A Study on Discovery Method of Hot Topics Based on Smart Campus Big Data Platform

To study and pose an modified discovery method for campus hot topics. The method is to extract word segmentation and keywords from the campus news collected on information big data platform through ICTCLAS word segmentation system, construct the text knowledge representation model based on vector space modal and finally add up the word frequency of headlines to determine k initial clustering centers and improve K-Means algorithm so as to obtain the campus hot topics.