A study performed by researchers in Europe found that a widely used music recommendation algorithm is more likely to choose music by male artists to the detriment of female artists.
Study authors tested a popular music recommendation algorithm across two song datasets. That process revealed that in both instances the algorithm ended up “reproducing” existing biases in the datasets. Results reveal both datasets displayed biases against women, with female artists only accounting for 25 percent of the songs.
In addition, the algorithm constructs a ranked list of songs for a user to sample. Across the board, the highest a female artist ever ranked on such lists was number six or seven.
The problem with this ranking is that researchers found most people usually end up listening to songs their playlists suggest to them. So, as an individual listens to more suggested songs, that reinforces the idea that the algorithm is doing something right, creating a “feedback loop” of gender bias.
The authors of the study propose a new approach that would allow greater exposure of female artists by manually reordering the recommendations, to ensure more female songs are suggested earlier.
In a simulation, the authors studied how classified recommendations would affect user behaviour in the long run. The results showed that, with the help of the reclassified algorithm, users would begin to change their behaviour and thus listen to more female artists than with other music recommendation algorithms and, moreover, the new algorithm, based on machine learning, would consolidate this change in behaviour.
It seems like a good suggestion to me. Algorithms are meant to be tweaked until you think you’ve got it right. So it will be interesting to see how the algorithm does in real life.
I checked in with Spotify and found what is referred to as Discover Weekly. Every Monday morning, Spotify listeners are greeted—some might say gifted—with a new Discover Weekly playlist to help set their soundtrack for the next seven days.
These algorithms look for how those songs are played and ordered in other Spotify users’ playlists. If it turns out that, when people play those songs together in their playlists, there’s another song sandwiched between them that someone has never heard before, that song will show up in your Discover Weekly.
So I did a bit of investigation and took a look at my latest Discover Weekly playlist. Of the 30 songs recommended for me, there were NO women artists.
Now I know that this is partly a function of what music I listen to, which is heavily dominated by male artists, but it’s certainly not void of women artists. But as the researchers pointed out, such a recommendation algorithm just perpetuates the bias.
How is a person to get exposed to new music, if these algorithms just recommend music and artists that are similar to what you already listen to?
So I can see how the manual reordering of the ranked list, with some female artists specifically included near the top of he Discover Weekly, would be helpful.
I would expect that these algorithms keep getting better and better, and so perhaps after a few months of manual reordering there may be an opportunity to let the algorithm work on its own.
In the meantime, I plan to play Gaslighter by the Chicks a few times every day over the next week to see what impact that will have on next Monday’s Discovery Weekly.
And if you haven’t listened to the song or seen the video, well, I think they are both marvelous (I have shared this before):
Sources used:
https://dittomusic.com/en/blog/how-to-get-your-music-on-discover-weekly-and-release-radar/
I wonder if it will make a difference? Will be interesting to read next week what the outcome of your exercise is.
LikeLiked by 1 person
I’ll let you know!
LikeLiked by 1 person
I started thinking about the ratio of female to male artists I listen to. I would say it’s 60% male to 40% female. I don’t know what that all means—probably nothing. In the meantime, I’ll listen to Gaslighter. That tune qualifies as one of my 40%.
LikeLiked by 2 people
I think that’s a much more balanced ratio than I have, but like you, I’m not sure what it means…
LikeLiked by 1 person
Sometimes I wonder, Pete, whether a natural inclination in each gender isn’t viewed as gender bias when it favours men. My experience of women in the work place is that they don’t want equality, they want concessions. They want to be paid the same but accommodated over things like sick children and other home-making related decisions and circumstances. Equality means the same. I would never say that there isn’t mistreatment of women, because it abounds. It has to be addressed through education and upliftment and not ratios in the workplace that prejudice men. Just my two pennies, but I feel strongly about this because I bear witness to what is happening in corporates and it is becoming more and more unfair towards men.
LikeLiked by 2 people
as always, Robbie, I appreciate your perspective on things. I never thought about women wanting more concessions than men, and that the business world is becoming more unfair towards men.
LikeLiked by 2 people
Hi Jim, I can’t speak for how it is in the USA, but I certainly see this happening more and more in the UK and here in South Africa. Women either have to behave in the same manner as men [which some do] and employ someone to look after their children and run the home, or they need to step back and accept earning less and having a lessor role. You can’t have it all.
LikeLiked by 2 people
and is it never the man who would accept such responsibilities with the children and the house?
LikeLiked by 1 person
There are men that do take on the role of home maker and their wives are usually the bread winner and do often compete on a level playing field. Among the younger generation though, the demand for concessions for many things, one of which is family commitments is big. Ladies want to work reduced hours and flexi-time but then complain if they don’t get put on the big transactions and don’t progress as quickly. You have to make choices in life. There is also a lot of unreliability among women who just drop the ball if something comes up in their home life. You just can’t do that in corporate. Meanwhile, the men are expected to pick up all the slack in the teams and are then told they won’t be promoted because the firm needs more female executives in their statistics. It just isn’t right.
LikeLiked by 2 people
no one said it was easy to try and balance work and family life…
LikeLiked by 2 people
such an interesting observation and i’ll love to see how this plays out – something i’ve never really consciously thought about, but i’d say my playlist and suggested list is definitely heavily titled to the male side.
LikeLiked by 1 person
I never gave it much thought to it either, but like you, my guess is that my usual music is mostly male…
LikeLiked by 1 person
I, too, will be looking forward to hearing what results your experiment will produce on the songs suggested to you in your Discover Weekly. I stream my music from Sirius XM at work and on the motorcycle, but I am not aware of any suggested playlist as with other streaming services. In the car, I listen to the radio (commercials and all) just to have the chance to listen to new music that may not make it on the channels I prefer on XM. It is sad to think that there is yet another hurdle for women artists to overcome in an already difficult industry. Great post where I again learn something I didn’t know. Thanks for bringing this into the light!
