An Ontological Approach to Detecting Irrelevant and Unreliable Information on Web-Resources and Social Networks

Author(s):  
Mykola Dyvak ◽  
Andriy Melnyk ◽  
Svitlana Mazepa ◽  
Mykola Stetsko
Sensors ◽  
2021 ◽  
Vol 21 (19) ◽  
pp. 6611
Author(s):  
Alexandra Cernian ◽  
Nicoleta Vasile ◽  
Ioan Stefan Sacala

The exponential increase in social networks has led to emergent convergence of cyber-physical systems (CPS) and social computing, accelerating the creation of smart communities and smart organizations and enabling the concept of cyber-physical social systems. Social media platforms have made a significant contribution to what we call human behavior modeling. This paper presents a novel approach to developing a users’ segmentation tool for the Romanian language, based on the four DISC personality types, based on social media statement analysis. We propose and design the ontological modeling approach of the specific vocabulary for each personality and its mapping with text from posts on social networks. This research proposal adds significant value both in terms of scientific and technological contributions (by developing semantic technologies and tools), as well as in terms of business, social and economic impact (by supporting the investigation of smart communities in the context of cyber-physical social systems). For the validation of the model developed we used a dataset of almost 2000 posts retrieved from 10 social medial accounts (Facebook and Twitter) and we have obtained an accuracy of over 90% in identifying the personality profile of the users.


Author(s):  
MOHAMMED NAZIM UDDIN ◽  
TRONG HAI DUONG ◽  
KYEONG-JIN OH ◽  
JIN-GUK JUNG ◽  
GEUN-SIK JO

Experts finding, one of the most important tasks in social networks, is aimed at identifying individuals with relevant expertise or experience in a given topic. Several approaches have been proposed for finding experts in social networks from documents or web repositories. However, the semantic approach for modeling the information to find experts has not yet been explored. In this paper, we propose a novel method to index the academic information in an ontology-based model for finding and ranking the experts in a particular domain. Additionally, we propose an effective method to construct the academic social network by exploring the relations among the experts and measuring the score of each expert. The score of an expert is measured considering the contributions of relevant publications and relationships among other expert candidates. It is very efficient to find and ranking experts to take advantage of the millions of candidate experts being with relationships. An experiment conducted to evaluate our model shows that experts finding and ranking with an ontological approach integrated with the social network is more effective than other approaches.


Author(s):  
Mark E. Dickison ◽  
Matteo Magnani ◽  
Luca Rossi

2006 ◽  
Vol 27 (2) ◽  
pp. 108-115 ◽  
Author(s):  
Ana-Maria Vranceanu ◽  
Linda C. Gallo ◽  
Laura M. Bogart

The present study investigated whether a social information processing bias contributes to the inverse association between trait hostility and perceived social support. A sample of 104 undergraduates (50 men) completed a measure of hostility and rated videotaped interactions in which a speaker disclosed a problem while a listener reacted ambiguously. Results showed that hostile persons rated listeners as less friendly and socially supportive across six conversations, although the nature of the hostility effect varied by sex, target rated, and manner in which support was assessed. Hostility and target interactively impacted ratings of support and affiliation only for men. At least in part, a social information processing bias could contribute to hostile persons' perceptions of their social networks.


Sign in / Sign up

Export Citation Format

Share Document