A novel method for personalized music recommendation

With the development of digital music technologies, it is an interesting and useful issue to recommend the ‘favored music’ from large amounts of digital music. Some Web-based music stores can recommend popular music which has been rated by many people. However, three problems that need to be resolved in the current methods are: (a) how to recommend the ‘favored music’ which has not been rated by anyone, (b) how to avoid repeatedly recommending the ‘disfavored music’ for users, and (c) how to recommend more interesting music for users besides the ones users have been used to listen. To achieve these goals, we proposed a novel method called personalized hybrid music recommendation, which combines the content-based, collaboration-based and emotion-based methods by computing the weights of the methods according to users’ interests. Furthermore, to evaluate the recommendation accuracy, we constructed a system that can recommend the music to users after mining users’ logs on music listening records. By the feedback of the user’s options, the proposed methods accommodate the variations of the users’ musical interests and then promptly recommend the favored and more interesting music via consecutive recommendations. Experimental results show that the recommendation accuracy achieved by our method is as good as 90%. Hence, it is helpful for recommending the ‘favored music’ to users, provided that each music object is annotated with the related music emotions. The framework in this paper could serve as a useful basis for studies on music recommendation.     

Music recommendation using text analysis on song requests to radio stations

Recommending appropriate music to users has always been a difficult task. In this paper, we propose a novel method in recommending music by analyzing the textual input of users. To this end, we mine a large corpus of documents from a Korean radio station’s online bulletin board. Each document, written by the listener, is composed of a song request associated with a brief, personal story. We assume that such stories are closely related with the background of the song requests and thus, our system performs text analysis to recommend songs that were requested from other similar stories. We evaluate our system using conventional metrics along with a user evaluation test. Results show that there is close correlation between document similarity and song similarity, indicating the potential of using text as a source to recommending music.     

Develop a personalized intelligent music selection system based on heart rate variability and machine learning

Music often plays an important role in people’s daily lives. Because it has the power to affect human emotion, music has gained a place in work environments and in sports training as a way to enhance the performance of particular tasks. Studies have shown that office workers perform certain jobs better and joggers run longer distances when listening to music. However, a personalized music system which can automatically recommend songs according to user’s physiological response remains absent. Therefore, this study aims to establish an intelligent music selection system for individual users to enhance their learning performance. We first created an emotional music database using data analytics classifications. During testing, innovative wearable sensing devices were used to detect heart rate variability (HRV) in experiments, which subsequently guided music selection. User emotions were then analyzed and appropriate songs were selected by using the proposed application software (App). Machine learning was used to record user preference, ensuring accurate and precise classification. Significant results generated through experimental validation indicate that this system generates high satisfaction levels, does not increase mental workload, and improves users’ performance. Under the trend of the Internet of Things (IoT) and the continuing development of wearable devices, the proposed system could stimulate innovative applications for smart factory, home, and health care.     

Learning to embed music and metadata for context-aware music recommendation

Contextual factors greatly influence users’ musical preferences, so they are beneficial remarkably to music recommendation and retrieval tasks. However, it still needs to be studied how to obtain and utilize the contextual information. In this paper, we propose a context-aware music recommendation approach, which can recommend music pieces appropriate for users’ contextual preferences for music. In analogy to matrix factorization methods for collaborative filtering, the proposed approach does not require music pieces to be represented by features ahead, but it can learn the representations from users’ historical listening records. Specifically, the proposed approach first learns music pieces’ embeddings (feature vectors in low-dimension continuous space) from music listening records and corresponding metadata. Then it infers and models users’ global and contextual preferences for music from their listening records with the learned embeddings. Finally, it recommends appropriate music pieces according to the target user’s preferences to satisfy her/his real-time requirements. Experimental evaluations on a real-world dataset show that the proposed approach outperforms baseline methods in terms of precision, recall, F1 score, and hitrate. Especially, our approach has better performance on sparse datasets.     

基于协同过滤和标签的混合音乐推荐算法研究

结合音乐这一特定的推荐对象,针对传统单一的推荐算法不能有效解决音乐推荐中的准确度问题,提出一种协同过滤技术和标签相结合的音乐推荐算法。该算法先通过协同过滤技术确定相似用户,再通过相似用户对某一歌手的标签评分预测另一用户对该歌手的偏好程度,从而选择更符合用户喜好的音乐进行推荐,以此提升个性化推荐效率,为优化音乐推荐系统提供参考方法。     

Toward an understanding of the history and impact of user studies in music information retrieval

Most Music Information Retrieval (MIR) researchers will agree that understanding users’ needs and behaviors is critical for developing a good MIR system. The number of user studies in the MIR domain has been gradually increasing since the early 2000s, reflecting this growing appreciation of the need for empirical studies of users. However, despite the growing number of user studies and the wide recognition of their importance, it is unclear how great their impact has been in the field: on how systems are developed, how evaluation tasks are created, and how MIR system developers in particular understand critical concepts such as music similarity or music mood. In this paper, we present our analysis on the growth, publication and citation patterns, topics, and design of 198 user studies. This is followed by a discussion of a number of issues/challenges in conducting MIR user studies and distributing the research results. We conclude by making recommendations to increase the visibility and impact of user studies in the field.     

