多模态音乐流派分类研究

针对自动的音乐流派分类这一音乐信息检索领域的热点问题,提出了多模态音乐流派分类的概念。针对传统的基于底层声学特征的音乐流派分类中的特征选择环节,实现了一种全新的特征选择算法——基于特征间相互影响的前向特征选择算法(IBFFS)。开创性地使用LDA(latent Dirichlet allocation)模型处理音乐标签,将标签属于每个流派的概率通过计算转换为对应的音乐属于每个流派的概率。     

Regional classification of Chinese folk songs based on CRF model

Music regional classification, which is an important branch of music automatic classification, aims at classifying folk songs according to different regional style. Chinese folk songs have developed various regional musical styles in the process of its evolution. Regional classification of Chinese folk songs can promote the development of music recommendation systems which recommending proper style of music to users and improve the efficiency of the music retrieval system. However, the accuracy of existing music regional classification systems is not high enough, because most methods do not consider temporal characteristics of music for both features extraction and classification. In this paper, we proposed an approach based on conditional random field (CRF) which can fully take advantage of the temporal characteristics of musical audio features for music regional classification. Considering the continuity, high dimensionality and large size of the audio feature data, we employed two ways to calculate the label sequence of musical audio features in CRF, which are Gaussian Mixture Model (GMM) and Restricted Boltzmann Machine (RBM). The experimental results demonstrated that the proposed method based on CRF-RBM outperforms other existing music regional classifiers with the best accuracy of 84.71% on Chinese folk songs datasets. Besides, when the proposed methods were applied to the Greek folk songs dataset, the CRF-RBM model also performs the best.

     

基于标签的歌曲相似性研究

基于音乐语义标签的音乐相似计算研究是音乐信息检索领域的另一个新的热点。该文提出一种基于标签挖掘的歌曲分类方法,以Last.fm音乐网站上的用户标签为特征进行歌曲相似性研究。文中将文本聚类中常用的潜在语义分析（LSA）方法和改进的K-means聚类方法相结合,应用于音乐语义标签的自动抽取;从音乐网站last.fm上抽取了6大类600首歌曲的8000多个用户标签作为音乐语义特征,并利用LSA进行歌曲向量的降维,形成了一个表示歌曲间相似关系的600×150维向量矩阵。最后利用K均值,根据音乐歌曲间的相似度进行歌曲分类,完成歌曲相似性比较。实验结果同没有LSA降维前及已有的HCC结果比较表明,使用文中提出的基于音乐标签的模型对歌曲进行分类,能得到较好的分类效果。     

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。     

On a serendipity-oriented recommender system based on folksonomy

This paper proposes a recommendation method that focuses on not only predictive accuracy but also serendipity. On many of the conventional recommendation methods, items are categorized according to their attributes (a genre, an authors, etc.) by the recommender in advance, and recommendation is made using the results of the categorization. In this study, impressions of users to items are adopted as a feature of the items, and each item is categorized according to the feature. Impressions used in such categorization are prepared using folksonomy, which classifies items using tags given by users. Next, the idea of “concepts” was introduced to avoid synonym and polysemy problems of tags. “Concepts” are impressions of users on items inferred from attached tags of folksonomy. The inferring method was also devised. A recommender system based on the method was developed in java language, and the effectiveness of the proposed method was verified through recommender experiments.     

Effective Tag Recommendation System Based on Topic Ontology Using Wikipedia and WordNet

