论文信息 - Word Unigram Weighing for Author Profiling at PAN 2018: Notebook for PAN at CLEF 2018

Word Unigram Weighing for Author Profiling at PAN 2018: Notebook for PAN at CLEF 2018

We present our system for the author profiling task at PAN 2018 on gender identification on Twitter. The submitted system uses word unigrams, character 1to 5-grams and emoji unigrams as features to train a logistic regression classifier. We explore the impact of three different word unigram weighing schemes on our system’s performance. Our submission achieved accuracies of 77.42% for English, 74.64% for Spanish, and 73.20% for Arabic tweets. It ranked 15th out of 23 competitors.

Mark Cieliebak | Pius von Däniken | Ralf Grubenmann

[1] Gaël Varoquaux,et al. Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[2] Bela Gipp,et al. TF-IDuF : A Novel Term-Weighting Scheme for User Modeling based on Users’ Personal Document Collections , 2017 .

[3] Benno Stein,et al. Overview of the 5th Author Profiling Task at PAN 2017: Gender and Language Variety Identification in Twitter , 2017, CLEF.

[4] Benno Stein,et al. Overview of PAN 2018 - Author Identification, Author Profiling, and Author Obfuscation , 2018, CLEF.

[5] Kilian Q. Weinberger,et al. Feature hashing for large scale multitask learning , 2009, ICML '09.

[6] Benno Stein,et al. Overview of the 4th Author Profiling Task at PAN 2016: Cross-Genre Evaluations , 2016, CLEF.

[7] Steven Bird,et al. NLTK: The Natural Language Toolkit , 2002, ACL.

[8] Proceedings of the Fourth Workshop on NLP for Similar Languages, Varieties and Dialects, VarDial 2017, Valencia, Spain, April 3, 2017 , 2017, VarDial.

[9] Benno Stein,et al. Overview of the 6th Author Profiling Task at PAN 2018: Multimodal Gender Identification in Twitter , 2018, CLEF.

[10] Malvina Nissim,et al. N-GrAM: New Groningen Author-profiling Model , 2017, CLEF.