Towards Improved Music Recommendation: Using Blogs and Micro-Blogs

2. Methodology

Our research starts with a review of the current literature. This review focuses on literature regarding recommendation engines, their approaches, evaluation processes and success, but also on blogs and micro-blogs.

From the literature review, important topics in music recommendation are selected. These topics are used to create a semi-structured interview for potential or current users of music-recommender systems. In addition, the topics help to define questions for several expert and user interviews. These interviews complemented the reviewed literature with personal insights.

The interviews are analyzed and combined with the literature. This analysis serves as input for the prototype requirements, in which technical and time-restriction are taken into account. Subsequently, a prototype that implements the requirements is designed and developed.

After developing the prototype, the user-evaluation is performed. A questionnaire is combined with the prototype to evaluate its quality. Hereby, the perceived quality of recommendations is measured. These results are used to draw a conclusion, propose improvements to the prototype and discuss areas for further research. Figure 1 gives a visualization of the methodology of this study.

Figure 1.

A visualization of the methodology

3. Related Work

This section presents a literature review on relevant research in Music Recommendation, music sharing through blogs and micro-blogs, improvements of perceived quality, combination of information and evaluation of recommendation systems.