Home // MMEDIA 2023, The Fifteenth International Conference on Advances in Multimedia // View article

Co-occurring Word Determination Used for Estimating Best Times for Viewing Cherry Blossoms

Authors:
Yusuke Takamori
Kenji Terada
Masaki Endo
Shigeyoshi Ohno
Hiroshi Ishikawa

Keywords: SNS, Twitter, estimating, cherry blossoms, co-occurrence.

Abstract:
As described herein, we propose a method to increase the amount of data used to estimate the best time for viewing seasonal organisms, particularly flowers. Observations of the best viewing times of seasonal organisms, conducted by the Japan Meteorological Agency, are required by those in the tourism industry and tourists, but the numbers of such official observations are decreasing: an accurate alternative is needed. As one alternative, we have investigated estimation of the best viewing seasons and times of biological organisms using Twitter, which is widely used in Japan. Specifically addressing difficulties posed by the decrease in data for estimating the best viewing times of biological organisms, which was a difficulty of earlier research, we propose a new method for improvement. The proposed method using co-occurring words is used for obtaining tweets related to seasonality. Combining the proposed method with the conventional method for estimating the best viewing times of seasonal organisms showed improved accuracy compared to that achieved using the conventional method alone.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2023

Publication date: April 24, 2023

Published in: conference

ISSN: 2308-4448

ISBN: 978-1-68558-038-4

Location: Venice, Italy

Dates: from April 24, 2023 to April 28, 2023