Authors: Abdul Shahid, Danny Diamond, James McDermott
Part of: fonn
Source: SPARQL endpoint
Export
{'sparql_endpoint': 'https://polifonia.disi.unibo.it/fonn/sparql', 'template_mode': 'statistics', 'title': 'Statistics on The Session (annotated subset) and Meertens Tune Collections (MTC-ANN) Pattern KG', 'color_code': ['#39c6b4', '#17cf31'], 'section_name': 'fonn', 'subtitle': '', 'curator': 'Abdul Shahid, Danny Diamond, James McDermott', 'description': 'MTC-ANN is a small, carefully selected corpus of tunes from the Meertens Tune Collections of Dutch folk music, with some annotations. We are using this as a demonstration of our pattern methods and pattern knowledge graph. We are also using a subset of The Session. The Session is 40k tunes, crowd-sourced, from the Irish tradition. Our annotated subset contains 315 tunes in 10 tune families.', 'dynamic_elements': [{'position': 0, 'operations': [], 'type': 'text', 'text': "<p>Next, let's look at the locations where patterns occur in the tune. This is measured in beats. Tunes are usually less than 100 beats in duration. This is shown on the x-axis. Let's compare this to pattern complexity. We have a simple measure of complexity here, where complexity is the percentage of unique notes in the pattern.</p>", 'extra_queries': [], 'map_filters': []}, {'position': 1, 'operations': [], 'type': 'chart', 'chart_type': 'scatterplot', 'chart_title': 'Pattern complexity versus location', 'chart_query': 'PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>\r\nPREFIX xsd: <http://www.w3.org/2001/XMLSchema#>\r\nprefix jams: <http://w3id.org/polifonia/ontology/jams/> \r\nprefix mc: <http://w3id.org/polifonia/ontology/musical-composition/>\r\nSELECT (xsd:float(?loc) AS ?x) (xsd:integer(?comp * 100) as ?y)\r\nWHERE {\r\n?ja rdf:type jams:JAMSAnnotation;\r\n jams:includesObservation ?ob.\r\n?ob jams:hasPatternLocation ?loc; \r\n jams:hasPatternComplexity ?comp\r\n} ORDER BY ?x', 'chart_series': 'Pattern occurrences', 'chart_legend': {'x': 'Location (beats)', 'y': 'Complexity (%)'}, 'extra_queries': [], 'map_filters': []}, {'position': 2, 'operations': [], 'type': 'table', 'table_title': 'Most frequent patterns', 'table_query': 'PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>\r\nPREFIX jams: <http://w3id.org/polifonia/ontology/jams/>\r\n\r\nSELECT ?pattern (COUNT(?pattern) AS ?occurrences)\r\nWHERE {\r\n ?observation rdf:type jams:JAMSObservation ;\r\n jams:ofPattern ?pattern .\r\n}\r\nGROUP BY ?pattern\r\nORDER BY DESC(?occurrences)\r\nLIMIT 10\r\n', 'extra_queries': [], 'map_filters': []}, {'position': 3, 'operations': [], 'type': 'text', 'text': '<p>Next, let\'s look at the most frequent patterns in the corpus. We\'ll see that the most frequent ones have a "stop-word" quality, with many repeated notes. And indeed, in the scatterplot below, we see that all of those with more than 400 occurrences have very low pattern complexity.</p>', 'extra_queries': [], 'map_filters': []}, {'position': 4, 'operations': [], 'type': 'chart', 'chart_type': 'scatterplot', 'chart_title': 'Pattern complexity versus frequency', 'chart_query': 'PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>\r\nPREFIX xsd: <http://www.w3.org/2001/XMLSchema#>\r\nprefix jams: <http://w3id.org/polifonia/ontology/jams/> \r\nprefix mc: <http://w3id.org/polifonia/ontology/musical-composition/>\r\nSELECT (COUNT(?pattern) AS ?x) (xsd:integer(?comp * 100) as ?y)\r\nWHERE {\r\n?observation rdf:type jams:JAMSObservation ;\r\n jams:ofPattern ?pattern .\r\n?observation jams:hasPatternComplexity ?comp\r\n} GROUP BY ?pattern ?comp', 'chart_series': '', 'chart_legend': {'x': 'Occurrences', 'y': 'Complexity'}, 'extra_queries': [], 'map_filters': []}, {'position': 5, 'operations': [], 'type': 'count', 'count_query': 'prefix jams: <http://w3id.org/polifonia/ontology/jams/> \r\nSELECT (COUNT(DISTINCT ?tuneID) AS ?count) \r\nWHERE {\r\n ?comp jams:tuneId ?tuneID.\r\n}\r\n', 'count_label': 'Number of tunes', 'extra_queries': [], 'map_filters': []}, {'position': 6, 'operations': [], 'type': 'count', 'count_query': 'prefix jams: <http://w3id.org/polifonia/ontology/jams/> \r\nSELECT (COUNT(DISTINCT ?pattern) as ?count) \r\nWHERE {\r\n ?observation jams:ofPattern ?pattern .\r\n}\r\n', 'count_label': 'Number of distinct patterns', 'extra_queries': [], 'map_filters': []}, {'position': 7, 'operations': [], 'type': 'count', 'count_query': 'prefix jams: <http://w3id.org/polifonia/ontology/jams/> \r\nSELECT (COUNT(?pattern) as ?count) \r\nWHERE {\r\n ?observation jams:ofPattern ?pattern .\r\n}\r\n', 'count_label': 'Number of pattern occurrences (non-distinct)', 'extra_queries': [], 'map_filters': []}]}
MTC-ANN is a small, carefully selected corpus of tunes from the Meertens Tune Collections of Dutch folk music, with some annotations. We are using this as a demonstration of our pattern methods and pattern knowledge graph. We are also using a subset of The Session. The Session is 40k tunes, crowd-sourced, from the Irish tradition. Our annotated subset contains 315 tunes in 10 tune families.