Music Recommendation Research Articles

Personalized Music Recommendation Based on Interest and Emotion: A Comparison of Multiple Algorithms

Recommendation algorithms can greatly improve the efficiency of information retrieval for users. This article briefly introduced recommendation algorithms based on association rules and algorithms based on interest and emotion analysis. After crawling music and comment data from the NetEase Cloud platform, a simulation experiment was conducted. Firstly, the performance of the Back-Propagation Neural Network (BPNN) in the interest and emotion-based algorithm for recommending music was tested, and then the impact of the proportion of emotion weight between comments and music on the emotion analysis-based algorithm was tested. Finally, the three recommendation algorithms based on association rules, user ratings, and interest and emotion analysis were compared. The results showed that when the BPNN used the dominant interest and emotion and secondary interest and emotion as judgment criteria, the accuracy of interest and emotion recognition for music and comments was higher. When the proportion of interest and emotion weight between comments and music was 6:4, the interest and emotion analysis-based recommendation algorithm had the highest accuracy. The interest and emotion-based recommendation algorithm had higher recommendation accuracy than the association rule-based and user rating-based algorithms, and could provide users with more personalized and emotional music recommendations.

Applications of AI and NLP to advance Music Recommendations on Voice Assistants

This paper builds on the popular use case of music requests on voice assistants like Siri, Google Assistant, Alexa, and others and explores the different AI and NLP techniques. The paper particularly focuses on how each of these techniques can be applied in the context of musical recommendations and experiences on voice assistants. It enumerates specific problems in the space of music recommendations and illustrates how specific techniques like multi-armed bandits can be applied.

MPL-TransKR: Multi-Perspective Learning Based on Transformer Knowledge Graph Enhanced Recommendation

The emergence of recommender system is aimed at solving the problems brought by information explosion to human life and even the development of human society. As a traditional recommendation technique, collaborative filtering often encounters sparsity and cold start problems in many recommendation scenarios. Therefore, researchers have found that the introduction of side information can solve these problems to a certain extent and improve the performance of recommender systems. The knowledge graph is a heterogeneous graph that contains rich semantic relationships among items. The Multi-Perspective Learning based on Transformer Knowledge Graph Enhanced Recommendation (MPL-TransKR) proposed in this paper uses the knowledge graph as the side information for input and introduces the multi-head self-attention mechanism by reasonably combining the transformer idea. While learning the high-order neighborhood information of the items, the long-distance information between the items is captured, and the weight value is assigned to the user through the attention mechanism to strengthen the user representation to realize user-item multi-perspective learning to enhance the performance of the recommendation model. Through extensive experiments using public datasets, we demonstrated that MPL-TransKR performs well in book and music recommendations, surpassing state-of-the-art baselines on several metrics.

DCF-MLSTM: a deep security content-based filtering scheme using multiplicative BiLSTM for movie recommendation system

Recently, the demand for online and offline recommendation systems has increased drastically. These systems are widely used in tourism, music, and video or movie recommendations. Currently, online movie streaming applications have gained huge attention. Providing better recommendations to the user is a challenging task for these applications. The content-based filtering (CBF) recommender system is a promising technique for these systems. However, traditional systems suffer from challenges such as cold-start problems, sparsity, and scalability. Consequently, we strengthen content-based recommendation algorithms by enriching the user-related and relevant product models with effective tendencies. The majority of previous work on classifiers has been in recommendation systems. To overcome these issues, we present a deep learning model that uses a deep neural network mechanism and a multiplicative BiLSTM model. This scheme uses embedding, weight updating, and preference learning processes to improve the recommendation system's performance. The performance of the proposed approach is measured in terms of MAE, MAP, Precision, Recall, and F-measure. The comparative performance shows that the proposed approach achieves better performance when compared with state-of-art movie recommendation techniques.

Research on long and short term music preference recommendation method integrating music emotional attention

Research on long and short term music preference recommendation method integrating music emotional attention

Research on long- and short-term music preference recommendation method integrating music emotional attention

Research on long- and short-term music preference recommendation method integrating music emotional attention

Background Music for Studying: A Naturalistic Experiment on Music Characteristics and User Perception

Despite the advances in context-aware background music (BM) recommendation, automated BM selection for studying-related contexts is still challenging in that the BM has to not only increase users' activation and task engagement but also avoid distraction. This study investigated how characteristics of BM linked to users' perceptions on task engagement and distraction. In a one-week <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">naturalistic</i> user experiment, thirty participants performed their everyday learning-related tasks with music selected by a BM player. We captured participants' learning contexts and perceptions via pop-up surveys and extracted fine-grained acoustic features for each song in their music listening history via audio processing techniques. Our findings support the power of music in fostering positive studying experience (e.g., perceived engagement) and reveal how several BM characteristics may link to perceived engagement in certain (but not all) conditions. Findings are discussed in relation to theoretical BM studies and implications for generating personalized and context-sensitive BM selections in music-enhanced learning environments.

