Analisis Prediktif Bitcoin dengan Metode SVM serta Pembobotan TIF-IDF Berbasis Data Narrative

Abstract

The cryptocurrency market has experienced significant volatility in recent years, making it challenging for investors to make informed decisions. This study aims to develop a predictive model for cryptocurrency price increases using TF-IDF (Term Frequency-Inverse Document Frequency) and SVM (Support Vector Machine) based on narrative data. Narrative data, such as news articles and social media posts, can provide valuable insights into investor sentiment and market trends. The proposed model extracts relevant features from narrative data using TF-IDF and employs SVM to classify cryptocurrency price movements into positive, negative, or neutral categories. Experimental results demonstrate the effectiveness of the proposed model in predicting cryptocurrency price increases, with an accuracy of over 70%. The findings suggest that narrative data can be a valuable source of information for cryptocurrency price prediction and that TF-IDF and SVM are effective methods for analyzing narrative data.

Keywords: Cryptocurrency, Price Prediction, TF-IDF, SVM, Narrative Data