Analisis Sistem Informasi Banjir Berbasis Media Twitter

Irza Utami, Marzuki Marzuki

Abstract


Uji keakurataan data media sosial Twitter sebagai sumber informasi banjir telah diteliti melalui penyaringan twit yang memuat informasi banjir di wilayah DKI Jakarta pada tahun 2015-2017. Twit yang memuat kata kunci banjir dikumpulkan untuk mendapatkan lokasi-lokasi banjir yang pernah dilaporkan oleh pengguna Twitter. Lokasi banjir ini selanjutnya dipetakan dan divalidasi menggunakan data curah hujan dari satelit GPM (Global Precipitation Measurement) yang disediakan oleh NASA (National Aeronautics and Space Administratration). Distribusi lokasi banjir dalam 3 tahun dianalisa berdasarkan intensitas rata-rata curah hujan tahunan. Kemudian, dilakukan uji regresi linear antara jumlah twit dengan intensitas curah hujan harian di setiap lokasi banjir. Hasil penelitian memperlihatkan sebaran lokasi banjir berada pada wilayah yang memiliki intensitas curah hujan yang tinggi.  Nilai uji regresi linear antara jumlah twit dengan intensitas curah hujan pada lokasi banjir sebesar 0.431. Nilai regresi 0,431 diperoleh setelah twit banjir kiriman dikeluarkan. Dengan demikian selain intensitas curah hujan, banjir kiriman juga memberikan dampak yang sangat besar sebagai penyebab banjir di Jakarta.

 

The utilization Twitter social media data as a source of flood information has been investigated through filtering tweets containing flood information in the DKI Jakarta area  during 2015-2017. Tweets containing the keyword flood are collected to get flood locations that have been reported by Twitter users. Furthermore, the location of this flood is mapped and validated using rainfall data from the GPM (Global Precipitation Measurement) satellite provided by NASA (National Aeronautics and Space Administration). The distribution of flood locations is analyzed based on the average annual rainfall intensity. Then the relationship between the number of tweets and the intensity of daily rainfall at each flood location was examined using a linear regression. The distribution of flood locations is concentrated in the areas with high rainfall intensity. The value of linear regression coefficient between the number of tweets with the intensity of rainfall at flood locations is 0.431. However, a regression coefficient of 0.431 was obtained after the tweet containing flood of submissions was excluded. Thus, in addition to the intensity of rainfall, flood of submissions also has a very large impact as a cause of flooding in Jakarta


Full Text:

PDF

References


Eilander, D., Trambauer, O., Wagemaker, J., Loenen, V. A., “Harvesting Social Media for Generation of Near Real-Time Food Maps”, Procedia Engineering ,176-183 (2016).

Wang, Z., Ye, X., Social Media Analytics for Natural Disaster Management, International Journal of Geographical Information Science, 49-72 (2018).

Fohringer, J., Dransch, D., Kreibich, H., Schröter, K., Social Media as an Information Source for Rapid Flood Inundation Mapping, Natural Hazards and Earth Sistem Sciences Journal, 2725-2738 (2015).

Indonesia Masuk 5 Besar Pengguna Twitter, http://www.beritasatu.com/digital-life/428591-indonesia-masuk-lima-besar pengguna-Twitter.html, diakses Januari (2019).

Jakarta dan Bandung 10 Besar Kota Teriuh Twitter, https://tekno.tempo.co/read/451350/Jakarta-dan-bandung-10-besar-kota-teriuh-twitter. diakses Januari (2019).




DOI: https://doi.org/10.25077/jfu.9.1.67-72.2020

Refbacks

  • There are currently no refbacks.


Copyright (c) 2020 Irza Utami, Marzuki Marzuki

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

Published by:

Jurusan Fisika, FMIPA Universitas Andalas

Kampus Unand Limau Manis Padang Sumatera Barat 25163

Telepon 0751-73307

Email:jfu@sci.unand.ac.id