Social Media Analytics to Understand the Construction Industry Sentiments

  • Shrestha, K. Joseph (Department of Engineering, Engineering Technology, and Surveying, East Tennessee State University) ;
  • Mani, Nirajan (Engineering Technology Department, Fitchburg State University) ;
  • Kisi, Krishna P. (Department of Engineering Technology, Texas State University) ;
  • Abdelaty, Ahmed (Department of Civil and Architectural Engineering and Construction Management, University of Wyoming)
  • Published : 2022.06.20

Abstract

The use of social media to disseminate news and interact with project stakeholders is increasing over time in the construction industry. Such social media data can be analyzed to get useful insights of the industry such as demands of new housing construction and satisfaction of construction workers. However, there has been a limited attempts to analyze social media data related to the construction industry. The objective of this study is to collect and analyze construction related tweets to understand the overall sentiments of individuals and organizations about the construction industry. The study collected 87,244 tweets from April 6, 2020, to April 13, 2020, which had hashtags relevant to the construction industry. The tweets were then analyzed to evaluate its sentiments polarity (positive or negative) and sentiment intensity or scores (-1 to +1). Descriptive statistics were produced for the tweets and the sentiment scores were visualized in a scatterplot to show the trend of the sentiment scores over time. The results shows that the overall sentiment score of all the tweets was slightly positive (0.0365). Negative tweets were retweeted and marked as favorite by more users on average than the positive ones. More specifically, the tweets with negative sentiments were retweeted by 2,802 users on average compared to the tweets with positive sentiments (247 average retweet count). This study can potentially be expanded in the future to produce a real time indicator of the construction market industry such as the increased availability of construction jobs, improved wage rates, and recession.

Keywords

Acknowledgement

The authors would like to thank Mr. Matthew Sweets for writing the Python script and collecting the data.