Emotion-based music recommendation by affinity discovery from film music

With the growth of digital music, the development of music recommendation is helpful for users to pick desirable music pieces from a huge repository of music. The existing music recommendation approaches are based on a user’s preference on music. However, sometimes, it might better meet users’ requirement to recommend music pieces according to emotions. In this paper, we propose a novel framework for emotion-based music recommendation. The core of the recommendation framework is the construction of the music emotion model by affinity discovery from film music, which plays an important role in conveying emotions in film. We investigate the music feature extraction and propose the Music Affinity Graph and Music Affinity Graph-Plus algorithms for the construction of music emotion model. Experimental result shows the proposed emotion-based music recommendation achieves 85% accuracy in average.     

Predicting user demographics from music listening information

Online activities such as social networking, online shopping, and consuming multi-media create digital traces, which are often analyzed and used to improve user experience and increase revenue, e. g., through better-fitting recommendations and more targeted marketing. Analyses of digital traces typically aim to find user traits such as age, gender, and nationality to derive common preferences. We investigate to which extent the music listening habits of users of the social music platform Last.fm can be used to predict their age, gender, and nationality. We propose a feature modeling approach building on Term Frequency-Inverse Document Frequency (TF-IDF) for artist listening information and artist tags combined with additionally extracted features. We show that we can substantially outperform a baseline majority voting approach and can compete with existing approaches. Further, regarding prediction accuracy vs. available listening data we show that even one single listening event per user is enough to outperform the baseline in all prediction tasks. We also compare the performance of our algorithm for different user groups and discuss possible prediction errors and how to mitigate them. We conclude that personal information can be derived from music listening information, which indeed can help better tailoring recommendations, as we illustrate with the use case of a music recommender system that can directly utilize the user attributes predicted by our algorithm to increase the quality of it’s recommendations.

     

Music similarity-based approach to generating dance motion sequence

In this paper, we propose a novel approach to generating a sequence of dance motions using music similarity as a criterion to find the appropriate motions given a new musical input. Based on the observation that dance motions used in similar musical pieces can be a good reference in choreographing a new dance, we first construct a music-motion database that comprises a number of segment-wise music-motion pairs. When a new musical input is given, it is divided into short segments and for each segment our system suggests the dance motion candidates by finding from the database the music cluster that is most similar to the input. After a user selects the best motion segment, we perform music-dance synchronization by means of cross-correlation between the two music segments using the novelty functions as an input. We evaluate our system’s performance using a user study, and the results show that the dance motion sequence generated by our system achieves significantly higher ratings than the one generated randomly.     

Implementing situation-aware and user-adaptive music recommendation service in semantic web and real-time multimedia computing environment

With the advent of the ubiquitous era, many studies have been devoted to various situation-aware services in the semantic web environment. One of the most challenging studies involves implementing a situation-aware personalized music recommendation service which considers the user’s situation and preferences. Situation-aware music recommendation requires multidisciplinary efforts including low-level feature extraction and analysis, music mood classification and human emotion prediction. In this paper, we propose a new scheme for a situation-aware/user-adaptive music recommendation service in the semantic web environment. To do this, we first discuss utilizing knowledge for analyzing and retrieving music contents semantically, and a user adaptive music recommendation scheme based on semantic web technologies that facilitates the development of domain knowledge and a rule set. Based on this discussion, we describe our Context-based Music Recommendation (COMUS) ontology for modeling the user’s musical preferences and contexts, and supporting reasoning about the user’s desired emotions and preferences. Basically, COMUS defines an upper music ontology that captures concepts on the general properties of music such as titles, artists and genres. In addition, it provides functionality for adding domain-specific ontologies, such as music features, moods and situations, in a hierarchical manner, for extensibility. Using this context ontology, we believe that logical reasoning rules can be inferred based on high-level (implicit) knowledge such as situations from low-level (explicit) knowledge. As an innovation, our ontology can express detailed and complicated relations among music clips, moods and situations, which enables users to find appropriate music. We present some of the experiments we performed as a case-study for music recommendation.     

