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.     

SFFS：低延迟的面向小文件的分布式文件系统

社交网站和电子商务等网络服务发展迅速,这类服务需要存储大量图片、音乐、微博文本等小文件。传统的分布式存储系统,如HDFS（Hadoop distributed file system）,是面向大文件而设计的,在存储小文件时会产生元数据开销过大,访问延迟较高等问题,不能适应存储海量小文件的应用环境。分析了TFS（Taobao file system）的系统架构和读写流程,发现TFS在每次读/写过程中至少要建立3次网络连接,增大了读写延迟。针对海量小文件存储带来的挑战和TFS存在的问题,提出了一种新的低延迟、高可用的面向海量小文件的分布式存储方案,并实现了分布式文件系统SFFS（small-file file system）。性能测试表明,SFFS和TFS相比,写延迟降低了76.6%,读延迟降低了约10%。通过对系统结构的分析,相比于TFS,SFFS在中心节点的负载更轻,失效恢复更快,在可用性方面更有优势。     

基于可拆分倒排索引的可搜索加密方案

为快速检索云环境下的加密数据,提出了一种高效的适用于批量数据处理场合的可搜索加密方案。首先,由客户端创建两个倒排索引,分别是存储了文件-关键词映射的文件索引和用于存储关键词-文件映射的空的搜索索引;然后,将这两个索引提交给云服务器。搜索索引是在用户检索过程中由云端根据搜索凭证和文件索引逐渐更新建立的,记录了已被检索关键词的检索结果,该方法将搜索索引的构建时间有效分摊了到了每次检索过程中并节省了存储空间。索引采用基于key-value结构的集合存储方式,支持索引的同时合并和拆分,即在添加和删除文件时,由客户端根据要添加或删除的文件集生成对应的文件索引和搜索索引,然后服务器对索引进行合并和拆分,从而实现文件的快速批量添加和删除。测试结果表明,所提方案极大提高了文件更新的效率,适用于批量数据处理。通过泄露函数证明了该方案能满足自适应动态选择关键词攻击下的不可区分性安全标准。     

基于特征的江面轮船识别算法*

本文提出了一种江面轮船目标的特征识别方法。首先对江面轮船图像进行预处理,然后通过二维小波变换,提取出边缘轮廓,将目标物体与背景分离开来。结合本文提出的FE（Feature Extract）算法提取出轮船图像的四个特征,根据这些特征建造一个知识库,通过选取适当的知识,采用产生式规则对目标物体进行判别,排除干扰目标,从而识别出轮船目标。最后本文从图片库中抽取几张图片进行实验,相比于之前的单特征方法和Adboost方法可以看出,该方法在识别率上要高于单特征方法,在识别速度上要快于Adboost方法。     

一种二胡音乐的WAVE转MIDI的设计方法

文中的目的是介绍一种将WAVE格式的二胡音乐文件转换成MIDI格式的设计过程。具体方法如下：首先,利用小波变换和快速傅里叶变换联合提取出音符的基频;其次,根据MIDI音乐的消息格式,将得到的各基频转换成相应的MII音符;再次,根据二胡音乐的特性,设定双门限阈值筛选出有效的音符。最后加入其他控制信息,最终结果是组合成MI-DI的音轨文件。经实验得出结论,该方法针对二胡音乐的音高提取有较高的准确率,同时能够生成所需的有效MIDI文件。     

The uses and gratifications of using Facebook music listening applications

Despite rapid adoption of social media as a means of music listening, little is known about users’ motivations. This study applies the uses and gratifications approach to users’ motivations for using music listening applications on Facebook. Participants completed an online survey, and 153 out of 576 respondents indicated that they used a Facebook music listening application. A principal axis factor analysis identified three different motivations for this usage, namely entertainment, communication, and habitual diversion gratifications. The entertainment and communication gratifications replicate those found in prior uses and gratifications research concerning other social networking features, illustrating the strong similarity between uses of music and social media. However, the habitual diversion gratification may serve to distinguish listening applications from other features. Identifying and explaining these factors is relevant to social media users, musicians and application designers, as they explain what motivates a means of music listening that is gaining prominence.     

Technology scruples: why intimidation will not save the recording industry and how enchantment might

While the recording industry continues to lobby for increasingly draconian laws to protect their interests, users of digital technology continue to share files and copy protected music. This paper considers the ethics of copying and argues that legal measures are unlikely to solve the music industry’s problems in the age of digital reproduction. It begins with a review of the legal arguments around copyright legislation and notes that the law is currently unclear and contested. Adapting the game “scruples” to questions of what is and is not considered theft, a qualitative study reflects on the ways that ethical positions around new media are reached and articulated. The findings relate ethical positions constructed around notions of resistance, intangibility and identity. It is argued that the global online population cannot be policed without consent and that mechanics of artist reimbursement must be developed that account for consumers’ technology scruples. File sharing is then considered not as a legal problem but as a design challenge and a strategy of enchantment is suggested. The design concept of a digital music box is outlined to illustrate strategies of enchantment rather than litigation and intimidation.     

