Difference between revisions of "2007:Audio Music Similarity and Retrieval Results"

From MIREX Wiki
(→‎Introduction)
(→‎Team ID)
 
(42 intermediate revisions by 9 users not shown)
Line 1: Line 1:
−
TEMPLATE: NEEDS TO BE UPDATED
 
−
 
 
[[Category: Results]]
 
[[Category: Results]]
 
==Introduction==
 
==Introduction==
−
These are the results for the 2007 running of the Audio Music Similarity and Retrieval task set. For background information about this task set please refer to the [[Audio Music Similarity and Retrieval]] page.
+
These are the results for the 2007 running of the Audio Music Similarity and Retrieval task set. For background information about this task set please refer to the [[2007:Audio Music Similarity and Retrieval]] page.
  
−
Each system was given XXXX songs chosen from "uspop", "uscrap" and "cover song" collections. Each system then returned a 5000x5000 distance matrix. 100 songs were randomly selected as queries and the first 5 most highly ranked songs out of the 5000 were extracted for each query (after filtering out the query itself, returned results from the same artist and members of the cover song collection). Then, for each query, the returned results (candidates) from all participants were grouped and were evaluated by human graders using the Evalutron 6000 grading system. Each individual query/candidate set was evaluated by a single grader. For each query/candidate pair, graders provided two scores. 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).  A description and analysis is provided below.
+
Each system was given 7000 songs chosen from IMIRSEL's "uspop", "uscrap" and "american" "classical" and "sundry" collections. Each system then returned a 7000x7000 distance matrix. 100 songs were randomly selected from the 10 genre groups (10 per genre) as queries and the first 5 most highly ranked songs out of the 7000 were extracted for each query (after filtering out the query itself, returned results from the same artist were also omitted). Then, for each query, the returned results (candidates) from all participants were grouped and were evaluated by human graders using the Evalutron 6000 grading system. Each individual query/candidate set was evaluated by a single grader. For each query/candidate pair, graders provided two scores. 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).  A description and analysis is provided below.
 +
 
 +
The systems read in 30 second audio clips as their raw data. The same 30 second clips were used in the grading stage.
  
 
===Summary Data on Human Evaluations (Evalutron 6000)===
 
===Summary Data on Human Evaluations (Evalutron 6000)===
 
'''Number of evaluators''' = 20<br />  
 
'''Number of evaluators''' = 20<br />  
−
'''Number of evaluation per query/candidate pair''' = 1<br />
+
'''Number of evaluations per query/candidate pair''' = 1<br />
 
'''Number of queries per grader''' = 5 <br />
 
'''Number of queries per grader''' = 5 <br />
−
'''Size of the candidate lists''' = Maximum 30 (with no overlap)<br />  
+
'''Size of the candidate lists''' = 48.32<br />  
 
'''Number of randomly selected queries''' = 100 <br />
 
'''Number of randomly selected queries''' = 100 <br />
 +
'''Number of query/candidate pairs graded''' = 4832
 
   
 
   
 
====General Legend====
 
====General Legend====
 
=====Team ID=====
 
=====Team ID=====
−
'''EP''' = [https://www.music-ir.org/evaluation/MIREX/2006_abstracts/AS_pampalk.pdf Elias Pampalk]<br />
+
'''BK1''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_bosteels.pdf Klaas Bosteels, Etienne E. Kerre 1]<br />
−
'''TP''' = [https://www.music-ir.org/evaluation/MIREX/2006_abstracts/AS_pohle.pdf Tim Pohle]<br />
+
'''BK2''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_bosteels.pdf Klaas Bosteels, Etienne E. Kerre 2]<br />
−
'''VS''' = Vitor Soares<br />  
+
'''CB1''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_bastuck.pdf Christoph Bastuck 1]<br />
−
'''LR''' = [https://www.music-ir.org/evaluation/MIREX/2006_abstracts/AS_lidy.pdf Thomas Lidy and Andreas Rauber]<br />
+
'''CB2''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_bastuck.pdf Christoph Bastuck 2]<br />
−
'''KWT''' = Kris West (Trans)<br />
+
'''CB3''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_bastuck.pdf Christoph Bastuck 3]<br />
−
'''KWL''' = Kris West (Likely)<br />
+
'''GT''' = [https://www.music-ir.org/mirex/abstracts/2007/AI_CC_GC_MC_AS_tzanetakis.pdf George Tzanetakis] <br />
 +
'''LB''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_barrington.pdf Luke Barrington, Douglas Turnbull, David Torres, Gert Lanskriet]<br />
 +
'''ME''' = [https://www.music-ir.org/mirex/abstracts/2007/AI_CC_GC_MC_AS_mandel.pdf Michael I. Mandel, Daniel P. W. Ellis]<br />
 +
'''PC''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_paradzinets.pdf Aliaksandr Paradzinets, Liming Chen]<br />
 +
'''PS''' = [https://www.music-ir.org/mirex/abstracts/2007/AS_pohle.pdf Tim Pohle, Dominik Schnitzer]<br />
 +
'''TL1''' = [https://www.music-ir.org/mirex/abstracts/2007/AI_CC_GC_MC_AS_lidy.pdf Thomas Lidy, Andreas Rauber, Antonio Pertusa, José Manuel Iñesta 1]<br />
 +
'''TL2''' = [https://www.music-ir.org/mirex/abstracts/2007/AI_CC_GC_MC_AS_lidy.pdf Thomas Lidy, Andreas Rauber, Antonio Pertusa, José Manuel Iñesta 2]<br />
  
−
=====Broad Categories====
+
====Broad Categories====
 
'''NS''' = Not Similar<br />
 
'''NS''' = Not Similar<br />
 
'''SS''' = Somewhat Similar<br />
 
'''SS''' = Somewhat Similar<br />
Line 39: Line 46:
  
 
===Overall Summary Results===
 
===Overall Summary Results===
 +
'''NB''': The results for BK2 were interpolated from partial data due to a runtime error.
  
−
<csv>mirex06_as_overalllist.csv</csv>
+
<csv>2007/ams07_overall_summary2.csv</csv>
−
 
 
−
http://staff.aist.go.jp/elias.pampalk/papers/mirex06/friedman.png
 
−
 
 
−
This figure shows the official ranking of the submissions computed using a Friedman test. The blue lines indicate significance boundaries at the p=0.05 level. As can be seen, the differences are not significant. For a more detailed description and discussion see [http://staff.aist.go.jp/elias.pampalk/papers/pam_mirex06.pdf].
 
  
−
====Audio Music Similarity and Retrieval Runtime Data====
 
  
−
<csv>as06_runtime.csv</csv>
 
−
 
−
For a description of the computers the submission ran on see [[MIREX_2006_Equipment]].
 