基于深度学习挖掘用户搜索主题研究

主要研究了基于深度学习技术挖掘用户搜索主题相关的感兴趣内容。通过深度挖掘算法分析用户搜索记录、查询历史以及用户感兴趣的相关文档视为用户搜索主题数据的来源,进而挖掘兴趣主题。挖掘模型主要采用向量空间模型,将用户搜索主题模型表示成用户搜索主题向量形式。形成主题和用户兴趣关系网,用户搜索主题向量的构造过程:选择一组用户查询词,并对它们进行深度挖掘分类,最后用它们构造用户搜索主题特征向量,进而分析用户兴趣点。结合用户随着时间的变化,以及过程中有不用的搜索词,以及无关的搜索噪声词去掉,调整兴趣度,用户搜索主题需要具有更新学习机制,动态跟踪了用户兴趣变化趋势。该用户搜索主题研究过程克服了数据稀疏、类别偏差、扩展性差等缺点。实验结果表明,该模型识别用户搜索主题准确率良好。     

A Novel framework of EEG-based user identification by analyzing music-listening behavior

This paper introduces a novel framework for user identification by analyzing neuro-signals. Studies regarding Electroencephalography (EEG) revealed that such bio-signals are sensitive, hard to forge, confidential, and unique which the conventional biometric systems like face, speaker, signature and voice lack. Traditionally, researchers investigated the neuro-signal patterns by asking users to perform various imaginary, visual or calculative tasks. In this work, we have analyzed this neuro-signal pattern using audio as stimuli. The EEG signals are recorded simultaneously while user is listening to music. Four different genres of music are considered as users have their own preference and accordingly they respond with different emotions and interests. The users are also asked to provide music preference which acts as a personal identification mechanism. The framework offers the benefit of uniqueness in neuro-signal pattern even with the same music preference by different users. We used two different classifiers i.e. Hidden Markov Model (HMM) based temporal classifier and Support Vector Machine (SVM) for user identification system. A dataset of 2400 EEG signals while listening to music was collected from 60 users. User identification performance of 97.50 % and 93.83 % have been recorded with HMM and SVM classifiers, respectively. Finally, the performance of the system is also evaluated on various emotional states after showing different emotional videos to users.     

面向推荐系统的音乐特征抽取

音乐推荐系统是指根据用户的历史浏览数据,从候选库中推荐给用户可能喜欢的音乐的一种新型网络服务。该系统的关键在于需要对整个数据库按照音乐风格进行分类,基于此提出一种新的音乐特征处理方法来完成音乐库分类,以有效实现音乐推荐。该方法首先为候选音乐库构建常规的音乐特征数据集,然后基于分形理论对数据集进行属性约简,获取每一首音乐的推荐特征向量,并且依据特征向量的特点,定义了一种新的距离度量方法。在包含六种风格的音乐数据库的实验中,仿真结果证明了提出的音乐推荐特征和距离度量的有效性,与现有的基于内容的音乐检索研究相比,音乐推荐特征的使用极大地降低了对数据库存储量的需求,对音乐推荐系统的网络开发具有很好的应用价值。     

Collaborative user modeling with user-generated tags for social recommender systems

With the popularity of social media services, the sheer amount of content is increasing exponentially on the Social Web that leads to attract considerable attention to recommender systems. Recommender systems provide users with recommendations of items suited to their needs. To provide proper recommendations to users, recommender systems require an accurate user model that can reflect a user’s characteristics, preferences and needs. In this study, by leveraging user-generated tags as preference indicators, we propose a new collaborative approach to user modeling that can be exploited to recommender systems. Our approach first discovers relevant and irrelevant topics for users, and then enriches an individual user model with collaboration from other similar users. In order to evaluate the performance of our model, we compare experimental results with a user model based on collaborative filtering approaches and a vector space model. The experimental results have shown the proposed model provides a better representation in user interests and achieves better recommendation results in terms of accuracy and ranking.     

On the efficiency of user identification: a system-based approach

In the Internet era, users’ fundamental privacy and anonymity rights have received significant research and regulatory attention. This is not only a result of the exponential growth of data that users generate when accomplishing their daily task by means of computing devices with advanced capabilities, but also because of inherent data properties that allow them to be linked with a real or soft identity. Service providers exploit these facts for user monitoring and identification, albeit impacting users’ anonymity, based mainly on personal identifiable information or on sensors that generate unique data to provide personalized services. In this paper, we report on the feasibility of user identification using general system features like memory, CPU and network data, as provided by the underlying operating system. We provide a general framework based on supervised machine learning algorithms both for distinguishing users and informing them about their anonymity exposure. We conduct a series of experiments to collect trial datasets for users’ engagement on a shared computing platform. We evaluate various well-known classifiers in terms of their effectiveness in distinguishing users, and we perform a sensitivity analysis of their configuration setup to discover optimal settings under diverse conditions. Furthermore, we examine the bounds of sampling data to eliminate the chances of user identification and thus promote anonymity. Overall results show that under certain configurations users’ anonymity can be preserved, while in other cases users’ identification can be inferred with high accuracy, without relying on personal identifiable information.     

