Anuj@DPIL-FIRE2016: A Novel Paraphrase Detection Method in Hindi Language Using Machine Learning

摘要

Every language possesses plausible several interpretations. With the evolution of web, smart devices and social media it has become a challenging task to identify these syntactic or semantic ambiguities. In Natural Language Processing, two statements written using different words having same meaning is termed as paraphrasing. At FIRE 2016, we have worked upon the problem of detecting paraphrases for the given Shared Task DPIL (Detecting Paraphrases in Indian Languages) in Hindi Language specifically. This paper proposes a novel approach to identify if two statements are paraphrased or not using various machine learning algorithms like Random Forest, Support Vector Machine, Gradient Boosting and Gaussian Naive Bayes on the given training data set of two subtasks. In cross validation experiments, Random Forest outperforms the other methods with Fl-score of 0.94. We have extended our work by adding few more features and using the former best classifier resulting in improvement of Fl-score by 1%. The experimental results depict that our algorithm got the highest F1-score and accuracy and hence, secured the first rank in Hindi language in this shared task among all participants. Our novel approach can be used in various applications such as question-answering system, document clustering, machine translation, text summarization, plagiarism detection and many more.