  
 
===Friedman Test with Multiple Comparisons Results (p=0.05)===
 
===Friedman Test with Multiple Comparisons Results (p=0.05)===
−
The Friedman test was run in MATLAB against the Fine summary data over the 60 queries.<br />
+
The Friedman test was run in MATLAB against the Fine summary data over the 100 queries.<br />
 
Command: [c,m,h,gnames] = multcompare(stats, 'ctype', 'tukey-kramer','estimate', 'friedman', 'alpha', 0.05);
 
Command: [c,m,h,gnames] = multcompare(stats, 'ctype', 'tukey-kramer','estimate', 'friedman', 'alpha', 0.05);
−
<csv>AV_sum_friedman.csv</csv>
+
<csv>2007/ams07_sum_friedman_fine.csv</csv>
−
<csv>AV_fine_result.csv</csv>
+
<csv>2007/ams07_detail_friedman_fine.csv</csv>
 +
 
 +
[[Image:2007 ams broad scores friedmans.png]]
  
 
===Summary Results by Query===
 
===Summary Results by Query===
−
<csv>mirex06_as_uberlist.csv</csv>
+
These are the mean FINE scores per query assigned by Evalutron graders. The FINE scores for the 5 candidates returned per algorithm, per query, have been averaged. Values are bounded between 0.0 and 10.0. A perfect score would be 10. Genre labels have been included for reference.  
  
−
===Raw Scores===
+
<csv>2007/ams07_fine_by_query_with_genre.csv</csv>
−
The raw data derived from the Evalutron 6000 human evaluations are located on the [[Audio Music Similarity and Retrieval Raw Data]] page.
 
  
−
===Query Meta Data===
+
These are the mean BROAD scores per query assigned by Evalutron graders. The BROAD scores for the 5 candidates returned per algorithm, per query, have been averaged. Values are bounded between 0 (not similar) and 2 (very similar). A perfect score would be 2. Genre labels have been included for reference.
−
<csv>as06_queries.csv</csv>
+
<csv>2007/ams07_broad_by_query_with_genre.csv</csv>
  
−
==Results from Automatic Evaluation==
+
===Anonymized Metadata===
 +
[https://www.music-ir.org/mirex/results/2007/anonymizedAudioSim07metaData.csv Anonymized Metadata]<br />
  
−
<csv>as06_nonhuman_results.csv</csv>
+
===Raw Scores===
−
 
+
The raw data derived from the Evalutron 6000 human evaluations are located on the [[2007:Audio Music Similarity and Retrieval Raw Data]] page.
−
 
 
−
=== Other Results from Automatic Evaluation===
 
−
See [[Audio Music Similarity and Retrieval Other Automatic Evaluation Results]] page.
 
−
 
 
−
 
 
−
=== Introduction to automatic evaluation ===
 
−
Automated evaluation of music similarity techniques based on a metadata catalogue has several advantages:
 
−
* It does not require costly human ΓÇÿgradersΓÇÖ
 
−
* Allows testing of incremental changes in indexing algorithms
 
−
* Can achieve complete coverage over the test collection
 
−
* Provides a target for machine-learning, feature-selection and optimisation experiments
 
−
* Can predict the visualisation performance of an indexing technique
 
−
* Can identify indexing ΓÇÿanomoliesΓÇÖ in the indices tested
 
−
 
 
−
Automated ΓÇÿpseudo-objectiveΓÇÖ evaluation of music similarity estimation techniques was introduced by Logan & Saloman [1] and were shown to be highly correlated with careful human-based evaluations by Pampalk [2]. The results of this contest support the conclusions of Pampalk [2] although further work is required to fully understand the evaluation statistics.
 
−
 
 
−
 
 
−
=== Description of evaluation statistics ===
 
−
The evaluation statistics
 
−
 
 
−
;Neighbourhood clustering (artist, genre, album)
 
−
:average % of the top N results for each query in the collection with the same same label
 
−
 
 
−
;Artist-filtered genre neighbourhood
 
−
:average % of the top N results for each query belonging to the same genre label, ignoring matches from the same artist (ensures that results reflect musical not audio similarity)
 
−
 
 
−
;Mean Artist-filtered genre neighbourhood
 
−
:normalised form of the above statistic equally weighting each genre, penalising lop-sided performance.
 
−
 
 
−
;Normalised average distance between examples
 
−
:average distance between examples with the same label, indicates degree of clustering and potential for visual organisation of a collection
 
−
 
 
−
;Always similar (hubs)
 
−
:largest # of times an example appears in top N results for other queries, a result that appears too often will adversely affect performance without affecting other statistics
 
−
 
 
−
;Never similar (orphans)
 
−
:% of examples that never appear in a top N result list and cannot be retrieved by search
 
−
 
 
−
;Triangular inequality (metric space)
 
−
:indicates whether the function produces a metric distance space and therefore what visualisation techniques may be applied to it
 
−
 
 
−
 
 
−
=== Normalisation ===
 
−
Each of the neighbourhood statistics, described above, has been normalised by the number of examples of each class (a genre, album or artist) that is available in the test database. E.g. if the collection contained 20 tracks by a particular artist and a particular system retrieved 10 of those examples in its top 50 results it would normally achieve an artist neighbourhood score of 20%, while the normalised form of the metric would report a score of 50% (of the available matches were retrieved). Such normalisation is intended to avoid bias introduced into the results by skewed distribution of the examples according to each label set.
 
−
 
 
−
The mean artist-filtered genre neighbourhood is a normalised form of the artist-filter genre neighbourhood metric, which gives equal weight to performance of a system on each genre class. This version of the statistic is intended to match the prior probabilities or distribtuion of examples according to genre labels used as queries in the human listening test (where an equal number of examples from each class was selected - stratified random sampling) instead of the prior probabilities or distribution of examples appearing in the database.
 
−
 
 
−
 
 
−
=== Music-similarity evaluation issues ===
 
−
Care must be taken with all evaluations of audio musical similarity estimation techniques as there is a great potential for over-fitting in these experiments and for over-optimistic estimates of the performance of a system on novel test data to be produced.
 
−
 
 
−
The metadata catalog used to conduct automated evaluations should be as accurate as possible. However, this technique seems relatively robust to a degree of noise in the catalogue, parhaps due to its coarse granularity.
 
−
 
 
−
Small test collections do not allow us to accurately predict performance on larger test collections, for example:
 
−
* Indexing anomalies (ΓÇÿhubsΓÇÖ and ΓÇÿorphansΓÇÖ) cannot yet be understood.
 
−
** a single ΓÇÿhubΓÇÖ was found in the results of one system
 
−
*** appeared in nearly 2/5 of result lists
 
−
*** removing this one example from the collection of 5000 tracks makes it appear that the system does not suffer from indexing anomolies.
 
−
** What will be the number and coverage of ΓÇÿhubsΓÇÖ in a 100,000 song DB?
 
−
[[User:Kriswest|Kriswest]]
 
−
 
 
−
 
 
−
=== Directions for further work on evaluating audio music similarity ===
 
−
* Establish whether stratified sampling used in the human evals is optimal for producing results that reflect human perception of the quality of music indexes or whether the database should be sampled randomly.
 
−
** will influence selection of a statistic for use in automated evaluations or optimisation exps (artist-filtered genre or the mean artist-filtered genre).
 
−
* Explain the indexing anomalies in some techniques.
 