LikeLiked by 1 person
It seems like there is a bit of gender bias in the music industry. So how does it work listening to music while on your motorcycle? Do you have it playing through your helmet? I would have just thought it would be hard to hear music while out riding…
LikeLiked by 1 person
I have a fairing on the bike that has two speakers. The volume increases with the speed, so the engine and wind noise never drown it out. Technology is an amazing thing!
LikeLiked by 1 person
that is a cool technology!
LikeLiked by 1 person
Interesting findings but sadly they didn’t come as a surprise to me. I listen to a lot of female singers but, having checked back through my past few Tuesday Tunes posts, I’m under-representing them too. Two out of six in today’s post – not great, but I’m getting there! I’ll be interested to see the effect of your experiment. Maybe worth trying it again another week but with different songs by different female artists, to see if that changes the results? As you know, I love The Chicks and have played this one on my blog – always good to have another excuse to watch that video though 😊
LikeLiked by 1 person
I would think your playlist would be the most balanced of anyone I know; I wonder what the percent of male vs female artists is in the music industry.
I’m listening to The Chicks while typing this… 🙂
LikeLiked by 1 person
I checked YouTube recommendations after writing that. They create mixes for you based on what you’ve listened to before. The first eight songs on today’s mix were all sung by females, so maybe I do listen to them a lot!
That’s an interesting point but it would be very difficult to assess. Would you go just by those who have a recording contract? I think that might be too narrow, but it’s a huge field to study.
Good to know you’re keeping the experiment going. You do know they have other songs too, right? If you have access to YouTube try to find their video for Goodbye Earl – it will make you laugh.
LikeLiked by 1 person
I’m not surprised you had so many women artists in your mix – were they country artists?
And you are right, that would be quite difficult to measure.
And I actually listened to the whole album that Gaslighter was on – quite good…
LikeLiked by 1 person
Surprisingly, only one: Caroline Jones. I’ve been listening to a lot of female folk singers over the past year and it reflected that. Kate Rusby was there, and the German band Faun who I featured some weeks ago – they are led by two female vocalists. The rest followed from there, and there was some Foxes and Fossils in the list too.
I think it would be a massive undertaking to research that – maybe someone could do it for a PhD?
It’s good that you listened – should help your stats. It’s a good album but I actually prefer their earlier ones.
LikeLiked by 1 person
that’s quite a list of female singers; I recognize all their names thanks to you!
LikeLiked by 1 person
Good to know my work is getting results 😉
LikeLiked by 1 person
👍
LikeLiked by 1 person
and I’ll check out Goodbye Earl 🙂
LikeLiked by 1 person
Have fun 😊
LikeLiked by 1 person
just watched, that was fun. catchy tune. poor earl… 🙂
LikeLiked by 1 person
It’s not a classic song, but everyone has a lot of fun with it in that video. I saw them play it live, and they all did those dance moves – some of their backing band joined in too. A fun moment…
Earl got what he deserved – underneath the fun and catchy tune there is a serious message there. And anything with Lauren Holly in it is good for me 😉
LikeLike
That’s a huge bias! I wonder what the ratio of male/female artists in general is. It will be even more interesting if overall, there are more female artists than men.
LikeLiked by 1 person
I wish I knew that ratio as well; it may be hard to measure…
LikeLiked by 1 person
Yes I wonder if the Spotify suggestions actually relate very much to what you play as an individual or to what subscribers play as a whole. Also the ratio of male to female subscribers would be interesting.
LikeLiked by 1 person
I think the suggestions are based on a combo of what I listen to and what others listen to. And that would be nice to know that ratio, but it could be tough to measure…
LikeLiked by 1 person
If we do not want the matrix to control us we must escape from it.
LikeLiked by 1 person
🙂
LikeLiked by 1 person
You’re smiling now…
LikeLiked by 1 person
Between robots and algorithms there is not much hope for humans.
LikeLiked by 1 person
we will be their servants before you know it 🙂
LikeLiked by 1 person
The percentage of female singers to male singers is quite low plus girls tend to be soloists rather than in bands
or as part of a male band be interesting and something for you Jim to explore further as you love your stats…I did agree with Robbies comment about women though…:) x
LikeLiked by 1 person
my guess would be the same as yours, Carol, that there are many more men making a living in the music industry than women. and yes, Robbie’s comments are eye-opening…
LikeLiked by 1 person
When I was a country music deejay, back in the 80s, our policy was to never play two female performers back-to-back. The theory was that listeners preferred male performers to female. I never felt comfortable with this limitation, but on the other hand there were at least two male performers per one female, so mathematically it kind of worked out.
LikeLiked by 1 person
this is the first I’ve heard of you being a dj…
that seems like a strange policy; I wonder if it still exists, or if just naturally works out that way…
LikeLiked by 1 person
I was a deejay for about 8 years. No, it doesn’t naturally work out that way. Deejays follow a system. Sometimes it’s rotational, and sometimes it’s a “clock,” where certain types of tunes are played at certain times. I don’t know if the policy of no two females in a row still exists, since it’s been over 30 years since the last time I worked the turntables.
LikeLiked by 1 person
so who set the policy?
LikeLiked by 1 person
The Program Director. I was the Music Director, but he made the final decisions.
LikeLiked by 1 person
and it sounds like there was no room for debate…
LikeLiked by 1 person
Not much, no. On that issue, the PD was very adamant.
LikeLiked by 1 person
I guess he was going by what he thought the audience wanted…
LikeLiked by 1 person
Yep, though I think he was wrong. Then again, I tend to be skeptical of experts.
LikeLiked by 1 person
and I am sure your skepticism is often warranted…
LikeLiked by 1 person