Authorship attribution in the wild

Koppel, Moshe; Schler, Jonathan; Argamon, Shlomo

doi:10.1007/s10579-009-9111-2

Cited by 190 publications

(131 citation statements)

References 15 publications

Supporting

Mentioning

121

Contrasting

Unclassified

Order By: Relevance

“…Our ensemble-based algorithm is a variation of the method presented by Koppel et al [8] for the task of author identification. In the original approach, there is only one training example for each author and a number of simple classifiers is learned based on random feature subspacing.…”

Section: Ensemble-based Algorithmmentioning

confidence: 99%

“…It is possible to use this algorithm in combination with any text similarity measure. The cosine distance has provided good results in the experiments of [8] and is also used in this study.…”

Section: Ensemble-based Algorithmmentioning

confidence: 99%

“…With respect to RFSE, following the suggestion of Koppel et al [8] we used k1 =100 in each experiment. Using more than 100 iterations to build the ensemble does not improve significantly the results.…”

Section: -Wordsmentioning

confidence: 99%

“…We compare two different open-set classification methods; (i) One-class SVM (OC-SVM), which builds one model per genre using only the positive examples and (ii) a Random Feature Subspacing Ensemble (RFSE) method, originally proposed for author identification [8], a task with many similarities with AGI. Two of the most popular web-genre collections used in previous studies are used to evaluate these methods: the 7-genre and KI-04 corpora [9].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Open-Set Classification for Automated Genre Identification

Pritsos

Stamatatos

2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Automated Genre Identification (AGI) of web pages is a problem of increasing importance since web genre (e.g. blog, news, eshops, etc.) information can enhance modern Information Retrieval (IR) systems. The state-of-the-art in this field considers AGI as a closed-set classification problem where a variety of web page representation and machine learning models have intensively studied. In this paper, we study AGI as an open-set classification problem which better formulates the real world conditions of exploiting AGI in practice. Focusing on the use of content information, different text representation methods (words and character n-grams) are tested. Moreover, two classification methods are examined, one-class SVM learners, used as a baseline, and an ensemble of classifiers based on random feature subspacing, originally proposed for author identification. It is demonstrated that very high precision can be achieved in open-set AGI while recall remains relatively high.

show abstract

Section: Ensemble-based Algorithmmentioning

confidence: 99%

Section: Ensemble-based Algorithmmentioning

confidence: 99%

Section: -Wordsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Open-Set Classification for Automated Genre Identification

Pritsos

Stamatatos

2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…We discussed the state of the art of the authorship attribution problem in our previous work on application of syntactic n-grams for authorship attribution [1]; also see various related works on authorship attribution [7,8,9,10], among many others. Here we will just briefly state the problem.…”

Section: Authorship Attribution Problemmentioning

confidence: 99%

Syntactic Dependency-Based N-grams: More Evidence of Usefulness in Classification

Sidorov

Velásquez

Stamatatos

et al. 2013

Computational Linguistics and Intelligent Text Processing

View full text Add to dashboard Cite

Abstract. The paper introduces and discusses a concept of syntactic n-grams (sn-grams) that can be applied instead of traditional n-grams in many NLP tasks. Sn-grams are constructed by following paths in syntactic trees, so sngrams allow bringing syntactic knowledge into machine learning methods. Still, previous parsing is necessary for their construction. We applied sn-grams in the task of authorship attribution for corpora of three and seven authors with very promising results.

show abstract

Using psycholinguistic features for profiling first language of authors

Torney

Vamplew

Yearwood

2012

J Am Soc Inf Sci Tec

View full text Add to dashboard Cite

This study empirically evaluates the effectiveness of different feature types for the classification of the first language of an author. In particular, it examines the utility of psycholinguistic features, extracted by the Linguistic Inquiry and Word Count (LIWC) tool, that have not previously been applied to the task of author profiling. As LIWC is a tool that has been developed in the psycholinguistic field rather than the computational linguistics field, it was hypothesized that it would be effective, both as a single type feature set because of its psycholinguistic basis, and in combination with other feature sets, because it should be sufficiently different to add insight rather than redundancy. It was found that LIWC features were competitive with previously used feature types in identifying the first language of an author, and that combined feature sets including LIWC features consistently showed better accuracy rates and average F measures than were achieved by the same feature sets without the LIWC features. As a secondary issue, this study also examined how effectively first language classification scaled up to a larger number of possible languages. It was found that the classification scheme scaled up effectively to the entire 16 language collection from the International Corpus of Learner English, when compared with results achieved on just 5 languages in previous research.

show abstract

Authorship attribution in the wild

Cited by 190 publications

References 15 publications

Open-Set Classification for Automated Genre Identification

Open-Set Classification for Automated Genre Identification

Syntactic Dependency-Based N-grams: More Evidence of Usefulness in Classification

Using psycholinguistic features for profiling first language of authors

Contact Info

Product

Resources

About