Lab9 - nikhitasharma/RTBigDataAnalytics_Project GitHub Wiki

Task: Audio Classification

1. Extract audio features from a video. We extracted audio features from a video input file using Apache Spark and stored features as a text file.
2. Send the audio features to Storm through Kafka as streaming data.
   We deployed the Storm Topology after creating Kafka Spout for incoming streams of audio features and two bolts for processing these features for the purpose of classification using the decision tree algorithm.
3. Try to use a Classification Model for audio features.
   We tried to use Decision tree algorithm for classification, trained and tested the new model. We reported F-Measure, Precision, Recall and Confusion Matrix as below.
4. Implement Storm topology for audio recognition.
   We tried to create a topology for processing audio data and give audio recognition results.

Screenshots for the above lab task can be found at: https://github.com/nikhitasharma/RTBigDataAnalytics_Project/issues/6