−
* Determine a safe minimum size for a test collection to be used to predict performance on an ΓÇÿindustrial-sizedΓÇÖ collection
 
−
* Establish optimal granularity or range of granularities for a genre catalogue to be used in this type of evaluation (8, 32 or 256 classes?) and integrate a confusion-cost matrix to reduce the penalisation of confusion between similar genres of music (e.g Punk and Heavy Metal) relative to confusion between highly dissimilar genres (e. Classical and Heavy Metal).
 
−
[[User:Kriswest|Kriswest]]
 
−
 
 
−
 
 
−
=== Evaluation Tools in Music-2-Knowledge (M2K) ===
 
−
The tools used to produce the evaluation statistics for MIREX 2006 will be released as part of M2K 1.2 (forthcoming). These tools provide services to:
 
−
* import collection metadata and distance matrices
 
−
* generate a stratified query set
 
−
* extract artist-filtered results (for use in human evaluation exps)
 
−
* calculate any of the evaluation statistics described above.
 
−
 
 
−
These tools may be used on the command-line by implementing the MIREX distance matrix file format, with M2K in the Data-2-Knowledge toolkit (D2K) or integrated into existing Java code with the new M2K API.
 
−
 
 
−
To obtain a copy of the evaluation tools prior to the M2K 1.2 release, contact [mailto:kw@cmp.uea.ac.uk Kris West]
 
−
 
 
−
 
 
−
=== Comments ===
 
−
The evaluation statistics for the MIREX 2006 Audio music similarity contest seem to support the contention that genre, artist and artist-filtered genre neighbourhood statistics are correlated with the human perception of the performance of music similarity estimators as they all reproduce the ranking produced by the human evaluation. However, the differences between systems in that evaluation are not statistically significant, so no firm conclusion can be made. Average distance statistics produce a different ranking but are intended to correlate with visualisation performance and not search.
 
−
[[User:Kriswest|Kriswest]]
 
−
 
 
−
 
 
−
=== A statistic for evaluation and use in selection & optimization experiments ===
 
−
As each statitic was found to be correlated with the results of the listening test, any *may* be used  to evaluate performance and to guide model optimisation or feature selection/weighting experiments. However, unfiltered genre and artist identification statistics are known to allow overfitting to produce over-optimistic performance estimates. In a model optimisation or feature selection experiment these statistics will be more likely to indicate '''Audio-similarity''' performance rather than actual '''Music-similarity''' performance and may lead to the selection of sub-optimal features or models. The artist-filtered genre neighbourhood can be used to avoid this effect.
 
−
 
 
−
The results from MIREX 2006 do not show a significant drop in performance using the artist-filtered genre statistic as would normally be expected. This may be due to the excessively skewed distribution of examples in the database (roughly 50% of examples are labelled as Rock/Pop, while a further 25% are Rap & Hip-Hop). Hence, the difference between the results produced and the random baseline are not well emphasized. Normalising this statistic by the prior probabilities of examples in the database (taking the mean of the diagonal of the artist-filtered genre confusion matrix) equally weights the contribution of each class to the final statistic and prevents performance on a single class dominating the statistic. This normalised statistic shows a drastic reduction in the performance estimates for each system and increases the relative distance between each of the systems in the evaluation.
 
−
[[User:Kriswest|Kriswest]]
 
−
 
 
−
=== References ===
 
−
# [http://gatekeeper.research.compaq.com/pub/compaq/CRL/publications/logan/icme2001_logan.pdf Logan and Salomon (ICME 2001), '''A Music Similarity Function Based On Signal Analysis'''].<br>One of the first papers on this topic. Reports a small scale listening test (2 users) which rate items in a playlists as similar or not similar to the query song. In addition automatic evaluation is reported: percentage of top 5, 10, 20 most similar songs in the same genre/artist/album as query.
 
−
# [http://www.ofai.at/~elias.pampalk/publications/pampalk06thesis.pdf E. Pampalk, '''Computational Models of Music Similarity and their Application in Music Information Retrieval.'''] <br>PhD thesis, Vienna University of Technology, Austria, March 2006
 

Latest revision as of 11:44, 26 July 2010

Introduction

These are the results for the 2007 running of the Audio Music Similarity and Retrieval task set. For background information about this task set please refer to the 2007:Audio Music Similarity and Retrieval page.

Each system was given 7000 songs chosen from IMIRSEL's "uspop", "uscrap" and "american" "classical" and "sundry" collections. Each system then returned a 7000x7000 distance matrix. 100 songs were randomly selected from the 10 genre groups (10 per genre) as queries and the first 5 most highly ranked songs out of the 7000 were extracted for each query (after filtering out the query itself, returned results from the same artist were also omitted). Then, for each query, the returned results (candidates) from all participants were grouped and were evaluated by human graders using the Evalutron 6000 grading system. Each individual query/candidate set was evaluated by a single grader. For each query/candidate pair, graders provided two scores. 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). A description and analysis is provided below.

The systems read in 30 second audio clips as their raw data. The same 30 second clips were used in the grading stage.

Summary Data on Human Evaluations (Evalutron 6000)

Number of evaluators = 20
Number of evaluations per query/candidate pair = 1
Number of queries per grader = 5
Size of the candidate lists = 48.32
Number of randomly selected queries = 100
Number of query/candidate pairs graded = 4832

General Legend

Team ID

BK1 = Klaas Bosteels, Etienne E. Kerre 1
BK2 = Klaas Bosteels, Etienne E. Kerre 2
CB1 = Christoph Bastuck 1
CB2 = Christoph Bastuck 2
CB3 = Christoph Bastuck 3
GT = George Tzanetakis
LB = Luke Barrington, Douglas Turnbull, David Torres, Gert Lanskriet
ME = Michael I. Mandel, Daniel P. W. Ellis
PC = Aliaksandr Paradzinets, Liming Chen
PS = Tim Pohle, Dominik Schnitzer
TL1 = Thomas Lidy, Andreas Rauber, Antonio Pertusa, José Manuel Iñesta 1
TL2 = Thomas Lidy, Andreas Rauber, Antonio Pertusa, José Manuel Iñesta 2

Broad Categories

NS = Not Similar
SS = Somewhat Similar
VS = Very Similar

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

NB: The results for BK2 were interpolated from partial data due to a runtime error.

