Publication Type : Conference Paper
Publisher : Advances in Intelligent Systems and Computing
Source : Advances in Intelligent Systems and Computing, Springer Verlag, Volume 750, p.489-498 (2019)
ISBN : 9789811318818
Keywords : Adaptive boosting, Author Profiling, Big data, Cloud computing, Data mining, Decision trees, Inverse problems, Matrix algebra, N-grams, On-line communities, Population statistics, Random forests, Social networking (online), Support vector machines, Term frequency-inverse document frequencies, Term frequencyinverse document frequency (TF-IDF), Text processing, Text representation, Variety identification
Campus : Coimbatore
School : School of Engineering
Center : Computational Engineering and Networking
Department : Electronics and Communication
Year : 2019
Abstract : Social media has evolved to be a crucial part of life today for everyone. With such a global population communicating with each other, comes the accumulation of large amounts of social media data. This data can be categorized as “Big Data”, owing to its large quantity. It contains valuable information in the form of the demographics of authors on online platforms; the analysis of which is required in certain scenarios to maintain decorum in the online community. Here, we have analyzed Twitter data, which is the training data of the PAN@CLEF(link sends e-mail) 2017 shared task contest, to identify the gender, as well as the language variety of the author. It is available in four different languages, namely, English, Spanish, Portuguese, and Arabic. Both Document-Term Matrix (DTM) and Term Frequency-Inverse Document Frequency (TF-IDF) have been used for text representation. The classifiers used are SVM, AdaBoost, Decision Tree, and Random Forest. © 2019, Springer Nature Singapore Pte Ltd.
Cite this Research Publication : R. V. Chacko, M. Kumar, A., and Dr. Soman K. P., “Experimental Study of Gender and Language Variety Identification in Social Media”, in Advances in Intelligent Systems and Computing, 2019, vol. 750, pp. 489-498.