PENENTUAN RELEVANSI BERITA PADA TWITTER MENGGUNAKAN ALGORITMA DENSITY BASED SPATIAL CLUSTERING APPLICATION WITH NOISE DAN K-NEAREST NEIGHBOR
No Thumbnail Available
Teknik Informatika, Universitas Widyatama
Twitter makes the development of communications and information technology to be fast and easy to access, allowing users to hold and share any information freely. With Twitter every user can convey and share whatever happens either in the form of tweet information or news that is retweet through certain news media without knowing whether the news tweet is true, relevant or irrelevant. Based on the problem, we need a system of determining the relevance of news using Density Based Spatial Clustering Application with Noise (DBSCAN) and K-Nearest Neighbor (KNN) algorithms to find out relevant news and irrelevant news. Implementation using PHP programming language. Determination of News Relevance resulted 2 clusters with data testing using K-Nearest Neighbor (KNN) and test data using Clustering Applications with Noise (DBSCAN) to 3 data collection showed that the bigger the relevant data rate of a bigger news.
Twitter, News Relevance, Clustering, DBSCAN, Classification, KNN