SocioNews Aggregation Analysis

• Published in: 2022 International Conference on Computer Communication and Informatics (ICCCI) pp. 1 - 6
• Main Authors: S, Nithiya, G, Parimala
• Format: Conference Proceeding
• Language: English
• Published: IEEE 25.01.2022
• Subjects: Blogs; Graph clustering; K-means; Real-time systems; SNS; Social networking (online); Software; Support vector machines; Technological innovation; tweets; Urban areas
• DOI: 10.1109/ICCCI54379.2022.9740806

In Social networking, networks are now one of present-day's most emerging innovations. This is a need for each of us, with its own combination of virtues and vices. Today fake news and disinformation have been a major issue causing across the globe. Creating an algorithm with great accuracy would also be a surprise, as it will have a humongous effect on the prevailing social issues as well as on the real-time political scene. Social networking and web news stories act as the primary source of news and statistics for people because it can be quickly viewed, has a low budget cost and is accessible only one click away immediately. It has so many negative impacts, though, including no verification of the origins or credibility and legitimacy of the views shown. Therefore, we have suggested a new approach for the identification of false news that incorporates empathy as an important function to improve accuracy. It also looks at the efficiency of the approach proposed using different data sets. The result suggests that the solution suggested is working well. In addition, the distinction between the structures is often carefully observed and analysed. News aggregators have the capacity to handle the load of tremendous volumes of regular produced news inputs. However, many of such systems mainly concentrate on finding and showing the common floating news, yet are unable to express the generalist opinion on the same. Different news media outlets give biased news on which different people will have multiple views. This might result in negative effects. So, to help in this question we provide a broad range to news topics which are divided into different main headings or major topics. By giving these a person can have a diverse news understanding on the same topic. Previously the news was taken only from mainstream news channels. And due to this, some news is heavily highlighted than the others as per the media bias. In older models, there were a lot of shortcomings. We address these shortcomings and we have made a few changes. We have included a bigger data set with multiple new additions such as emoticons, city, country, intensifiers and abbreviations. We would be using tweets as a main source for tweets along with traditional news sources. The most popular tweet with the maximum number of retweets is selected and categorised. From those, we then find a keyword particularly use in similar topical tweets. Based on these tweets we prepare a list of top twenty tweets and after that, we compare them with the news media outlets to verify and confirm their content. Once that is done, we create a matrix. This matrix is based on the user analysis and their reaction to that particular post. Then a graph is created based on multiple users input and this graph is analysed. This will tell us about the news and the sentiments attached to that news.