we use a knowledge graph-based approach to obtain and compare context models of document fragments in different languages. Experimental results in German-English and Spanish-English crosslanguage plagiarism detection indicate that our knowledge graph-based approach offers a better performance compared to other state-of-the-art models.

A systematic study of knowledge graph analysis for cross-language plagiarism detection

“…The Cross-Language Knowledge Graph Analysis (CL-KGA) model (Franco-Salvador et al, 2013) uses a multilingual semantic network to create knowledge graphs that model the context of documents. The model achieved interesting results for CL plagiarism detection, also in cases of paraphrasing (Franco-Salvador et al, 2014a). However, it left unanswered questions -relationship with CL-ESA, contributions of WSD, vocabulary expansion, etc.…”

Author Profiling and Plagiarism Detection

“…Evaluation Corpora In the author profiling task at PAN 2013 [58] participants approached the task of identifying age and gender in a large corpus collected from social media, and age was annotated with three classes: 10s (13)(14)(15)(16)(17), 20s (23)(24)(25)(26)(27), and 30s (33)(34)(35)(36)(37)(38)(39)(40)(41)(42)(43)(44)(45)(46)(47). At PAN 2014, we continued to study the gender and age aspects of the author profiling problem, however, four corpora of different genres were considered-social media, blogs, Twitter, and hotel reviews-both in English and Spanish.…”

“…The granularity for CL-KGA is the closest to 1.0, the best possible value. A more detailed analysis can be found in [16].…”

“…Schler et al [61] studied the effect of age and gender in the writing style in blogs; they gathered over 71,000 blogs and obtained a set of stylistic features like non-dictionary words, parts-of-speech, function words and hyperlinks, combined with content features, such as word unigrams with the highest information gain. They modeled age in three classes -10s (13)(14)(15)(16)(17), 20s (23)(24)(25)(26)(27) and 30s (33-47)-obtaining an accuracy of about 80% for gender identification and about 75% for age identification. They showed that language features in blogs correlate with age, as reflected in, for example, the use of prepositions and determiners.…”

A Cross-domain and Cross-language Knowledge-based Representation of Text and its Meaning

“…The Cross-Language Knowledge Graph Analysis (CL-KGA) model (FrancoSalvador et al, 2013a,b) uses a multilingual semantic network to create knowledge graphs that model the context of documents. The model achieved interesting results for CL plagiarism detection, also in cases of paraphrasing (Franco-Salvador et al, 2014a). However, it left unanswered questions -relationship with CL-ESA, contributions of WSD, vocabulary expansion, etc.…”