MUSIPER: a system for modeling music similarity perception based on objective feature subset selection

We explore the use of objective audio signal features to model the individualized (subjective) perception of similarity between music files. We present MUSIPER, a content-based music retrieval system which constructs music similarity perception models of its users by associating different music similarity measures to different users. Specifically, a user-supplied relevance feedback procedure and related neural network-based incremental learning allows the system to determine which subset of a set of objective features approximates more accurately the subjective music similarity perception of a specific user. Our implementation and evaluation of MUSIPER verifies the relation between subsets of objective features and individualized music similarity perception and exhibits significant improvement in individualized perceived similarity in subsequent music retrievals.     

User Personality and User Satisfaction with Recommender Systems

In this study, we show that individual users’ preferences for the level of diversity, popularity, and serendipity in recommendation lists cannot be inferred from their ratings alone. We demonstrate that we can extract strong signals about individual preferences for recommendation diversity, popularity and serendipity by measuring their personality traits. We conducted an online experiment with over 1,800 users for six months on a live recommendation system. In this experiment, we asked users to evaluate a list of movie recommendations with different levels of diversity, popularity, and serendipity. Then, we assessed users’ personality traits using the Ten-item Personality Inventory (TIPI). We found that ratings-based recommender systems may often fail to deliver preferred levels of diversity, popularity, and serendipity for their users (e.g. users with high-serendipity preferences). We also found that users with different personalities have different preferences for these three recommendation properties. Our work suggests that we can improve user satisfaction when we integrate users’ personality traits into the process of generating recommendations.     

Large-scale user modeling with recurrent neural networks for music discovery on multiple time scales

The amount of content on online music streaming platforms is immense, and most users only access a tiny fraction of this content. Recommender systems are the application of choice to open up the collection to these users. Collaborative filtering has the disadvantage that it relies on explicit ratings, which are often unavailable, and generally disregards the temporal nature of music consumption. On the other hand, item co-occurrence algorithms, such as the recently introduced word2vec-based recommenders, are typically left without an effective user representation. In this paper, we present a new approach to model users through recurrent neural networks by sequentially processing consumed items, represented by any type of embeddings and other context features. This way we obtain semantically rich user representations, which capture a user’s musical taste over time. Our experimental analysis on large-scale user data shows that our model can be used to predict future songs a user will likely listen to, both in the short and long term.

     

Collaborative user modeling for enhanced content filtering in recommender systems

Recommender systems, which have emerged in response to the problem of information overload, provide users with recommendations of content suited to their needs. To provide proper recommendations to users, personalized recommender systems require accurate user models of characteristics, preferences and needs. In this study, we propose a collaborative approach to user modeling for enhancing personalized recommendations to users. Our approach first discovers useful and meaningful user patterns, and then enriches the personal model with collaboration from other similar users. In order to evaluate the performance of our approach, we compare experimental results with those of a probabilistic learning model, a user model based on collaborative filtering approaches, and a vector space model. We present experimental results that show how our model performs better than existing alternatives.     

Improving music recommendation by incorporating social influence

In the past decades, a large number of music pieces are uploaded to the Internet every day through social networks, such as Last.fm, Spotify and YouTube, that concentrates on music and videos. We have been witnessing an ever-increasing amount of music data. At the same time, with the huge amount of online music data, users are facing an everyday struggle to obtain their interested music pieces. To solve this problem, music search and recommendation systems are helpful for users to find their favorite content from a huge repository of music. However, social influence, which contains rich information about similar interests between users and users’ frequent correlation actions, has been largely ignored in previous music recommender systems. In this work, we explore the effects of social influence on developing effective music recommender systems and focus on the problem of social influence aware music recommendation, which aims at recommending a list of music tracks for a target user. To exploit social influence in social influence aware music recommendation, we first construct a heterogeneous social network, propose a novel meta path-based similarity measure called WPC, and denote the framework of similarity measure in this network. As a step further, we use the topological potential approach to mine social influence in heterogeneous networks. Finally, in order to improve music recommendation by incorporating social influence, we present a factor graphic model based on social influence. Our experimental results on one real world dataset verify that our proposed approach outperforms current state-of-the-art music recommendation methods substantially.