A method for personalized music recommendation based on emotional multi-label

A method for personalized music recommendation based on emotional multi-label

Music recommendation based on affective image content analysis

Music has the ability to invest even the tritest scenes with so much meaning when added to them. Human perceptions of music and image can be closely related to each other, as both can incite similar sensations and emotions. Advertising agencies often make use of audio and music over their visuals to engage more audiences and to convey the emotions associated with their content more effectively. Matching visuals and music to comparable feelings might help people perceive emotions more vividly and strongly. This paper proposes an effective cross-modal neural network that provides music recommendations to a user by generating matches between images and music over a common emotional vector space. Using the valence and arousal values, a combined image-music pair dataset has been created. The images incorporated in this dataset are leveraged from the OASIS dataset while the music part is queried using Spotify API and YouTube. A Transfer Learning approach is proposed with Convolution Neural Network architecture for training on this dataset using MobileNetV3, ResNet-18 and EfficientNetB4 for the images and SampleCNN for the raw audio clips. For any given image input, a list of top-n music recommendations shall be outputted. This concept thus aims to generate music and image matching based on various deep hidden features over the emotion space of the two modalities.

Music recommender system using facial feature recognition

Facial expressions convey a lot of information about the emotional state of a person and even a small change in it can help to detect changes in an individual's mood. Researchers have conducted numerous studies in the fields of machine learning and computer vision, training ML models to detect various human emotions or moods based on captured facial features. In this research, a system was proposed which is capable of identifying the user's emotions and moods and suggesting a list of appropriate songs to improve their mood. Incorporating mood detection offers significant benefits to customers' mental health and personal satisfaction. The objective of this system is to capture the user's image, detect their mood and create a music player system that recommends music based on the user's real-time mood, utilizing a web camera and deep learning algorithms.

A Novel Feature-Based Music Recommendation System Considering the Uniqueness of Musical Items

A Novel Feature-Based Music Recommendation System Considering the Uniqueness of Musical Items

Automated Music Genre Classification through Deep Learning Techniques

Music Genre Classification (MGC) automatically categorizes music into different genres based on various musical attributes and features in a small number of music files. This is a crucial problem in the field of music information retrieval as it provides a way to organize and analyse large amounts of music files. MGC can be performed using conventional machine learning algorithms such as SVM, k-nearest neighbours, Decision trees, and neural networks. These algorithms learn to recognize different musical features and attributes to categorize the music files into different genres. The literature shows that the performance of conventional machine learning algorithms is inferior to deep learning algorithms such as CNN, RNN, etc., in various applications. Hence, the CNN algorithm is adapted to implement the classification of music files. This aims to classify music genres using CNN deep learning techniques. The performance of the algorithms for MGC can be evaluated using metrics such as accuracy, precision, recall, and F1-score. Additionally, the impact of different features and algorithms on the performance of MGC can be studied and compared. It has applications in areas such as automated music recommendation systems, music education, and music production. An accuracy of 83% is achieved by using CNN to accomplish the task of MGC.

Song Recommendation based on user’s Activity using Ensemble Learning and Clustering

The Song Recommendation System Based on User Schedule project is designed to provide users with personalized music recommendations that match their daily activities and mood swings. With a busy and hectic schedule, it can be challenging to find time to select music that matches a user’s current activity and mood. This project aims to provide a solution to this problem by analyzing the user’s daily schedule, including their planned activities and time of day, and using machine learning algorithms to recommend songs that fit their mood and energy level during each activity. The project utilizes a variety of technologies, such as React.js for the front-end and various machine learning algorithms using python for the back-end, to provide a user-friendly interface that allows users to input their schedules and receive song recommendations.

An Improved Music Recommendation System for Facial Recognition and Mood Detection

A user can choose the song using a number of strategies. This system finds the user’s mood and recommends music in accordance with that mood. The main objective is to accurately assess the user’s mood in a friendly way. Listeners are introduced to a machine learning application that plays song based on the user’s mood. Finding the user’s expression using Artificial Neural Network model, which expresses their emotions, can help with this. Compared to speaking, facial expressions reveal a lot more about a person. Users are finding it challenging to select the song that they want to listen to due to the daily increase in the number of songs being created. In order to address, our recommendation system gives users a selection of songs that are automatically added to their customized playlist based on the user’s mood.

Cross-Modal Variational Auto-Encoder for Content-Based Micro-Video Background Music Recommendation

In this paper, we propose a cross-modal variational auto-encoder (CMVAE) for content-based micro-video background music recommendation. CMVAE is a hierarchical Bayesian generative model that matches relevant background music to a micro-video by projecting these two multimodal inputs into a shared low-dimensional latent space, where the alignment of two corresponding embeddings of a matched video-music pair is achieved by cross-generation. Moreover, the multimodal information is fused by the product-of-experts (PoE) principle, where the semantic information in visual and textual modalities of the micro-video are weighted according to their variance estimations such that the modality with a lower noise level is given more weights. Therefore, the micro-video latent variables contain less irrelevant information that results in a more robust model generalization. Furthermore, we establish a large-scale content-based micro-video background music recommendation dataset, TT-150k, composed of approximately 3,000 different background music clips associated to 150,000 micro-videos from different users. Extensive experiments on the established TT-150k dataset demonstrate the effectiveness of the proposed method. A qualitative assessment of CMVAE by visualizing some recommendation results is also included.

