Difference between revisions of "2006:Symbolic Melodic Similarity Results"
(→Introduction) |
|||
| Line 2: | Line 2: | ||
==Introduction== | ==Introduction== | ||
These are the results for the 2006 running of the Symbolic Melodic Similarity task set. For background information about this task set please refer to the [[Symbolic Melodic Similarity]] page. | These are the results for the 2006 running of the Symbolic Melodic Similarity task set. For background information about this task set please refer to the [[Symbolic Melodic Similarity]] page. | ||
| − | Each system was given a query and returned the 10 most melodically similar songs from a given collection where the collections were | + | |
| + | Each system was given a query and returned the 10 most melodically similar songs from a given collection where the collections were RISM (monophonic; 10,000), Karoke (polyphonic; 1,000), Mixed (polyphonic; 15,741). Then, for each query, the returned results from all participants were grouped and were evaluated by human graders, each query being evaluated by 3 different graders with two scores (using the Evalutron 6000 system). Graders were asked to provide 1 categorical score with 3 categories: NS,SS,VS as explained below, and one fine score (in the range from 0 to 10). | ||
====Evalutron 6000 Summary Data==== | ====Evalutron 6000 Summary Data==== | ||
| − | '''Number of evaluators''' = | + | '''Number of evaluators''' = 20<br /> |
'''Number of evaluations per query/candidate pair''' = 3<br /> | '''Number of evaluations per query/candidate pair''' = 3<br /> | ||
'''Number of queries per grader''' = 15<br /> | '''Number of queries per grader''' = 15<br /> | ||
Revision as of 18:45, 3 October 2006
Contents
Introduction
These are the results for the 2006 running of the Symbolic Melodic Similarity task set. For background information about this task set please refer to the Symbolic Melodic Similarity page.
Each system was given a query and returned the 10 most melodically similar songs from a given collection where the collections were RISM (monophonic; 10,000), Karoke (polyphonic; 1,000), Mixed (polyphonic; 15,741). Then, for each query, the returned results from all participants were grouped and were evaluated by human graders, each query being evaluated by 3 different graders with two scores (using the Evalutron 6000 system). Graders were asked to provide 1 categorical score with 3 categories: NS,SS,VS as explained below, and one fine score (in the range from 0 to 10).
Evalutron 6000 Summary Data
Number of evaluators = 20
Number of evaluations per query/candidate pair = 3
Number of queries per grader = 15
Ave. size of the candidate lists = 15
Ave. number of query/candidate pairs evaluated per grader: 225
Number of queries (across all subtasks = 17
General Legend
Team ID
Prefix R = RISM collection, K = Karaoke collection, M = Polyphonic collection
FH = Pascal Ferraro and Pierre Hanna
NM = Kjell Lemström, Niko Mikkilä, Veli Mäkinen and Esko Ukkonen
RT = Rainer Typke, Frans Wiering and Remco C. Veltkamp
KF = Klaus Frieler
AU = Alexandra Uitdenbogerd
Broad Categories
NS = Not Similar
SS = Somewhat Similar
VS = Very Similar
Table Headings
ADR = Average Dynamic Recall
NRGB = Normalize Recall at Group Boundaries
AP = Average Precision (non-interpolated)
PND = Precision at N Documents
Calculating Summary Measures
Fine(1) = Sum of fine-grained human similarity decisions (0-10).
PSum(1) = Sum of human broad similarity decisions: NS=0, SS=1, VS=2.
WCsum(1) = 'World Cup' scoring: NS=0, SS=1, VS=3 (rewards Very Similar).
SDsum(1) = 'Stephen Downie' scoring: NS=0, SS=1, VS=4 (strongly rewards Very Similar).
Greater0(1) = NS=0, SS=1, VS=1 (binary relevance judgement).
Greater1(1) = NS=0, SS=0, VS=1 (binary relevance judgement using only Very Similar).
(1)Normalized to the range 0 to 1.
Overall Summary Results
Visualizations
Rainer Typke has created a series of Symbolic Melodic Similarity Graphs that help us visualize the results.
Rainer Typke has also created a set of detailed representations of the results that is definitely with exploring at [http://rainer.typke.org/mirex06.0.html].
Task I: RISM Overall Summary
file /nema-raid/www/mirex/results/sms06_rism_sum.csv not found
Task I: RISM Runtime Data
file /nema-raid/www/mirex/results/sms06_rism_runtime.csv not found
Task IIa: Karaoke Overall Summary
file /nema-raid/www/mirex/results/sms06_karaoke_sum.csv not found
Task IIa: Karaoke Runtime Data
file /nema-raid/www/mirex/results/sms06_karaoke_runtime.csv not found
Task IIb: Mixed Polyphonic Overall Summary
file /nema-raid/www/mirex/results/sms06_mixed_sum.csv not found
Task IIb: Mixed Polyphonic Runtime Data
file /nema-raid/www/mirex/results/sms06_mixed_runtime.csv not found
Task I: RISM Collection Summary Results
There is an error with this data set...please stand by. file /nema-raid/www/mirex/results/sms06_rism_results3.csv not found
Task IIa: Karaoke Collection Summary Results
file /nema-raid/www/mirex/results/sms06_kar_results3.csv not found
Task IIb: Mixed Polyphonic Collection Summary Results
file /nema-raid/www/mirex/results/sms06_mix_results3.csv not found
Raw Scores
The raw data derived from the Evalutron 6000 human evaluations are located on the Symbolic Melodic Similarity Raw Data page.
