• Annual Conference of KIPS
• /
• 2013.05a
• /
• pp.204-207
• /
• 2013

Music genre classification is an essential component for music information retrieval system. There are two important components to be considered for better genre classification, which are audio feature extraction and classifier. This paper incorporates two different kinds of features for genre classification, timbral texture and rhythmic content features. Timbral texture contains several spectral and Mel-frequency Cepstral Coefficient (MFCC) features. Before choosing a timbral feature we explore which feature contributes less significant role on genre discrimination. This facilitates the reduction of feature dimension. For the timbral features up to the 4-th order central moments and the covariance components of mutual features are considered to improve the overall classification result. For the rhythmic content the features extracted from beat histogram are selected. In the paper Extreme Learning Machine (ELM) with bagging is used as classifier for classifying the genres. Based on the proposed feature sets and classifier, experiment is performed with well-known datasets: GTZAN databases with ten different music genres, respectively. The proposed method acquires the better classification accuracy than the existing approaches.

• Journal of Information Processing Systems
• /
• v.20 no.1
• /
• pp.13-23
• /
• 2024

Music brings pleasure and relaxation to people. Therefore, it is necessary to classify musical genres based on scenes. Identifying favorite musical genres from massive music data is a time-consuming and laborious task. Recent studies have suggested that machine learning algorithms are effective in distinguishing between various musical genres. However, meeting the actual requirements in terms of accuracy or timeliness is challenging. In this study, a hybrid machine learning model that combines a deep residual auto-encoder (DRAE) and support vector machine (SVM) for musical genre recognition was proposed. Eight manually extracted features from the Mel-frequency cepstral coefficients (MFCC) were employed in the preprocessing stage as the hybrid music data source. During the training stage, DRAE was employed to extract feature maps, which were then used as input for the SVM classifier. The experimental results indicated that this method achieved a 91.54% F1-score and 91.58% top-1 accuracy, outperforming existing approaches. This novel approach leverages deep architecture and conventional machine learning algorithms and provides a new horizon for musical genre classification tasks.

• Journal of Music and Human Behavior
• /
• v.15 no.1
• /
• pp.51-68
• /
• 2018

The purpose of this study was to examine the use of music by international college students in Korea. A questionnaire was administered to foreign college students. The questionnaire was composed of 25 items regarding use of music, the purpose of music use, and demographic information. A total of 150 questionnaires were distributed, and 81 questionnaires were returned. Incomplete questionnaires were excluded, and the remaining 69 questionnaires were analyzed. The results of this study were as follows. First, international students in Korea preferred listening to music over playing instruments or singing and mostly listened to popular music in their dormitory alone. They mostly listened to popular songs with love related themes and preferred music from their own country over Korean music. Second, in terms of the purpose behind music use, comfort was the reason reported most frequently, followed by mood change, enjoyment, sense of belonging, and sense of achievement. Third, there were no significant differences in use of music depending on individual factors (e.g., gender, length of residence in Korea, length of previous music education), but significant differences were found for using music for the purpose of mood change and sense of belonging based on respondents' length of residence in Korea. The results of this study may contribute to the development of musical programs for cultural adaptation and psycho-emotional support for international students in Korea.

• Journal of Music and Human Behavior
• /
• v.11 no.2
• /
• pp.59-80
• /
• 2014

Helen Lindquist Bonny developed the Bonny Method of Guided Imagery and Music (BMGIM), a music oriented self-exploration of consciousness, with the influence of humanistic and client-centered approaches. BMGIM can help people manage pain, anxiety, relationship issues, depression, and other conditions through self-awareness and self-potential. The purpose of this biographical study was to explore Bonny's early life experiences and their contribution to the humanistic and transpersonal approach to music therapy. The study was organized in chronological order: Bonny's childhood and young adulthood; inspirations that led to development of BMGIM; and research and clinical events that helped solidify the development of BMGIM. An interview with Bonny and a review of the literature supported the significance of these periods. The years between 1972 and 1979 mark the development of the first series of BMGIM music programs, which Bonny said served as the base for many later GIM music programs developed. Bonny's contribution to the field of music therapy includes the provision of strong foundation for music therapy based on Maslow's and Rogerian approach, and the introduction of time-valued music to the music therapy field.