A method for personalised music recommendation based on emotional multi-label

In this paper, a personalised music recommendation method based on emotion multi-label was proposed. First is the analysis of music emotion and music emotional label, then, the principal component analysis method is used to reduce the dimension to process the music features and complete the preprocessing. Secondly, construct the music emotion multi-label, and combine the cosine method to calculate the emotional multi-label similarity. Finally, the interest degree of emotional multi-label is calculated to obtain the user's interest degree of music resources, and the personalised recommendation method is optimised to realise the personalised recommendation of music. Experimental results show that the average coverage rate of personalised music recommendation of the proposed method is as high as 99.5%, the accuracy is 98.3%, and the recommendation time of 500 music items is only 18.9 s. Therefore, the recommendation effect of the proposed method is good, the accuracy of personalised music recommendation is improved, and the recommendation time is shortened.

A Robust Emotion Detection and Music Recommendation System using Mini Xception CNN

Abstract: Due to the emerging developments in artificial intelligence(AI) and machine learning(ML) Technology various systems are developed in recent days that late the human emotions and real time aspects of human psychology detection. Facial recognition based music recommendation system (MRS) is a interesting area of research where he plays an important role in handling the psychology patients. Face recognition system is extensively applied in security systems surveillance systems fault identification etc. Based on Emotion of the human, the music recommendation needs to be provided to analyse the phycology changes with the patients. The proposed approach is focused on considering the constraints available with the facial recognition system in existing frameworks such as deep feature extraction processing delay need to be reduce the hair using deep convolutional neural network(DCNN) architecture based Mini exception algorithm is developed. The system considered FER2013 image dataset that contains 35000 face images with automated labels that would be helpful for the presented approach to identify the emotion class accurately. The Mini exception algorithm used in CNN layers act as a lightweight system compared with various states of approaches. The proposed system removes the barrier between the existing frameworks and achieved the accuracy of 92%. The recommended music is derived from the music database and further mapped with respect to the algorithm result.

Emotion Based Music Recommendation System

Abstract: We propose a new approach for playing music automatically using facial emotion. Most of the existing approaches involve playing music manually, using wearable computing devices, or classifying based on audio features. Instead, we propose to change the manual sorting and playing. We have used a Convolutional Neural Network for emotion detection. For music recommendations, Pygame &amp; Tkinter are used. Our proposed system tends to reduce the computational time involved in obtaining the results and the overall cost of the designed system, thereby increasing the system’s overall accuracy. Testing of the system is done on the FER2013 dataset. Facial expressions are captured using an inbuilt camera. Feature extraction is performed on input face images to detect emotions such as happy, angry, sad, surprise, and neutral. Automatically music playlist is generated by identifying the current emotion of the user. It yields better performance in terms of computational time, as compared to the algorithm in the existing literature

Recommendation of Music Based on DASS-21 (Depression, Anxiety, Stress Scales) Using Fuzzy Clustering

The present study proposes a music recommendation service in a mobile environment using the DASS-21 questionnaire to distinguish and measure certain psychological state instability symptoms—viz. anxiety, depression, and stress—that anyone can experience regardless of job or age. In general, the outcome of the DASS-21 from almost every participant did not reveal any single psychological state out of the abovementioned three states. Therefore, the weighted scores were calculated for each scale and fuzzy clustering was used to cluster users into groups with similar states. For the initial dataset’s generation, we used the DASS inventory collected from the Open-Source Psychometrics Project conducted from 2017 to 2019 on approximately 39,000 respondents, and the results of the survey showed that the average scores for each scale were 23.6 points for depression, 17.4 for anxiety, and 23.3 for stress. Based on the datasets collected from fuzzy clustering, the individuals were classified into three groups: Group 1 was recommended with music for “high” depression, “high” anxiety, and “low” stress; Group 2 was recommended with music for “normal” depression, “low” anxiety, and “normal” stress; and Group 3 was recommended with music for “high” depression, “high” anxiety, and “high” stress. Especially, the largest numbers of recommended music in the three groups were for Group 1 with “High” depressive (4.64), Group 2 for “Low” anxiety (4.54), and Group 3 for “High” anxiety (4.76). In addition, to compare the results of fuzzy clustering with other data, the silhouette coefficient of the samples extracted with the same severity ratio and those generated by simple random sampling were 0.641 and 0.586, respectively, which were greater than 0. The proposed service can recommend not only the music of users with similar trends at all psychological states, but also the music of users with similar psychological states in part.

Emotion based Music Recommendation System

Emotion based Music Recommendation System