Published Paper Details:

Abstract - In recent years, with the development and use of big data, deep learning has begun to attract more and more attention. Convolutional neural network, a deep learning neural network, plays an important role in facial image recognition. This paper combines convolutional neural networks' knowledge of micro interpretation technology with an automatic music recognition algorithm to create patterns that recognize micro faces, speak, and recommend music based on your mood. The facial micro expression recognition model developed in this article uses FER 2013 and the recognition rate is 62.1%. After determining the similarity, the content-based music recognition algorithm was used to extract the feature vector of the song, and the cosine similarity algorithm was used for music recognition. This research helps improve the effectiveness of music recognition, and r elated results can also be applied to the use of music recognition in areas such as emotion regulation. Keywords: deep learning, face macro recognition, CNN, FER2013, CB, music recommendation algorithm