BK1 BK2 CB1 CB2 CB3 GT LB ME PC PS TL1 TL2
Fine 0.412 0.178 0.539 0.446 0.439 0.554 0.541 0.512 0.377 0.568 0.519 0.491
Psum 0.417 0.147 0.586 0.468 0.457 0.597 0.582 0.539 0.376 0.619 0.556 0.521
WCsum 0.361 0.115 0.528 0.414 0.407 0.534 0.520 0.480 0.325 0.560 0.491 0.459
SDsum 0.333 0.099 0.499 0.387 0.382 0.503 0.489 0.451 0.299 0.531 0.459 0.428
Greater0 0.586 0.244 0.760 0.630 0.608 0.786 0.768 0.716 0.530 0.796 0.750 0.708
Greater1 0.246 0.050 0.412 0.304 0.306 0.408 0.396 0.362 0.222 0.442 0.362 0.334

download these results as csv


Friedman Test with Multiple Comparisons Results (p=0.05)

The Friedman test was run in MATLAB against the Fine summary data over the 100 queries.
Command: [c,m,h,gnames] = multcompare(stats, 'ctype', 'tukey-kramer','estimate', 'friedman', 'alpha', 0.05);

Friedman's ANOVA Table
Source SS df MS Chi-sq Prob>Chi-sq
Columns 3.0110e+003 11 273.7314 232.0407 0
Error 1.1263e+004 1089 10.3425
Total 1.4274e+004 1199

download these results as csv

TeamID TeamID Lowerbound Mean Upperbound Significance
LB CB1 -1.7998 -0.1350 1.5298 FALSE
LB CB2 0.1302 1.7950 3.4598 TRUE
LB CB3 0.1002 1.7650 3.4298 TRUE
LB BK1 -0.0598 1.6050 3.2698 FALSE
LB BK2 3.4352 5.1000 6.7648 TRUE
LB TL1 -1.4748 0.1900 1.8548 FALSE
LB TL2 -0.9048 0.7600 2.4248 FALSE
LB ME -0.9098 0.7550 2.4198 FALSE
LB PC 1.2052 2.8700 4.5348 TRUE
LB PS -2.3198 -0.6550 1.0098 FALSE
LB GT -2.2748 -0.6100 1.0548 FALSE
CB1 CB2 0.2652 1.9300 3.5948 TRUE
CB1 CB3 0.2352 1.9000 3.5648 TRUE
CB1 BK1 0.0752 1.7400 3.4048 TRUE
CB1 BK2 3.5702 5.2350 6.8998 TRUE
CB1 TL1 -1.3398 0.3250 1.9898 FALSE
CB1 TL2 -0.7698 0.8950 2.5598 FALSE
CB1 ME -0.7748 0.8900 2.5548 FALSE
CB1 PC 1.3402 3.0050 4.6698 TRUE
CB1 PS -2.1848 -0.5200 1.1448 FALSE
CB1 GT -2.1398 -0.4750 1.1898 FALSE
CB2 CB3 -1.6948 -0.0300 1.6348 FALSE
CB2 BK1 -1.8548 -0.1900 1.4748 FALSE
CB2 BK2 1.6402 3.3050 4.9698 TRUE
CB2 TL1 -3.2698 -1.6050 0.0598 FALSE
CB2 TL2 -2.6998 -1.0350 0.6298 FALSE
CB2 ME -2.7048 -1.0400 0.6248 FALSE
CB2 PC -0.5898 1.0750 2.7398 FALSE
CB2 PS -4.1148 -2.4500 -0.7852 TRUE
CB2 GT -4.0698 -2.4050 -0.7402 TRUE
CB3 BK1 -1.8248 -0.1600 1.5048 FALSE
CB3 BK2 1.6702 3.3350 4.9998 TRUE
CB3 TL1 -3.2398 -1.5750 0.0898 FALSE
CB3 TL2 -2.6698 -1.0050 0.6598 FALSE
CB3 ME -2.6748 -1.0100 0.6548 FALSE
CB3 PC -0.5598 1.1050 2.7698 FALSE
CB3 PS -4.0848 -2.4200 -0.7552 TRUE
CB3 GT -4.0398 -2.3750 -0.7102 TRUE
BK1 BK2 1.8302 3.4950 5.1598 TRUE
BK1 TL1 -3.0798 -1.4150 0.2498 FALSE
BK1 TL2 -2.5098 -0.8450 0.8198 FALSE
BK1 ME -2.5148 -0.8500 0.8148 FALSE
BK1 PC -0.3998 1.2650 2.9298 FALSE
BK1 PS -3.9248 -2.2600 -0.5952 TRUE
BK1 GT -3.8798 -2.2150 -0.5502 TRUE
BK2 TL1 -6.5748 -4.9100 -3.2452 TRUE
BK2 TL2 -6.0048 -4.3400 -2.6752 TRUE
BK2 ME -6.0098 -4.3450 -2.6802 TRUE
BK2 PC -3.8948 -2.2300 -0.5652 TRUE
BK2 PS -7.4198 -5.7550 -4.0902 TRUE
BK2 GT -7.3748 -5.7100 -4.0452 TRUE
TL1 TL2 -1.0948 0.5700 2.2348 FALSE
TL1 ME -1.0998 0.5650 2.2298 FALSE
TL1 PC 1.0152 2.6800 4.3448 TRUE
TL1 PS -2.5098 -0.8450 0.8198 FALSE
TL1 GT -2.4648 -0.8000 0.8648 FALSE
TL2 ME -1.6698 -0.0050 1.6598 FALSE
TL2 PC 0.4452 2.1100 3.7748 TRUE
TL2 PS -3.0798 -1.4150 0.2498 FALSE
TL2 GT -3.0348 -1.3700 0.2948 FALSE
ME PC 0.4502 2.1150 3.7798 TRUE
ME PS -3.0748 -1.4100 0.2548 FALSE
ME GT -3.0298 -1.3650 0.2998 FALSE
PC PS -5.1898 -3.5250 -1.8602 TRUE
PC GT -5.1448 -3.4800 -1.8152 TRUE
PS GT -1.6198 0.0450 1.7098 FALSE

download these results as csv

2007 ams broad scores friedmans.png

Summary Results by Query

These are the mean FINE scores per query assigned by Evalutron graders. The FINE scores for the 5 candidates returned per algorithm, per query, have been averaged. Values are bounded between 0.0 and 10.0. A perfect score would be 10. Genre labels have been included for reference.