In this paper, we proposed a novel approach based on topic ontology for tag recommendation. The proposed approach intelligently generates tag suggestions to blogs. In this approach, we construct topic ontology through enriching the set of categories in existing small ontology called as Open Directory Project. To construct topic ontology, a set of topics and their associated semantic relationships is identified automatically from the corpus‐based external knowledge resources such as Wikipedia and WordNet. The construction relies on two folds such as concept acquisition and semantic relation extraction. In the first fold, a topic‐mapping algorithm is developed to acquire the concepts from the semantic of Wikipedia. A semantic similarity‐clustering algorithm is used to compute the semantic similarity measure to group the set of similar concepts. The second is the semantic relation extraction algorithm, which derives associated semantic relations between the set of extracted topics from the lexical patterns between synsets in WordNet. A suitable software prototype is created to implement the topic ontology construction process. A Jena API framework is used to organize the set of extracted semantic concepts and their corresponding relationship in the form of knowledgeable representation of Web ontology language. Thus, Protégé tool provides the platform to visualize the automatically constructed topic ontology successfully. Using the constructed topic ontology, we can generate and suggest the most suitable tags for the new resource to users. The applicability of topic ontology with a spreading activation algorithm supports efficient recommendation in practice that can recommend the most popular tags for a specific resource. The spreading activation algorithm can assign the interest scores to the existing extracted blog content and tags. The weight of the tags is computed based on the activation score determined from the similarity between the topics in constructed topic ontology and content of the existing blogs. High‐quality tags that has the highest activation score is recommended to the users. Finally, we conducted experimental evaluation of our tag recommendation approach using a large set of real‐world data sets. Our experimental results explore and compare the capabilities of our proposed topic ontology with the spreading activation tag recommendation approach with respect to the existing AutoTag mechanism. And also discuss about the improvement in precision and recall of recommended tags on the data sets of Delicious and BibSonomy. The experiment shows that tag recommendation using topic ontology results in the folksonomy enrichment. Thus, we report the results of an experiment mean to improve the performance of the tag recommendation approach and its quality.     

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.

     

A Cascade-Hybrid Music Recommender System for mobile services based on musical genre classification and personality diagnosis

In this paper, we present a Cascade-Hybrid Music Recommender System intended to operate as a mobile service. Specifically, our system is a middleware that realizes the recommendation process based on a combination of music genre classification and personality diagnosis. A mobile user is able to query for music files by simply sending an example music file from his/her mobile device. In response to the user query, the system recommends music files that not only belong to the same genre as the user query, but also an attempt has been made to take into account both the user preferences as well as ratings from other users for candidate results. The recommendation mechanism is realized by applying the collaborative filtering technique of personality diagnosis. Using the minimum absolute error and the ranked scoring criteria, our approach is compared to existing recommendation techniques that rely on either collaborative filtering or content-based approaches. The outcome of the comparison clearly indicates that our approach exhibits significantly higher performance.     

结合节拍语义和MFCC声学特征的音乐流派分类

由于音乐节拍的强度、快慢、持续时间等是反映音乐不同流派风格的重要语义特征,而音乐节拍多属于由打击乐器所产生的低频部分,为此利用小波变换对音乐信号进行6层分解来提取低频节拍特征;针对节拍特征差异不明显的音乐流派,提出用描述频域能量包络的MFCC声学特征与节拍特征结合,并用基于音乐流派机理分析的8阶MFCC代替常用的12阶MFCC。对8类音乐流派实验仿真结果表明,基于语义特征和声学特征结合的方法,总体分类准确率可达68.37%,同时特征维数增加对分类时间影响很小。     

Co-clustering for Auditory Scene Categorization

Auditory scenes are temporal audio segments with coherent semantic content. Automatically classifying and grouping auditory scenes with similar semantics into categories is beneficial for many multimedia applications, such as semantic event detection and indexing. For such semantic categorization, auditory scenes are first characterized with either low-level acoustic features or some mid-level representations like audio effects, and then supervised classifiers or unsupervised clustering algorithms are employed to group scene segments into various semantic categories. In this paper, we focus on the problem of automatically categorizing audio scenes in unsupervised manner. To achieve more reasonable clustering results, we introduce the co-clustering scheme to exploit potential grouping trends among different dimensions of feature spaces (either low-level or mid-level feature spaces), and provide more accurate similarity measure for comparing auditory scenes. Moreover, we also extend the co-clustering scheme with a strategy based on the Bayesian information criterion (BIC) to automatically estimate the numbers of clusters. Evaluation performed on 272 auditory scenes extracted from 12-h audio data shows very encouraging categorization results. Co-clustering achieved a better performance compared to some traditional one-way clustering algorithms, both based on the low-level acoustic features and on the mid-level audio effect representations. Finally, we present our vision regarding the applicability of this approach on general multimedia data, and also show some preliminary results on content-based image clustering.     

基于短语的维吾尔文文本分类

文本特征表示是在文本自动分类中最重要的一个环节。在基于向量空间模型（VSM）的文本表示中特征单元粒度的选择直接影响到文本分类的效果。对于基于词袋模型（BOW）的维吾尔文文本分类效果不理想的问题,提出了一种基于统计方法的维吾尔语短语抽取算法并将抽取到的短语作为文本特征项,采用支持向量机（SVM）算法对维吾尔文文本进行了分类实验。实验结果表明,与以词为特征的文本分类相比,短语作为文本特征能够提高维吾尔文文本分类的准确率和召回率。     

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.

