• Anusha Villuri
• Yugandhar Bokka
• P. Haritha
• R Prasanthi Kumari

Author Profiling (AP) is a type of text classification and information extraction technique, which extracts the author’s hidden information from their written texts. This technique extracts the author’s demographic features like gender, age, location, nativity language, educational background, personality traits etc., by analysing their written texts. The author profiling techniques are used in various applications like marketing, security, forensic analysis etc. Researchers proposed several methods for differentiating the writing style of authors by using stylistic features, machine learning techniques and deep learning techniques. In this work, we proposed two approaches based on word embedding techniques with machine learning algorithms and word embedding techniques with similarity measures for author profiling. We are concentrating on prediction of gender and age group of authors. The PAN competition 2014 dataset is used in this work for experimentation. In the first proposed approach, we used word embedding techniques for representing words as vectors. These word vectors are used for representing the documents as vectors. The machine learning algorithms are used for training these document vectors. These algorithms develop the classification model to predict the accuracy of gender and age prediction. In the second approach, we used word embedding techniques for representing words as vectors and similarity measures are used to find the similarity among the documents. For gender and age dimensions, the second approach attained good accuracies than first approach. The proposed approaches attained best accuracies for gender and prediction when compared with other popular approaches to author profiling.

Author Profiling, Gender Prediction, Age Prediction, Word2Vec, Doc2Vec, Similarity Measures.

"Word embedding Technique with Similarity Measures based Approach for Author Profiling", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 6, page no.f447-f453, June-2023, Available :http://www.jetir.org/papers/JETIR2306562.pdf

"Word embedding Technique with Similarity Measures based Approach for Author Profiling", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 6, page no. ppf447-f453, June-2023, Available at : http://www.jetir.org/papers/JETIR2306562.pdf