Abstract: Video is an information-intensive media with much redundancy. Therefore, it is desirable to mine structure or semantics of video data for efficient browsing, indexing and highlight extraction. A naive user is interested in querying at the semantic level, rather than having to use features to describe his (her) concept. In most cases it is difficult to express concepts using feature matching and even a good match in terms of feature metrics may yield poor query results for the user. Researchers propose a novel video indexing using semantic pattern. First video preprocessing techniques are applied for processing the video to get audio and video cues. Then mining technique is applied to classify the events based on the semantic patterns. The task to discover useful and interesting patterns from a video is the main goal of video data mining. Data mining on multimedia is a challenging task, due to the complicated contents of multimedia data. Particularly the discovered patterns need to be supported by their semantic features. However, low-level features often have little meaning for naive users, who much prefer to identify content using high level semantics or concepts. This creates a gap between systems and their users that must be bridged for these systems to be used effectively. So, researchers propose to develop a knowledge-based video indexing using fuzzy based classification for mining semantic patterns.