     

An approach of genetic programming for music emotion classification

In this paper, we suggest a new approach of genetic programming for music emotion classification. Our approach is based on Thayer’s arousal-valence plane which is one of representative human emotion models. Thayer’s plane which says human emotions is determined by the psychological arousal and valence. We map music pieces onto the arousal-valence plane, and classify the music emotion in that space. We extract 85 acoustic features from music signals, rank those by the information gain and choose the top k best features in the feature selection process. In order to map music pieces in the feature space onto the arousal-valence space, we apply genetic programming. The genetic programming is designed for finding an optimal formula which maps given music pieces to the arousal-valence space so that music emotions are effectively classified. k-NN and SVM methods which are widely used in classification are used for the classification of music emotions in the arousal-valence space. For verifying our method, we compare with other six existing methods on the same music data set. With this experiment, we confirm the proposed method is superior to others.     

A Music Recommendation System Based on Music and User Grouping

In this paper, we present a music recommendation system, which provides a personalized service of music recommendation. The polyphonic music objects of MIDI format are first analyzed for deriving information for music grouping. For this purpose, the representative track of each polyphonic music object is first determined, and then six features are extracted from this track for proper music grouping. Moreover, the user access histories are analyzed to derive the profiles of user interests and behaviors for user grouping. The content-based, collaborative, and statistics-based recommendation methods are proposed based on the favorite degrees of the users to the music groups, and the user groups they belong to. A series of experiments are carried out to show that our approach performs well.Part of the contents of this paper has been published in ACM proceedings of International Conference on Information and Knowledge Management, 2001     

Cross attention fusion for knowledge graph optimized recommendation

Knowledge Graph has attracted a wide range of attention in the field of recommendation, which is usually applied as auxiliary information to solve the problem of data sparsity. However, most recommendation models cannot effectively mine the associations between the items to be recommended and the entities in the Knowledge Graph. In this paper, we propose CAKR, a knowledge graph recommendation method based on the cross attention unit, which is similar to MKR, a multi-task feature learning general framework that uses knowledge graph embedding tasks to assist recommendation tasks. Specifically, we design a new method to optimize the feature interaction between the items and the corresponding entities in the Knowledge Graph and propose a feature cross-unit combined with the attention mechanism to enhance the recommendation effect. Through extensive experiments on the public datasets of movies, books, and music, we prove that CAKR is better than MKR and other knowledge graph recommendation methods so that the new feature cross-unit designed in this paper is effective in improving the accuracy of the recommendation system.

     

Automatic mood detection and tracking of music audio signals

Music mood describes the inherent emotional expression of a music clip. It is helpful in music understanding, music retrieval, and some other music-related applications. In this paper, a hierarchical framework is presented to automate the task of mood detection from acoustic music data, by following some music psychological theories in western cultures. The hierarchical framework has the advantage of emphasizing the most suitable features in different detection tasks. Three feature sets, including intensity, timbre, and rhythm are extracted to represent the characteristics of a music clip. The intensity feature set is represented by the energy in each subband, the timbre feature set is composed of the spectral shape features and spectral contrast features, and the rhythm feature set indicates three aspects that are closely related with an individual's mood response, including rhythm strength, rhythm regularity, and tempo. Furthermore, since mood is usually changeable in an entire piece of classical music, the approach to mood detection is extended to mood tracking for a music piece, by dividing the music into several independent segments, each of which contains a homogeneous emotional expression. Preliminary evaluations indicate that the proposed algorithms produce satisfactory results. On our testing database composed of 800 representative music clips, the average accuracy of mood detection achieves up to 86.3%. We can also on average recall 84.1% of the mood boundaries from nine testing music pieces.     

基于体裁的中文网页分类的特征选取

探讨基于体裁的中文网页分类的特征项选取问题.词汇特征方面,结合自动抽取和人工归纳的方式来获得.通过改进PAT树存储结构,进行序列挖掘来获得频繁字符串特征,使得文本分类系统摆脱对切词处理和词典的依赖,并提出了模糊字符串模式的特征表达方式.此外,特征集中融入了文本的形式特征,并根据网页的特点,引入链接信息特征.实现了基于体裁的中文网页分类系统,结果表明分类效果得到了有效的改善.