Approach to processing of data from social networks for detecting public opinion on quality of educational services

А. Д. Хомоненко, V. L. Dashonok, Kseniya Ivanova, Dinara Kassymova

2017

Closed access Metadata (CSL-JSON)
DOI 10.1109/scm.2017.7970707
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

An approach is proposed to identify public opinion on the basis of large data extracted from social networks, including: latent-semantic analysis, identification of computer-generated texts, automatic classification of text fragments and identification of public opinion based on Mamdani's fuzzy inference algorithm. Examples are given of assessing the semantic proximity of documents, the classification of documents based on the method. The indicators for assessing public opinion on the evaluation of the quality of educational services are substantiated.