一种云存储服务客户端增量同步算法

随着云计算技术的飞速发展,越来越多的用户选择使用云存储服务来保存个人文件。云存储共享与协作技术允许用户之间共享云端文件,支持其他用户通过各种智能终端上的客户端对文件进行读写操作。云存储共享与协作技术带来了文件历史版本大规模共享场景的需求,这对云存储系统的并发I/O性能是极大的考验。针对云存储服务共享场景的特点,挖掘文件历史版本之间的关系,采用基于增量传输的优化技术来提升云存储系统的传输性能。在此基础上,优化算法中强弱校验过程的内存占用和磁盘读写,利用文件历史版本数据优化同步流程,有效的减少数据传输量,并且提高系统的存储性能,适用于带宽有限和网络不稳定和大规模共享同步等极端场景。     

A privacy-preserving multi-keyword ranked retrieval scheme in cloud computing

ABSTRACT

Big data and cloud computing could bring security problems. In order to ensure data security and user privacy, people would choose to store data in the cloud with ciphertext. How to search data efficiently and comprehensively without decryption has become the focus of this paper. In this paper, we propose an efficient privacy protection scheme. In this scheme, Elliptic Curve Cryptography (ECC) is adopted to encrypt the data. It can reduce the computing cost of encryption and decryption uploading the encrypted files and indexes to the cloud server. Then it can authorize users to generate trap door using hash conflict function, and send it to Cloud Service Provider (CSP) for searching for matched ciphertext. The CSP uses the Apriori algorithm to extend keywords and search index to match the ciphertext. In this paper, we will use the Apriori algorithm to extend the keywords’ semantics, match the index list based on these keywords, and return the requested file-set which is more consistent with the user’s search. Experiments show that compared with traditional methods, files can be encrypted, decrypted, and recovered more quickly when we use this method. It can also ensure the privacy of data and reduce the communication overhead.     

中文音乐情感词典构建及情感分类方法研究

作为仅次于及时通信和搜索引擎的中国互联网网民第三大应用,网络音乐及其应用技术受到业界学者的青睐。音乐作为人类最重要的交流媒介,携带着丰富的情感信息,计算机音乐情感分析更是得到人机情感交互技术领域的高度重视。在基于歌词文本的音乐情感分析过程中,一部合理的音乐领域情感词典,将提供更加细致、更加准确的分析结果。以改进后的Hevner情感环模型为基础,借助HowNet所提供的语义资源和从网络爬取的歌词文本语料库,构建了一部树形层次结构的音乐领域中文情感词典,并利用LRC歌词携带的时间标签获取歌曲的语速信息,实现了基于情感向量空间模型和情感词典的歌词情感分类。实验表明与人工构建的情感词典相比,所构建的情感词典更适用于音乐领域。     

Design and evaluation of mood pictures in social music discovery service

This paper describes the iterative design process and evaluation of mood pictures in a social music discovery service. The service enables users to consume and collaboratively create playlists based on the pictures. In total, 45 Finnish users took part in the qualitative evaluations. This paper presents the results regarding the preset mood picture design and introduces user-generated mood picture playlists. Based on the results, a set of design implications for mood pictures is introduced. In addition, the consistency of user responses from a quantitative picture–music association test shows the applicability of mood picture–music associations to a music discovery service.     

数字化民俗博物馆建设中数据存储方案研究

在数字博物馆的建设中,存在大量的图片、视频、声音、大文本或大二进制文件、3D图像、声音解说、电影录像等多媒体数据和大量的展品信息数据。为方便对这些数据进行存储、管理、检索等,并使用户方便快捷地浏览到博物馆中信息,本文对博物馆中大量的数据信息的存储方案进行了研究,并对不同存储方案进行了测试比较,根据数字博物馆中各种多媒体数据特点的不同,分别采用了不同的存储方案。     

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.     

A musical system for emotional expression

The automatic control of emotional expression in music is a challenge that is far from being solved. This paper describes research conducted with the aim of developing a system with such capabilities. The system works with standard MIDI files and develops in two stages: the first offline, the second online. In the first stage, MIDI files are partitioned in segments with uniform emotional content. These are subjected to a process of features extraction, then classified according to emotional values of valence and arousal and stored in a music base. In the second stage, segments are selected and transformed according to the desired emotion and then arranged in song-like structures.The system is using a knowledge base, grounded on empirical results of works of Music Psychology that was refined with data obtained with questionnaires; we also plan to use data obtained with other methods of emotional recognition in a near future. For the experimental setups, we prepared web-based questionnaires with musical segments of different emotional content. Each subject classified each segment after listening to it, with values for valence and arousal. The modularity, adaptability and flexibility of our system’s architecture make it applicable in various contexts like video-games, theater, films and healthcare contexts.     

