Feature Exploration for Authorship Attribution of Lithuanian Parliamentary Speeches

Kapočiūtė-Dzikienė, Jurgita; Utka, Andrius; Šarkutė, Ligita

doi:10.1007/978-3-319-10816-2_12

Cited by 3 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, this approach could be explored across languages to demonstrate that its usefulness is not limited to an American context. There is already work that explores classifying authors by political affiliation in multipole languages and we would hope that here too reference to domain specific knowledge would be of use (Abd et al, 2020;Kapočiūtė-Dzikienė et al, 2014;Lapponi et al, 2018).…”

Section: Future Directionsmentioning

confidence: 99%

Style over substance: A psychologically informed approach to feature selection and generalisability for author classification

Holmes

Cribbin

Ferenczi

2023

Computers in Human Behavior Reports

View full text Add to dashboard Cite

Section: Future Directionsmentioning

confidence: 99%

Style over substance: A psychologically informed approach to feature selection and generalisability for author classification

Holmes

Cribbin

Ferenczi

2023

Computers in Human Behavior Reports

View full text Add to dashboard Cite

“…On the other hand, the number of prefixes is not large: in Lithuanian there are only thirty six prefixes [28], while modern English has fifty seven prefixes. Lithuanian is highly inflective, ambiguous (47 per cent of words are ambiguous), has rich vocabulary (0.5 million headwords) and has complex word derivation system (e.g., seventy eight suffixes for diminutives) [29]. Verbs have 3 conjugations, and are inflected by four tenses, three persons, two numbers, and three moods.…”

Section: Language Specific Featuresmentioning

confidence: 99%

“…There is a significant difference between frequency of unigrams in Lithuanian and in other (English, Polish, Serbian) languages (see Table I). Previous experiments in authorship identification using Lithuanian texts have demonstrated that content-features are more useful compared with function words or POS tags [29], while best results were obtained with word-level character tetra-grams and a set of lexical, morphological, and character features [30].…”

Section: Language Specific Featuresmentioning

confidence: 99%

Open Class Authorship Attribution of Lithuanian Internet Comments using One-Class Classifier

Venčkauskas

Karpavičius

Damaševičius

et al. 2017

Annals of Computer Science and Information Systems

Self Cite

View full text Add to dashboard Cite

Internet can be misused by cyber criminals as a platform to conduct illegitimate activities (such as harassment, cyber bullying, and incitement of hate or violence) anonymously. As a result, authorship analysis of anonymous texts in Internet (such as emails, forum comments) has attracted significant attention in the digital forensic and text mining communities. The main problem is a large number of possible of authors, which hinders the effective identification of a true author. We interpret open class author attribution as a process of expert recommendation where the decision support system returns a list of suspected authors for further analysis by forensics experts rather than a single prediction result, thus reducing the scale of the problem. We describe the task formally and present algorithms for constructing the suspected author list. For evaluation we propose using a simple Winner-Takes-All (WTA) metric as well as a set of gain-discount model based metrics from the information retrieval domain (mean reciprocal rank, discounted cumulative gain and rank-biased precision). We also propose the List Precision (LP) metric as an extension of WTA for evaluating the usability of the suspected author list. For experiments, we use our own dataset of Internet comments in Lithuanian language and consider the use of language-specific (Lithuanian) lexical features together with general lexical features derived from English language. For classification we use one-class Support Vector Machine (SVM) classifier. The results of experiments show that the usability of open class author attribution can be improved considerably by using a set of language-specific lexical features together with general lexical features, while the proposed method can be used to reduce the number of suspected authors thus alleviating the work of forensic linguists.

show abstract

Authorship Attribution of Internet Comments with Thousand Candidate Authors

Kapočiūtė-Dzikienė

Utka

Šarkutė

2015

Communications in Computer and Information Science

View full text Add to dashboard Cite

Feature Exploration for Authorship Attribution of Lithuanian Parliamentary Speeches

Cited by 3 publications

References 9 publications

Style over substance: A psychologically informed approach to feature selection and generalisability for author classification

Style over substance: A psychologically informed approach to feature selection and generalisability for author classification

Open Class Authorship Attribution of Lithuanian Internet Comments using One-Class Classifier

Authorship Attribution of Internet Comments with Thousand Candidate Authors

Contact Info

Product

Resources

About