Genre Query BK1 BK2 CB1 CB2 CB3 GT LB ME PC PS TL1 TL2
BAROQUE d004553 1.500 2.920 8.500 7.740 8.300 8.920 8.740 8.960 7.420 8.800 6.960 5.880
BAROQUE d006674 7.740 2.400 9.300 7.840 9.420 9.380 9.000 9.280 7.060 9.200 9.700 9.680
BAROQUE d008942 6.400 1.000 7.000 5.400 7.800 7.200 8.400 5.800 2.600 6.980 6.480 6.980
BAROQUE d010361 0.340 1.640 6.880 5.420 6.940 5.640 5.300 6.700 4.840 8.880 8.360 7.680
BAROQUE d011850 9.040 1.760 8.920 8.100 7.900 9.080 8.760 7.560 2.980 9.060 9.020 9.380
BAROQUE d012288 0.360 2.280 8.900 2.740 5.660 7.140 8.600 7.800 4.400 8.760 8.400 9.160
BAROQUE d017287 1.060 1.620 9.480 8.900 8.180 9.600 9.140 9.220 5.260 7.800 8.840 8.660
BAROQUE d018106 0.340 0.900 6.000 3.560 6.720 8.020 7.300 7.660 7.240 7.800 5.860 4.720
BAROQUE d018433 1.680 2.200 9.020 7.000 7.680 9.360 8.900 9.420 6.360 7.640 8.880 8.660
BAROQUE d018533 0.120 1.940 7.140 6.660 6.340 7.440 7.700 6.720 6.160 7.480 6.300 7.220
BLUES e000039 6.160 5.080 2.660 2.780 1.400 5.060 5.200 5.700 5.600 4.880 5.820 2.340
BLUES e001388 5.880 3.960 4.000 2.320 6.700 8.180 7.020 7.780 8.460 7.920 5.380 5.380
BLUES e004743 8.800 7.100 5.500 2.800 0.700 6.900 2.800 8.400 0.900 5.600 4.800 4.800
BLUES e004899 3.300 3.120 3.000 2.820 1.380 3.100 5.360 5.700 5.440 3.120 2.460 1.500
BLUES e006420 1.960 1.120 3.600 3.120 1.860 3.140 3.260 2.220 2.300 2.960 0.400 0.820
BLUES e011446 6.860 5.800 7.980 8.200 8.580 7.560 7.440 6.260 6.420 7.100 7.340 6.800
BLUES e012584 4.040 1.540 1.480 2.220 1.660 3.840 2.240 1.160 2.200 4.200 4.780 2.620
BLUES e013173 5.200 5.320 4.920 1.500 1.760 5.180 7.340 6.340 7.880 7.100 7.060 4.680
BLUES e015590 3.000 2.560 4.780 2.600 2.620 3.680 3.440 3.720 1.320 1.700 2.620 2.480
BLUES e018053 1.360 2.120 4.560 2.460 1.380 3.080 4.720 0.760 2.980 3.980 2.720 1.980
CLASSICAL d000429 4.640 1.960 4.780 3.280 4.840 5.820 5.900 5.140 0.360 5.840 4.780 4.880
CLASSICAL d003313 1.020 2.880 7.480 8.120 5.940 4.360 7.420 6.380 5.200 7.580 5.740 5.980
CLASSICAL d011706 3.280 3.240 9.040 8.940 9.060 8.920 9.020 9.100 1.680 8.140 7.920 5.460
CLASSICAL d012929 0.760 1.760 8.000 7.940 7.720 7.820 7.240 7.540 7.340 8.060 7.660 7.380
CLASSICAL d014060 1.320 2.660 7.500 6.860 6.740 7.320 8.360 6.520 5.440 7.040 6.560 6.420
CLASSICAL d014334 1.100 2.600 9.200 9.200 8.860 8.800 8.000 8.100 5.400 8.160 8.200 8.200
CLASSICAL d015555 5.220 0.000 5.100 0.400 1.940 5.140 4.900 3.900 0.100 6.940 5.440 4.580
CLASSICAL d017438 0.000 0.740 2.900 2.640 4.480 5.700 4.380 6.940 3.460 4.080 3.220 1.900
CLASSICAL d017826 0.000 1.380 2.080 1.140 2.640 4.000 1.860 4.800 1.420 3.980 3.700 2.120
CLASSICAL d019684 0.120 0.500 8.380 7.400 8.200 7.240 6.940 7.560 5.700 7.420 8.340 8.560
COUNTRY b008293 7.200 1.800 7.800 3.000 3.200 5.200 7.000 5.000 3.000 4.200 7.000 7.000
COUNTRY e003167 2.940 3.240 5.880 4.140 6.140 7.460 5.160 3.360 2.260 3.460 3.340 2.120
COUNTRY e005922 1.240 1.320 2.540 0.360 0.240 3.580 0.860 1.820 1.900 1.400 2.360 2.120
COUNTRY e008013 5.920 2.980 7.100 7.480 6.240 5.320 5.720 2.400 5.180 5.920 6.620 5.200
COUNTRY e008637 6.340 2.960 5.460 3.220 0.780 4.220 3.580 6.040 1.400 5.820 5.980 6.480
COUNTRY e009438 5.600 1.140 4.020 3.120 3.420 3.520 4.120 5.780 2.960 7.240 4.660 3.520
COUNTRY e011578 6.620 3.580 6.360 4.220 2.100 1.620 5.500 4.340 5.080 3.620 1.720 1.940
COUNTRY e011702 0.220 0.920 0.380 1.560 0.000 0.040 0.000 0.440 7.140 0.120 0.000 1.760
COUNTRY e014903 5.340 2.200 4.960 1.900 2.400 3.500 2.840 1.340 4.320 2.740 3.980 3.320
COUNTRY e018325 5.080 3.240 4.760 4.020 3.100 5.720 4.280 3.120 2.340 4.200 4.300 2.800
EDANCE a002630 4.840 0.100 3.460 2.800 3.300 3.360 6.220 5.200 2.880 3.760 5.480 4.720
EDANCE a003317 6.200 0.480 5.280 5.660 2.640 7.300 3.880 3.520 4.360 6.860 5.320 6.480
EDANCE a003588 8.780 0.000 5.400 6.340 6.840 6.040 4.380 4.000 1.560 9.100 6.720 2.120
EDANCE a006326 1.800 0.500 1.540 2.140 3.840 1.900 2.720 3.580 1.500 3.140 1.840 2.500
EDANCE b007008 7.420 0.000 4.420 4.980 5.960 8.000 7.180 8.200 5.960 8.400 6.020 5.620
EDANCE b008903 5.020 0.480 1.620 2.760 5.020 5.260 4.580 6.900 5.480 5.500 6.760 5.600
EDANCE b009057 1.940 0.060 4.820 4.360 5.360 2.780 5.020 3.240 1.260 3.000 2.840 6.840
EDANCE f000987 0.200 0.600 3.700 2.100 2.100 2.600 3.700 5.400 3.300 5.700 0.200 0.400
EDANCE f005146 2.540 1.000 5.320 3.740 2.280 3.320 3.580 2.740 1.360 6.600 2.620 6.500
EDANCE f010067 7.000 0.800 5.400 4.400 6.200 4.800 4.600 6.000 7.000 5.400 5.200 5.200
