Title: Estimating deflation representing people spreading in stream data and estimating a specific position

Authors: Takuma Toyoshima; Masaki Endo; Takuo Kikuchi; Shigeyoshi Ohno; Hiroshi Ishikawa

Addresses: Polytechnic Center Shiga, Ohtsu-shi, Shiga, Japan ' Division of Core Manufacturing, Polytechnic University, Kodaira-shi, Tokyo, Japan ' Division of Core Manufacturing, Polytechnic University, Kodaira-shi, Tokyo, Japan ' Division of Core Manufacturing, Polytechnic University, Kodaira-shi, Tokyo, Japan ' Graduate School of System Design, Tokyo Metropolitan University, Hino-shi, Tokyo, Japan

Abstract: With the expanded use of social media such as Twitter in recent years, it has become easy to add various information such as location data using mobile devices. Using those data, one can observe the real world without using physical sensors. Therefore, social media have high operational value as social sensors. As described herein, we aim to support decision-making for people who intend to visit a specific place at which an event or some trouble recently occurred. After proposing a method of real-time extraction of data reflecting a burst state showing people's concentration, their inactivity, and continuous flow and dispersion, we confirm the method's effectiveness. We will also try to estimate the location information of tweets for the purpose of further improving the estimation accuracy. Since few tweets have accurate location information, we use the content text of the tweet to find the tweet posted at the event occurrence location by machine learning. We will study changes in the accuracy of the proposed method due to the increase in the data to be analysed.

Keywords: deflation; burst; Twitter; real-time extraction; social sensor.

DOI: 10.1504/IJIIDS.2022.120150

International Journal of Intelligent Information and Database Systems, 2022 Vol.15 No.1, pp.104 - 123

Received: 04 Nov 2020
Accepted: 12 Apr 2021
Published online: 07 Jan 2022 *