Retrieving Data From Social Network Platforms: a State-of-Art Review(Conference Proceedings)
Analyzing public concerns gives feedback to organizations about goods and services. The means of collecting social concerns differ for each social media platform. There is a need to explore the literature about automated tools applied by researchers to collect social issues. The goal of this paper is to provide an overview of data collection methods from social media platforms. This is to guide the collection of public concerns through machine learning approaches. Following the preferred reporting items for systematic reviews and meta-analyses standards, we collected 2180 articles from the Google Scholar database based on keywords. We screened the reviews for relevancy based on abstract and title. Considering the exclusion criteria, we removed all full articles not related to social media platforms. We manually analyzed only 298 articles to identify classifications within methods of data collection. From the reviews, we retrieved five categories of data collection. These include; manual observations, self-report surveys, public repositories, existing licensed tools, public application programming interfaces, and web crawlers or scrapers. Considering sample networks of Facebook, Twitter, and YouTube, we explored the trends of tools. And we stated the pros and cons. In conclusion, to collect social concerns at no cost and in large amounts, we recommend using open libraries of public application programming interfaces and web scrapers. In the future, we plan to extract public data using the recommended trendy automated tools for each category and sample social networks
Authoured by: Harriet Sibitenda , Awa Diattara, B. A. Cheikh, Assitan Traore
Academic units: Faculty of Science
Departments: Computer Science and Information Systems