STOCK PRICE PREDICTION USING A HYBRID CNN-LSTM MODEL WITH SENTIMENT DATA INTEGRATION FROM GOOGLE NEWS (Case Study: Tesla Stock)

Listianto, Demas Risdho (2025) STOCK PRICE PREDICTION USING A HYBRID CNN-LSTM MODEL WITH SENTIMENT DATA INTEGRATION FROM GOOGLE NEWS (Case Study: Tesla Stock). Tugas Akhir thesis, University of Technology Yogyakarta.

Text #Abstrak_Demas Risdho Listianto_5211811017_Sains Data.pdf Download (140kB)

Abstract

ABSTRACT This study aims to forecast Tesla’s stock price (TSLA) using a hybrid CNN-LSTM model integrated with financial news sentiment analysis. The dataset includes historical TSLA stock prices sourced from Yahoo Finance and over 100,000 news articles collected from Google News between 2020 and 2025. Sentiment analysis was conducted using the VADER algorithm to classify articles as positive, neutral, or negative. These sentiment scores were then integrated with the stock data and engineered into time-series features, including lag variables. The Convolutional Neural Network (CNN) component was used to extract local spatial patterns from the input data, while the Long Short-Term Memory (LSTM) component captured long-term temporal dependencies. The model was trained and evaluated across eight experimental scenarios to assess the impact of combining stock price data with sentiment information on prediction accuracy. Model performance was measured using RMSE, MAE, and MAPE metrics. Results indicate that incorporating sentiment data significantly reduces prediction error compared to using stock price data alone. The best performance was achieved in the Lag-7 configuration, with MAE, RMSE, and MAPE values of 14.9379, 10.7227, and 4.05%, respectively, substantially outperforming previous benchmark studies in terms of prediction error. Keywords: CNN, LSTM, TSLA, VADER, Hybrid Model, Sentiment.

Item Type:	Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects:	T Technology > T Technology (General)
Divisions:	Fakultas Sains Dan Teknologi > Data Science
Depositing User:	Sains Data
Date Deposited:	07 Aug 2025 03:48
Last Modified:	07 Aug 2025 03:48
URI:	http://eprints.uty.ac.id/id/eprint/18382

Actions (login required)

View Item