A Novel Framework for Discovering Emerging Topics in Streams of Social

Social networking became popular and they attract many other businesses as the social networks generate wealth of data that is useful for various organizations and researchers. The information being exchanged in social media include many forms of data known as multimedia. The conventional approaches do not work for mining such data as the data is of many types and from diverse sources. In order to handle such data and discover emerging topics, the recent approach was to use user links base on the tweets, mentions and replies. The probability model is used for anomaly detection in order to find the emergence of new topics. This knowledge will give provision for many associated businesses to make well informed decisions. For instance an advertisement agency can make use of this knowledge and post advertisements accordingly. When messages of being flown anomaly are detected and the anomaly score is computed and then aggregated to detect emerging topics accurately. This is practically implemented in this project. A prototype application is built to demonstrate the proof of concept. The empirical results show that the proposed approach is efficient and capable of finding emerging topics