Tagging Artists using Co-Occurrences on the Web
نویسندگان
چکیده
We present an efficient unsupervised approach in finding subjective artist meta-data on the world wide web. Since we are interested in the collective knowledge on artists as available on the web, our method is based on the extraction of information from multiple web pages. We use co-occurrences of pairs of artists on the web to identify similarity between artists. To determine the applicability of tags to artists we follow the same approach. We use Google to find the co-occurrences on the web, either by analyzing Google excerpts found by querying patterns or by scanning full documents. Since the same tags are often applicable to related artists, we use similarity between artists to improve the tagging. We tested and compared the two co-occurrence extraction methods on two different domains: finding the most appropriate genres for music artists, and finding art-styles for painters. The results are convincing and show that the use of similar artists indeed improves the precision of the tagging.
منابع مشابه
Web-Based Artist Categorization
We present a novel approach in categorizing artists into subjective categories such as genre. We base our method on co-occurrences on the web, found with the Google search engine. A direct mapping between artists and categories proved to be unreliable. We use the categories mapped to closely related artists to obtain a more reliable mapping. The method is tested on a genre classification test s...
متن کاملA Web-based Approach to Assessing Artist Similarity Using Co-occurrences
In this paper, we present a similarity measure for music artists based on search results of Google queries. Co-occurrences of artist names on web pages are analyzed to measure how often two artists are mentioned together on the same web page. We estimate conditional probabilities using the extracted page count. These conditional probabilities give a similarity measure which is evaluated using a...
متن کاملInstance Classification using Co-Occurrences on the Web
We present a novel unsupervised approach to mapping artrelated instances (such as music artists and painters) to subjective categories like genre and style. We base our approach on co-occurrences of the two on the web, found with Google. The co-occurrences are found using three methods: by identifying the search engine counts, by analyzing Google excerpts found by querying patterns and by scann...
متن کاملUncovering Affinity of Artists to Multiple Genres from Social Behaviour Data
In organisation schemes, musical artists are commonly identified with a unique ‘genre’ label attached, even when they have affinity to multiple genres. To uncover this hidden cultural awareness about multi-genre affinity, we present a new model based on the analysis of the way in which a community of users organise artists and genres in playlists. Our work is based on a novel dataset that we ha...
متن کاملInvestigating Different Term Weighting Functions for Browsing Artist-Related Web Pages by Means of Term Co-Occurrences
We present a user interface (UI) for browsing collections of web pages about music artists. Given such a collection, we use a term list to index its contents and to derive term co-occurrences. Based on these co-occurrences, we create a UI that employs a variant of the Sunburst visualization technique. The UI is embedded in CoMIRVA, our framework for music information retrieval and visualization...
متن کامل