Identification of Duplication in Questions Posed on Knowledge Sharing Platform Quora using Machine Learning Techniques

Quora, an online question-answering platform has a lot of duplicate questions i.e. questions that convey the same meaning. Since it is open to all users, anyone can pose a question any number of times this increases the count of duplicate questions. This paper uses a dataset comprising of question pairs (taken from the Quora website) in different columns with an indication of whether the pair of questions are duplicates or not. Traditional comparison methods like Sequence matcher perform a letter by letter comparison without understanding the contextual information, hence they give lower accuracy. Machine learning methods predict the similarity using features extracted from the context. Both the traditional methods as well as the machine learning methods were compared in this study. The features for the machine learning methods are extracted using the Bag of Words models- Count-Vectorizer and TFIDF-Vectorizer. Among the traditional comparison methods, Sequence matcher gave the highest accuracy of 65.29%. Among the machine learning methods XGBoost gave the highest accuracy, 80.89% when Count-Vectorizer is used and 80.12% when TFIDF-Vectorizer is used.

REPRESENTASI PENGETAHUAN PADA WEB SEMANTIK UNTUK MENINGKATKAN NILAI EFEKTIFITAS PENCARIAN DATA SURAT (STUDI KASUS: PT ANGKASA PURA I CABANG SEPINGGAN BALIKPAPAN )

PT. Angkasa Pura I branch Sepinggan, Balikpapan still uses the web with a little interactive (browsing or searching for certain information). The system still has some constraints between, the data searched only in accordance with the origin of the sender of the letter. Information gleaned from web searches will make it hard for users to find other broader content, consequently if the search keyword does not match the text contained in the information, the information will not appear. The method used is semantic ontology method, and use protege as ontology design, then use netbean and library Jena as interface design. The methodology used in this final project is the methodology methodology of Guidelines for Rapid APPlication Engineering (GRAPPLE). The GRAPPLE methodology is a simplified framework of the Rapid Application Development (RAD) system development method. GRAPPLE methodology is a system development methodology that each stage consists of several actions, and each action produces a UML diagram. UML is a good modeling language to apply in development. UML has the benefit of providing an overview of systems that are easily understood by developers. UML also makes it easy to translate diagram form into coding. The results obtained in making this application shows that the new application can cover the constraints existing in the previous application, so as to increase the effectiveness of searching a letter. Comparison of test results proved that new applications more minimize the constraints that occur. Keywords: Semantic Web, Ontology, Protege, Letter