JAZZ e001054 4.000 4.000 7.000 6.200 6.400 6.600 5.400 4.800 5.400 5.800 5.600 5.600
JAZZ e004497 6.120 5.900 3.620 2.760 1.520 6.420 6.020 5.520 1.840 5.020 5.780 6.000
JAZZ e005048 8.240 7.100 4.060 4.140 1.300 4.200 2.300 6.600 7.920 3.120 4.900 3.420
JAZZ e007252 1.400 1.140 2.060 3.140 2.700 1.180 1.000 4.680 2.320 5.380 2.900 2.520
JAZZ e007772 5.540 4.300 3.440 3.640 3.160 2.920 1.920 1.820 2.660 3.780 2.580 4.140
JAZZ e007968 6.600 4.000 6.100 4.640 4.540 6.940 6.200 7.880 7.340 7.600 7.000 7.340
JAZZ e009441 5.360 0.720 4.200 1.800 0.380 1.500 4.760 2.800 3.440 4.540 4.260 3.280
JAZZ e010478 6.200 5.680 3.020 3.900 6.040 3.820 2.340 4.820 6.220 3.700 3.460 4.680
JAZZ e016650 3.780 2.860 6.600 4.040 0.000 4.280 4.320 5.100 3.380 5.080 0.940 0.940
JAZZ e019093 7.140 6.340 9.040 5.800 4.060 4.900 8.880 5.120 6.840 7.960 6.520 5.940
METAL b001148 0.920 0.640 5.180 4.920 6.140 4.880 6.720 6.840 2.280 5.700 5.360 5.480
METAL b002842 6.560 0.080 6.680 5.680 2.560 7.540 5.340 5.320 5.480 6.360 4.400 2.780
METAL b005320 6.060 1.100 6.080 3.140 2.660 5.440 7.000 6.880 1.420 6.020 5.380 6.300
METAL b005832 3.380 0.480 6.780 4.120 4.140 4.180 6.360 6.040 4.240 3.440 4.900 5.600
METAL b006870 4.960 0.360 6.800 5.720 6.220 6.360 5.520 7.260 3.800 6.180 7.100 6.760
METAL b010213 0.960 0.560 1.020 2.020 2.060 3.500 1.140 0.200 1.660 2.400 2.200 2.000
METAL b010569 6.200 2.200 6.400 5.400 3.200 5.400 6.800 5.200 3.000 4.200 6.200 4.800
METAL b011148 7.860 0.300 6.740 6.300 7.200 5.920 7.640 3.320 5.080 6.420 6.960 7.620
METAL b011701 6.460 0.000 6.660 7.220 1.640 7.400 7.940 8.160 5.100 8.180 8.160 7.760
METAL f012092 6.560 0.440 7.280 8.260 6.780 7.320 7.900 4.360 1.040 5.600 6.740 5.020
RAPHIPHOP a000154 8.660 0.360 9.160 9.120 9.340 8.880 8.900 8.360 8.840 8.820 9.400 7.680
RAPHIPHOP a000895 7.200 1.340 6.820 7.200 6.620 7.140 7.540 6.780 5.320 7.880 7.140 7.300
RAPHIPHOP a001600 2.440 3.760 1.400 0.420 0.220 2.840 5.660 4.120 1.240 1.640 2.360 2.260
RAPHIPHOP a003760 6.660 0.420 7.200 6.040 7.080 7.620 7.280 4.640 1.460 7.140 6.720 7.320
RAPHIPHOP a004471 6.480 0.040 7.600 7.480 7.640 7.540 7.040 7.000 7.460 7.740 8.480 7.840
RAPHIPHOP a006304 5.500 0.000 5.420 6.300 5.980 5.920 6.240 4.040 0.200 6.580 4.340 6.280
RAPHIPHOP a008309 8.380 0.620 7.120 6.420 8.280 8.380 7.500 7.500 1.100 7.640 7.980 7.660
RAPHIPHOP b000472 6.060 0.000 5.820 2.740 6.240 6.720 4.880 6.080 5.400 4.400 6.340 3.300
RAPHIPHOP b001007 3.960 0.840 2.700 3.600 4.660 3.540 2.920 5.700 1.760 2.140 1.200 1.040
RAPHIPHOP b011158 7.840 0.800 8.140 9.080 8.820 8.900 7.360 7.920 8.480 8.940 8.940 8.900
ROCKROLL b000264 0.760 0.000 0.100 0.720 0.280 0.940 0.360 0.160 0.480 0.200 0.920 0.320
ROCKROLL b000459 2.640 1.100 3.900 4.520 3.180 3.780 3.060 4.780 1.540 2.300 5.400 3.780
ROCKROLL b001686 2.600 0.000 4.200 6.500 0.900 5.700 2.700 5.700 0.300 3.800 2.900 1.400
ROCKROLL b005863 1.740 0.220 4.600 1.120 1.460 2.900 2.500 0.200 1.380 1.300 2.700 2.200
ROCKROLL b008067 4.000 1.200 4.800 2.000 1.600 4.800 5.200 4.600 2.400 3.800 5.800 5.000
ROCKROLL b009038 2.640 0.060 1.980 2.700 2.020 2.120 3.060 0.820 0.400 2.140 2.200 2.400
ROCKROLL b012623 1.920 0.000 3.940 0.880 0.200 3.100 2.020 1.940 2.080 2.020 1.500 2.080
ROCKROLL b015975 4.560 0.760 5.820 5.980 3.540 4.460 6.100 3.600 2.740 5.940 6.540 5.540
ROCKROLL b016172 1.900 1.040 3.800 2.100 1.900 3.960 3.860 1.240 2.240 2.560 3.460 2.920
ROCKROLL b019890 3.420 0.480 4.520 1.960 1.060 3.380 2.300 1.080 3.360 3.560 3.760 3.820
ROMANTIC d000095 1.600 2.340 7.640 6.360 7.740 7.420 6.860 7.040 4.200 7.740 3.300 3.020
ROMANTIC d000491 0.980 2.440 5.080 5.020 7.280 8.060 6.920 6.120 1.320 8.960 7.160 5.340
ROMANTIC d003906 6.800 1.560 4.600 3.960 4.760 8.020 5.840 7.020 4.200 7.300 6.740 7.140
ROMANTIC d005430 2.200 1.100 2.100 0.960 5.260 7.680 4.600 7.200 7.480 8.780 1.880 1.880
ROMANTIC d008298 3.860 0.260 5.120 1.840 3.000 5.660 5.200 2.600 0.000 7.120 5.000 6.260
ROMANTIC d008438 2.160 3.100 7.520 7.020 8.560 7.240 7.200 8.320 7.660 7.840 7.460 7.020
ROMANTIC d009148 6.260 2.360 3.960 4.220 5.380 5.940 5.780 2.960 1.240 4.760 6.520 6.500
ROMANTIC d009455 2.660 2.080 3.220 4.620 2.480 6.240 4.900 2.160 6.180 8.360 5.080 6.140
ROMANTIC d010229 7.000 0.020 6.300 4.720 4.700 7.100 4.700 0.000 1.420 6.900 4.900 5.700
ROMANTIC d010380 0.000 0.380 7.280 5.340 1.700 4.700 7.120 4.500 0.440 8.380 4.840 5.640