嵌入式系统中PDF格式的高效压缩算法

提出的算法可以解决在嵌入式系统上显示含有文本和图像的PDF文件的要求。该算法先将PDF文件分解成图像和文本两部分,然后依照原文件的显示格式将分解出来的元素合成一种新的格式文件,最终实现将PC机上的PDF格式文件转化为适合在嵌入式系统上正确显示的格式。该算法缩减了文件的大小,使之适应于嵌入式系统有限的硬件资源,也扩大了嵌入式系统的文件来源。     

An investigation of the intention to share media files over peer-to-peer networks

File-sharing over peer-to-peer (P2P) networks once consisted largely of music files, which, when shared, were infringements of copyrights. For this reason, studies of the behavioural intentions (BIs) to share files over P2P networks have often focused on the piracy of music files. However, with improved technology and increased bandwidth, large files such as videos are routinely shared. As industry-led efforts may have had some success stemming illegal file-sharing, and as new and legitimate applications of P2P file-sharing are emerging, it is important to include media files of all types and consider file-sharing that is both legitimate and that which constitutes piracy within the scope of our study. To study the intention to share media files over P2P networks, we evaluate two alternative models. The theory of planned behaviour (TPB) is the underlying theory for both models, one of which is based on the original TPB and one on the decomposed TPB. We test both models using previously validated instruments and find that both models can explain a significant portion of the variance in the intentions to share media files over P2P networks. However, the model based on the decomposed TPB can explain more of the variance. A second advantage of this model is that it can be more readily translated to managerial actions, which are also explored.     

基于中文搜索引擎网络信息用户行为研究*

为了更好地理解中文搜索用户的检索行为,首先建立一个搜索引擎选择平台,主要是用来生成研究中所需的日志文件;然后从中英文用户的搜索行为差异的角度出发,对日志文件进行深入研究,包括各中文搜索引擎使用率比较以及中文用户输入查询行为的一些规律等。研究结果表明,对准确地评测搜索引擎检索的效果以及未来中文搜索引擎设计的改进都有较好的指导意义。     

StoryWizard: a framework for fast stylized story illustration

A fast stylized framework used for creating illustration of a given story is represented. The framework automatically searches proper online images according to the keywords of story, provides users some tools to make the search result precise, and helps users create a picture in every scene without considering the consistency of each character. With the friendly user interface, it provides user abundant interactions, which helps users express their ideas flexibly. Experimental results indicate that this framework allows users, without any art background, to produce personalized picture series with specified story. The fast process, effective interaction and generated delicate pictures can make a story more impressive.     

面向云环境的一致性可验证单关键词检索方法

在云环境资源外包服务模式下,数据拥有者不再参与对其外包数据的直接管理,这就使得验证数据使用者获得的检索结果是否满足一致性成为具有挑战性的问题。现有的研究工作重点聚焦于解决云服务提供商满足“诚实而好奇”模型假设下的隐私保护问题,但并不能解决恶意攻击威胁模型下的检索结果一致性验证问题。针对云服务提供商恶意攻击威胁模型,提出了一种面向云环境的基于偏序约束链的一致性可验证单关键词检索方法——IVSKS。数据拥有者根据文档与关键词的相关度的偏序关系,构造用于检索结果一致性验证的偏序约束链验证编码信息,并将该信息与文档集共同外包存储至云端;数据使用者在执行单关键词检索时,云端返回检索结果文档集以及相应的验证编码;最后,数据使用者根据获得的检索结果重构验证编码,实现针对检索结果的一致性验证。实验表明,与同类方法相比,IVSKS在检索结果冗余度以及一致性验证时间开销上具有更好的表现。     

Affective design of products using an audio-based protocol: Application to eyeglass frame

The measurement and understanding of user emotions elicited by product appearance are critical elements of the product development process. This paper proposes a new emotion measurement method, called Auditory Parameter Method. It is a non-verbal technique that uses auditory stimuli (music samples) and association tests for evaluating a set of products, given by their pictures. From user-tests, it provides an assessment of these products according to a series of emotional dimensions. We present the methodological framework used to build the links between user's emotional responses and geometrical features of the products. The method is described on an application case, an eyeglass frame. Analysis of Variance models are employed to examine how various shape factors influence users' emotional responses. To demonstrate the effectiveness of our protocol, we compare the proposed method with the conventional Semantic Differential using Principal Component Analysis and Generalized Procrustes Analysis. The new protocol demonstrates interesting qualities for collecting the intuitive emotions of users and for providing a discriminant measurement of emotions. It can also be used by designers to stimulate creativity.