Research of the estimated emotional components for the content analysis

Iskhakova, Anastasia; Iskhakov, Andrey; Meshcheryakov, Roman

doi:10.1088/1742-6596/1203/1/012065

Cited by 7 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the one hand, it is obvious that the resource can contain hidden sections that do not have any links from the mainframe of the project. This fact could be considered as insignificant, because, as a rule, a web page without any links addressed to it is not attractive from the point of view of the search of destructive materials [11,12]. We examined the main graph traversal algorithms:…”

Section: Results Of the Analysis Of Comment Threadsmentioning

confidence: 99%

Problems of Automation of the Aggression Analysis in Socio-Cyberphysical Environment

Kulagina¹,

Iskhakov²

2020

Proceedings of the 8th Scientific Conference on Information Technologies for Intelligent Decision Making Support (ITIDS 2020)

View full text Add to dashboard Cite

The article presents the results of the carried out interdisciplinary research that allowed creating algorithmic software to determine the structural composition of investigated cases of electronic communication that could be interpreted as aggressive. Such interpretation is offered to be carried out by means of the system analysis of the hypertext according to the following metrics: the intensity of discussion, density and reference. The research allowed us to define sociometric models of electronic communication accompanied by aggression; the primary detection of these models is the basis for carrying out the procedure of a detailed linguistic analysis of the text of the communication. Such an approach promotes the reduction of expenses for the management of aggression processes in the socio-cyberphysical environment. The received results prove the possibility to automate the analysis of information content and to raise the interpretability of phenomena in the socio-cyberphysical environment. We examined the aspects of crawling and parsing of modern webplatforms, focused data gathering, automatic and automated extraction of comments with the application of various technologies. The architecture of a system of multithreaded data gathering is offered.

show abstract

Section: Results Of the Analysis Of Comment Threadsmentioning

confidence: 99%

Problems of Automation of the Aggression Analysis in Socio-Cyberphysical Environment

Kulagina¹,

Iskhakov²

2020

Proceedings of the 8th Scientific Conference on Information Technologies for Intelligent Decision Making Support (ITIDS 2020)

View full text Add to dashboard Cite

show abstract

“…To solve this complex, multifaceted and ambiguous problem, it is necessary to consider a number of smaller private tasks, among which it is important to monitor the content of the virtual environment, create automatic means of recognizing and detecting possible destructive effect on the user and respond to such content in a timely manner. An example of a text material analysis is proposed in the work [1][2][3], the processing of audio materials (or audio parts of it) to detect perception anomalies is presented according to [4,5]. In this paper the authors give the results of research of graphic images, as an integral part of Internet content, which very often occurs on web resources of any orientation and can also negatively affect the condition of a person, change his mood, strengthen the impression of accompanying material of another type, and so on.…”

Section: Introductionmentioning

confidence: 99%

Detecting Images That Have a Destructive Impact on Users on the Internet

Iskhakova¹,

Meshcherykov²

2020

Proceedings of the 8th Scientific Conference on Information Technologies for Intelligent Decision Making Support (ITIDS 2020)

View full text Add to dashboard Cite

This article deals with the current problem of automated detection of facts of destructive influence of images on the Internet on the person. It should be noted that detecting aggressive content in images is more difficult than detecting an object and defining its category. The reason for this is that aggressive content has no specific color or object parameters, there are no common features for classification. In this case the use of convolutional neural networks with pre-learning is a successful solution. Due to increase in aggressive content on the Internet, the problem of identifying such objects in order to minimize its impact on users is acute. The paper presents the developed architecture of the neural network for solving the problem of image with aggressive content recognition. Experiments during 180 epochs showed that at the number of epochs ~ 95-100 during training and ~ during testing it is possible to achieve the classification accuracy 91.8% taking into account the top-5 results.

show abstract