Music recommendation and query-by-content using Self-Organizing Maps

Abstract

Abstract-The ever-increasing density of computer storage devices has allowed the average user to store enormous quantities of multimedia content, and a large amount of this content is usually music. Current search techniques for musical content rely on meta-data tags which describe artist, album, year, genre, etc. Query-by-content systems allow users to search based upon the acoustical content of the songs. Recent systems have mainly depended upon textual representations of the queries and targets in order to apply common string-matching algorithms. However, these methods lose much of the information content of the song and limit the ways in which a user may search. We have created a music recommendation system that uses Self-Organizing Maps to find similarities between songs while preserving more of the original acoustical content. We build on the design of the recommendation system to create a musical query-by-content system. We discuss the weaknesses of the naïve solution and then implement a quasi-supervised design and discuss some preliminary results.

A SOM-based multimodal system for musical query-by-content

“…Self-Organizing Maps (SOMs) produce an interpolated feature space and can easily be visualized, and, as a result, naturally suggest themselves as a way of doing this. In earlier work, we proposed the design of a system that uses a SOM to reduce the dimensionality Kyle of the feature space, allowing the use of simple (Euclidean) distance metrics [13]. Here we extend that work by fully implementing the proposed system and evaluating it for a variety of different query scenarios.…”

Syaritar intelligent system for the detection of illegal logging in the river basin Jeneberang

“…Cutting images or image cropping is a process to clean the sides of the image to fix orreframing a digital image or image [6]. These cuts are necessary to eliminate parts of the right side and bottom of a digital image that contains information derived from Google Earth.…”

“…The classification was intended to distinguish between areas of forest to non-forest area so there happen error identification results [6]. Program classifications retrieved from programs that have been created previously by Safaruddin A.…”

“…From 30 data tested, 24 of them is data logging takes place 6 remaining data liardan is not happening logging obtained from previous studies [6].…”

“…In addition, the RMSE values can also be used as a criterion for the epoch number system to stop learning. RMSE is the level of system output values error learning to value output data pairs of trainers or in other words the RMSE is the difference between the desired output and the actual output [6].…”