• Journal of Music and Human Behavior
• /
• v.16 no.1
• /
• pp.89-117
• /
• 2019

The purpose of this study was to explore and better understand special education music teachers' perceptions of their music curriculum. For this purpose, we conducted a focus group interview with seven special education music teachers. During the interview, four major themes and 14 sub-themes were identified. The main themes were the following: (a) types of applied activities and the goals of music classes (e.g., activities to be applied in various ways depending on the characteristics of the disability and intended outcome), (b) difficulty in implementing the music curriculum (e.g., lack of fit between textbook and students' chronological ages, lack of time and focus, self-evaluation of performance as a music teacher, (c) therapeutic experiences during music classes (e.g., expectation for positive effects through music therapy, joy of witnessing changes in students, and sense of togetherness), and (d) obstacles to the therapeutic approach of music classes and need for support (e.g., lack of professional knowledge regarding therapeutic approaches and problems with administrators and school environment). Based on these results, problems in implementing the music therapy approach as part of the music curriculum in special education schools are discussed and practical solutions for educators are offered.

• Journal of Information Processing Systems
• /
• v.13 no.2
• /
• pp.400-416
• /
• 2017

Music emotion is an important component in the field of music information retrieval and computational musicology. This paper proposes an approach for automatic emotion classification, based on rough set (RS) theory. In the proposed approach, four different sets of music features are extracted, representing dynamics, rhythm, spectral, and harmony. From the features, five different statistical parameters are considered as attributes, including up to the $4^{th}$ order central moments of each feature, and covariance components of mutual ones. The large number of attributes is controlled by RS-based approach, in which superfluous features are removed, to obtain indispensable ones. In addition, RS-based approach makes it possible to visualize which attributes play a significant role in the generated rules, and also determine the strength of each rule for classification. The experiments have been performed to find out which audio features and which of the different statistical parameters derived from them are important for emotion classification. Also, the resulting indispensable attributes and the usefulness of covariance components have been discussed. The overall classification accuracy with all statistical parameters has recorded comparatively better than currently existing methods on a pair of datasets.

• Journal of Intelligence and Information Systems
• /
• v.12 no.3
• /
• pp.111-126
• /
• 2006

The context-awareness is one of the core technologies in ubiquitous computing environment. In this research, we incorporated the capability of context-awareness in a case-based music recommendation system. Our proposed system consists of Intention Module and Recommendation Module. The Intention Module infers whether a user wants to listen to the music or not from the environmental context information. Then, the Recommendation Module selects songs from the songs that are listened by similar users in similar context, and recommends them to the user. The results showed that our proposed system outperformed the traditional case-based music recommendation system in accuracy by about 9% point.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.6
• /
• pp.1302-1309
• /
• 2014

In this paper, we propose a method to improve the performance of the direction-of-arrival (DOA) estimation of a speech source using a multiple signal classification (MUSIC)-based algorithm. Basically, the proposed algorithm utilizes a complex coefficient band pass filter to generate the narrow band signals for signal analysis. Also, reverberation component reduction and quadratic function-based response approximation in MUSIC spatial spectrum are utilized to improve the accuracy of DOA estimation. Experimental results show that the proposed method outperforms the well-known generalized cross-correlation (GCC)-based DOA estimation algorithm in the aspect of the estimation error and success rate, respectively.Abstract should be placed here. These instructions give you guidelines for preparing papers for JICCE.

• Journal of Information Technology Services
• /
• v.13 no.1
• /
• pp.161-179
• /
• 2014

Added value of music industry in Korea is not distributed and calculated properly, and this obstacle brings various problems in a creative environment. Meanwhile, a new business model such as Open Collaboration, Crowdsourcing and platform that makes decisions and innovation from external resources has been appeared in commercial area. This new model like a composer delivers to consumers directly through Youtube.com, and multi collaboration is applied to the music industry, and it enables a new type of mechanism for creation, distribution, division, and calculation of music. However there are not enough empirical study of the music market because existing relative researches has been centered around fundamental concepts and application methodologies. This research defines Open Collaboration Platform in the music industry, and studies affecting factors of Participation Intention for example Justice, Information System Quality and Perceived Value. For a survey we apply PLS(Partial Least Square) to analyse Equity, Information System Quality and structural equation between Perceived Value and Participation Intention. Analysis results show Distributive Justice and Procedural Justice affects Platform Trust, and Service Quality, Economical Value and Emotional Value affects Platform Usefulness. Also Platform Trust and Platform Usefulness affects Platform Participation Intention. We discussed academic and practical implication based on research results.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.712-716
• /
• 2008

In this paper, we introduce a way to generate and control the serial music using stochastical tools. A serialism, which is one of composition techniques in the 20th century modern music, is composed using uniformly distributed notes or durations. To conserve this property, we design the optimization process to generate the random numbers which can be used to compose the serial music. The optimization are designed using chi-square test and auto-correlation test. User can also apply the additional constraint to the objective function for controlling the serial music. Using the our method, we can compose and control the traditional serial music automatically.