A scientific seminar was held at the Institute of Information Technology of ANAS on the topic "Application of clustering methods to detect fake profiles in social networks." The report was presented by the graduate student of the Institute Khayala Ahmedova. She provided information on machine learning approaches used to detect fake profiles on social networks.
Talking about the types of fake profiles on social networks, she emphasized the purpose of creating these fake profiles, the dangers they pose and gave a brief overview of the methods used to detect fake profiles. The speaker brought to the attention of information on the rating of the use of social networking sites and applications.
Then the speaker gave information about machine learning methods used to detect fake profiles, the abundance of signs used in them, and also emphasized the main advantages and disadvantages of the most commonly used clustering methods.
She said that different machine learning methods are used to detect fake profiles on social networks, but researchers are more focused on finding more specialized types of fake profiles. This can lead to the fact that other types of fake profiles will not be detected, as well as to malicious infection of the network. In this regard, it is very important to develop common methods in social networks that allow to identify the maximum number of fake profiles. Clustering methods can offer similar approaches by grouping fake profiles. With their use, there will be a great need for well-scalable machine learning methods in high-volume networks.
She provided detailed information on research, research groups, work done and ongoing research projects.
In conclusion, an exchange of views on the report took place, answers to questions were announced. The head of the department, Ph.D. in engineering, associate professor Yadigar Imamverdiyev recommended continuing research on the development of methods and algorithms based on machine learning approaches for detecting fake profiles on social networks.
© All rights reserved. Citing to www.ict.az is necessary upon using news