download these results as csv

These are the mean BROAD scores per query assigned by Evalutron graders. The BROAD scores for the 5 candidates returned per algorithm, per query, have been averaged. Values are bounded between 0 (not similar) and 2 (very similar). A perfect score would be 2. Genre labels have been included for reference.

Genre Query BK1 BK2 CB1 CB2 CB3 GT LB ME PC PS TL1 TL2
BAROQUE d004553 1.6 0.0 1.2 1.4 0.4 1.8 0.4 0.6 0.6 1.8 1.4 1.8
BAROQUE d006674 0.4 0.6 0.2 0.0 0.0 0.6 1.2 0.8 0.2 0.2 0.4 0.4
BAROQUE d008942 1.6 0.0 1.4 1.8 1.8 1.8 1.8 1.0 0.0 2.0 1.0 1.6
BAROQUE d010361 0.2 0.0 0.0 0.2 0.6 0.0 0.4 0.4 0.2 0.4 0.2 0.2
BAROQUE d011850 1.8 0.0 2.0 2.0 2.0 1.6 2.0 2.0 1.6 2.0 1.8 1.8
BAROQUE d012288 0.8 0.0 0.4 0.2 0.4 0.6 1.4 1.0 0.4 0.4 1.0 1.0
BAROQUE d017287 1.6 0.0 2.0 1.4 1.8 2.0 1.8 1.2 0.2 1.8 1.6 2.0
BAROQUE d018106 2.0 0.0 1.0 1.0 1.4 1.2 0.8 0.8 0.4 2.0 1.4 0.4
BAROQUE d018433 1.2 0.0 1.4 1.4 1.4 1.4 1.2 1.2 1.6 1.4 1.8 1.4
BAROQUE d018533 2.0 0.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 1.6
BLUES e000039 0.8 0.0 0.4 0.8 0.8 0.4 0.4 1.2 0.2 0.2 0.0 0.0
BLUES e001388 2.0 0.0 1.8 1.6 2.0 2.0 2.0 2.0 0.0 1.8 1.8 1.8
BLUES e004743 0.4 0.0 1.2 0.6 0.8 0.8 1.2 1.2 1.0 0.6 1.0 1.2
BLUES e004899 0.0 0.0 1.0 1.0 1.0 1.0 1.2 1.2 0.4 1.0 1.0 1.0
BLUES e006420 1.0 0.0 1.0 0.4 0.4 1.0 1.2 1.2 0.0 1.0 1.0 1.0
BLUES e011446 1.4 0.0 1.4 1.4 0.6 1.4 1.0 1.0 1.0 1.2 0.8 0.6
BLUES e012584 0.6 0.0 0.6 1.0 0.8 0.8 1.0 1.0 0.6 0.4 1.0 1.2
BLUES e013173 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
BLUES e015590 0.4 0.0 0.8 1.6 0.2 1.2 0.4 1.4 0.0 0.6 0.8 0.2
BLUES e018053 1.4 0.0 1.2 0.4 1.4 1.4 1.2 1.6 1.2 1.0 1.4 0.8
CLASSICAL d000429 0.8 0.0 0.4 0.6 0.4 0.6 0.8 0.2 0.0 0.4 0.8 0.6
CLASSICAL d003313 1.4 0.4 1.6 1.4 0.8 1.4 1.8 1.2 0.6 1.0 1.6 1.2
CLASSICAL d011706 0.4 0.0 1.2 0.8 1.4 0.6 1.2 0.6 0.0 0.6 0.6 1.8
CLASSICAL d012929 0.2 0.0 1.2 0.0 0.2 0.6 0.6 0.0 0.2 0.4 0.6 0.6
CLASSICAL d014060 1.2 0.0 0.2 0.4 1.2 1.2 1.2 1.6 1.2 1.4 1.8 1.2
CLASSICAL d014334 0.0 0.0 0.0 0.2 0.0 0.4 0.0 0.0 0.0 0.2 0.0 0.0
CLASSICAL d015555 1.0 0.0 1.8 1.2 1.4 1.4 1.0 1.6 0.4 1.4 1.8 1.6
CLASSICAL d017438 1.6 0.0 0.8 1.0 1.2 1.6 1.4 1.8 1.0 1.8 1.4 1.2
CLASSICAL d017826 0.8 0.0 1.2 0.4 0.0 1.0 1.2 1.2 0.4 0.8 1.0 1.0
CLASSICAL d019684 1.8 0.4 1.8 0.6 0.6 1.2 1.8 1.2 0.6 1.0 1.6 1.6
COUNTRY b008293 1.0 0.0 1.2 1.0 0.6 0.8 1.2 0.8 0.6 1.2 1.2 1.0
COUNTRY e003167 0.8 0.0 1.0 0.4 0.2 0.6 0.4 0.2 0.8 0.8 0.8 0.8
COUNTRY e005922 0.2 0.4 2.0 1.4 1.8 1.4 1.6 1.2 1.0 1.8 0.6 0.6
COUNTRY e008013 0.2 0.0 0.8 0.0 0.0 0.6 0.4 0.2 0.4 0.2 0.2 0.4
COUNTRY e008637 1.6 0.0 1.6 1.8 0.4 1.8 2.0 2.0 1.2 2.0 2.0 1.6
COUNTRY e009438 0.0 0.4 1.0 0.8 1.2 1.6 1.6 1.0 0.0 2.0 1.6 1.0
COUNTRY e011578 0.8 0.4 0.8 0.6 1.0 1.2 1.2 1.0 0.0 1.2 1.0 1.0
COUNTRY e011702 1.0 0.0 1.8 2.0 2.0 2.0 1.8 1.6 2.0 2.0 2.0 2.0
COUNTRY e014903 1.8 0.0 1.6 1.4 1.6 1.2 1.8 0.4 1.2 1.6 1.4 1.6
COUNTRY e018325 0.2 0.0 0.6 0.4 0.2 0.6 0.6 0.0 0.2 0.2 0.6 0.4
EDANCE a002630 0.0 0.4 2.0 1.6 2.0 2.0 2.0 2.0 1.4 2.0 1.0 0.8
EDANCE a003317 0.4 0.2 0.4 0.8 0.2 1.2 1.0 0.4 1.4 1.8 0.8 1.2
EDANCE a003588 1.0 0.6 0.8 1.0 1.2 1.0 1.0 0.4 0.4 1.0 1.2 1.2
EDANCE a006326 0.2 0.2 0.4 0.2 0.8 1.6 0.6 1.4 1.4 2.0 0.2 0.2
EDANCE b007008 1.4 0.2 1.8 1.6 2.0 1.8 1.6 2.0 1.0 1.8 2.0 2.0
EDANCE b008903 1.2 0.2 1.6 1.2 1.8 1.6 2.0 1.4 0.6 1.4 1.4 1.6
EDANCE b009057 1.0 0.0 1.4 0.4 0.8 1.6 1.2 0.6 0.0 2.0 1.2 1.6
EDANCE f000987 0.0 0.4 1.6 1.8 1.2 0.8 1.6 1.2 1.2 1.4 1.2 1.2
EDANCE f005146 0.4 0.8 2.0 2.0 2.0 2.0 2.0 2.0 2.0 1.8 2.0 2.0
