A Machine Learning Based Social Network Data Mining System for Better Search Engine Algorithm

• Main Authors: Himu, Sheik Erfan Ahmed, Ikram, Arafat Ibne, Abdullah, Kazi Mohammed, Choity, Tasnia Fahrin, Parves, Md. Shahazan, Hasan, Md. Rabiul, Uddin, Md Imtiaz
• Format: Conference Proceeding
• Language: English
• Subjects: Big data; Blogs; Data Mining; Machine learning; Machine learning algorithms; Search Engine; Search engines; Searching Algorithms; Semantic Web; Social Media Analysis; Social networking (online); User-generated content
• ISSN: 2837-8245
• DOI: 10.1109/WIECON-ECE60392.2023.10456428

Social networks provide access to a vast amount of information in a variety of ways. However, these are poorly organized. On the other hand, data search engines include limited and occasionally inaccurate data. So, it is possible to find the right analytical answer to a problem based on what people say on social media. Consequently, a machine learning-based social network data mining system will aid in the development of a superior search engine. Google is thought to be the best search engine in the world right now. People used to do SEO after building a website to improve its search engine ranking. Google's AI decides which search results to show based on how much traffic that website gets. But we still aren't getting the right results. But if we configure our search algorithm to utilize social network user-generated content based on their sentiment, we can obtain accurate search results. This paper proposed a machine learning-based data mining system from social networks where all data is collected from social networks using linear regression, polynomial regression, and percentile machine learning techniques and stores unstructured and pre-structured data in big data for data validation. With the help of some techniques, we can show that the information from social networks is a good way to solve our problems.