Social Media Data in the Big Data Environment
| Autor: | O. OSAULENKO, O. HOROBETS |
|---|---|
| Jazyk: | ukrajinština |
| Rok vydání: | 2021 |
| Předmět: | |
| Zdroj: | Naukovij Vìsnik Nacìonalʹnoï Akademìï Statistiki, Oblìku ta Auditu, Iss 3-4, Pp 23-31 (2021) |
| Druh dokumentu: | article |
| ISSN: | 2520-6834 2521-1323 |
| Popis: | The article contains results of a study of social media data (SMD) which, being distinct from conventional data by their origin, require special methods for collection, processing and analysis. As shown by a literature review, in spite of great many research publications devoted to social media research and big data analysis, the SMD potential as a big data component still remains inadequately explored. Two approaches to research and analysis of SMD were highlighted in course of the study, in which SMD are addressed as an object of Internet statistics and an object of big data. When SMD are explored as an object of Internet statistics, collection of anonymized data is performed using the services that have network protocols for collection and analysis of data on social media customers using statistical methods. When SMD are explored as an object of big data, the collection is performed mostly by artificial intellect, whereas the storage and processing is operated by databases designed for large scopes of data and software with statistical data processing applications. The social media most popular with users in 2020 were identified in the study. Statistical indicators for assessment of users’ feedback, available now for statistical assessments of social media communities, are given. The study revealed several problems which solutions would require, apart from a multifaceted and complex approach to collection and processing, highly competent teams of specialists in various subject fields, including experts in computations, experts in machine learning and statisticians. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|