EDANCE f010067 1.4 0.4 1.0 0.6 0.8 2.0 1.2 1.4 1.0 1.8 1.6 1.6
JAZZ e001054 0.0 0.4 1.8 1.4 1.4 1.6 1.8 1.4 1.2 1.6 1.6 1.8
JAZZ e004497 0.0 0.2 2.0 2.0 1.8 2.0 2.0 2.0 1.0 1.6 2.0 2.0
JAZZ e005048 0.0 0.2 2.0 1.6 1.6 2.0 2.0 2.0 1.2 1.6 2.0 2.0
JAZZ e007252 1.0 1.0 1.8 1.4 1.4 1.6 1.0 1.0 1.0 1.2 1.2 1.2
JAZZ e007772 0.0 0.2 0.6 0.6 1.0 1.4 1.4 1.8 0.8 1.0 0.8 0.4
JAZZ e007968 0.0 0.2 0.2 0.0 0.4 1.0 0.4 1.0 0.2 1.0 0.8 0.2
JAZZ e009441 0.0 0.2 1.2 0.6 1.4 1.8 1.6 1.8 1.6 2.0 1.4 1.0
JAZZ e010478 0.0 0.0 2.0 1.8 2.0 1.8 1.6 1.6 1.2 1.6 2.0 2.0
JAZZ e016650 1.4 1.2 0.4 0.4 0.2 1.0 1.0 1.0 1.0 1.0 1.2 0.6
JAZZ e019093 1.2 0.8 1.0 0.6 1.4 2.0 1.6 1.6 1.8 1.8 1.2 1.2
METAL b001148 0.0 0.4 2.0 0.4 1.2 1.8 2.0 1.8 1.0 2.0 2.0 2.0
METAL b002842 2.0 0.2 2.0 1.8 2.0 2.0 1.8 1.6 0.6 2.0 2.0 2.0
METAL b005320 1.8 0.0 1.4 1.2 1.2 1.6 1.2 0.0 0.2 1.6 1.2 1.4
METAL b005832 0.0 0.2 1.4 1.0 1.4 1.0 1.0 1.2 1.0 2.0 1.6 1.2
METAL b006870 1.2 0.0 1.4 0.2 0.6 1.2 1.2 1.0 0.2 1.8 1.2 1.0
METAL b010213 0.0 0.4 2.0 2.0 1.8 1.8 1.6 1.8 1.8 2.0 1.8 1.6
METAL b010569 0.4 0.4 2.0 2.0 2.0 2.0 2.0 2.0 0.0 1.8 1.6 1.0
METAL b011148 0.0 0.4 2.0 2.0 1.8 2.0 1.8 1.6 1.2 1.8 1.8 1.8
METAL b011701 0.0 0.4 2.0 1.8 1.8 2.0 2.0 1.6 1.4 1.8 1.8 1.8
METAL f012092 0.0 0.0 1.6 1.2 0.0 1.0 1.6 0.8 0.0 1.8 1.0 1.4
RAPHIPHOP a000154 1.0 0.8 0.6 0.8 0.4 0.4 0.2 0.4 0.4 0.8 0.4 0.8
RAPHIPHOP a000895 0.8 0.8 0.4 0.6 0.2 0.8 1.0 1.0 0.8 0.8 0.4 0.2
RAPHIPHOP a001600 1.6 0.8 1.2 1.0 1.0 1.6 1.2 2.0 1.8 1.8 1.6 1.8
RAPHIPHOP a003760 0.0 0.0 0.4 0.4 0.4 0.0 0.0 0.8 0.2 1.2 0.4 0.2
RAPHIPHOP a004471 1.6 1.4 0.8 0.8 0.2 0.6 0.4 1.6 1.4 0.6 0.8 0.6
RAPHIPHOP a006304 1.0 1.0 0.8 0.4 0.0 1.2 0.8 1.2 0.2 1.0 0.8 1.0
RAPHIPHOP a008309 0.8 0.4 1.0 1.0 0.4 0.8 0.8 0.6 0.8 0.8 0.0 0.2
RAPHIPHOP b000472 0.4 0.6 1.0 0.8 1.0 1.8 1.0 0.6 0.2 0.6 0.6 0.2
RAPHIPHOP b001007 0.0 0.0 0.4 0.0 0.0 0.6 0.0 0.2 0.4 0.2 0.4 0.2
RAPHIPHOP b011158 2.0 1.6 1.2 0.4 0.0 1.8 0.6 2.0 0.2 1.2 1.2 1.2
ROCKROLL b000264 1.0 0.0 0.6 0.6 0.4 0.6 0.8 1.0 0.6 1.4 0.8 0.4
ROCKROLL b000459 1.6 0.6 1.2 0.8 0.2 1.0 0.6 1.4 0.2 1.2 1.4 1.6
ROCKROLL b001686 0.8 0.0 0.0 0.4 0.4 0.8 0.2 0.2 0.2 0.8 1.0 0.4
ROCKROLL b005863 1.2 1.2 1.0 0.4 0.4 1.0 1.6 1.4 1.8 1.6 1.6 0.8
ROCKROLL b008067 1.4 0.6 1.6 1.6 1.2 1.2 1.4 0.2 1.0 1.4 1.6 0.8
ROCKROLL b009038 1.4 0.6 1.4 0.8 0.4 0.4 1.0 0.6 0.8 0.6 0.4 0.4
ROCKROLL b012623 0.0 0.2 0.0 0.2 0.0 0.0 0.0 0.0 1.4 0.0 0.0 0.4
ROCKROLL b015975 1.0 0.8 0.4 0.6 1.2 0.8 0.4 0.8 1.6 0.6 0.8 1.0
ROCKROLL b016172 1.2 0.0 1.0 0.2 0.0 0.2 1.0 0.2 0.6 1.0 1.0 0.8
ROCKROLL b019890 1.2 1.0 1.8 2.0 2.0 1.8 1.6 1.2 1.4 1.4 1.6 1.4
ROMANTIC d000095 1.4 0.8 1.2 1.0 0.8 1.4 1.2 0.8 0.6 1.2 1.2 0.8
ROMANTIC d000491 0.0 0.0 0.8 0.4 0.4 0.4 0.8 1.4 0.4 1.2 0.0 0.0
ROMANTIC d003906 0.2 0.0 1.0 0.4 0.2 0.4 0.4 0.4 0.0 1.4 0.4 1.4
ROMANTIC d005430 0.8 0.4 1.4 0.8 0.0 0.8 1.0 1.0 0.8 1.2 0.2 0.2
ROMANTIC d008298 0.4 0.4 1.0 0.4 0.4 0.6 0.6 0.8 0.2 0.2 0.2 0.2
ROMANTIC d008438 0.4 0.6 1.4 0.6 0.4 0.6 1.2 0.2 0.8 1.2 1.0 0.6
ROMANTIC d009148 1.8 0.0 1.4 0.8 1.4 1.0 1.0 1.4 1.6 1.0 1.2 1.4
ROMANTIC d009455 1.6 1.6 2.0 1.2 0.8 1.0 2.0 1.2 1.6 2.0 1.2 1.2
ROMANTIC d010229 1.6 0.0 2.0 2.0 1.6 2.0 2.0 1.0 0.0 1.2 1.6 1.2
ROMANTIC d010380 1.2 0.4 0.8 0.4 0.4 0.8 0.6 0.2 1.0 0.4 0.6 0.6

download these results as csv

Anonymized Metadata

Anonymized Metadata

Raw Scores

The raw data derived from the Evalutron 6000 human evaluations are located on the 2007:Audio Music Similarity and Retrieval Raw Data page.

Retrieved from "https://music-ir.org/mirex/w/index.php?title=2007:Audio_Music_Similarity_and_Retrieval_Results&oldid=7411"