Band of Skulls

This Southampton formed three-piece have been making a lot of noise across the pond. Check out this chart that shows the the top locations for the bands followers on MySpace.

Austin comes in at 7th place for social network fan locations, so South by South West should be a good gig.

Band of Skulls: From Southampton to L.A. - Click to Enlarge

The next chart is a good visualisation of where people have been downloading the latest Band of Skulls Album – Baby Darling Doll Face Honey using Bittorent, as you can see – they are pretty popular in the USA on Bittorent too. Maybe some of those people will attend the gig ?

Band of Skulls - Baby Darling Doll Face Honey : Bittorent download locations

And finally,  here is a chart showing the top countries for Band of Skulls, also based on where people are downloading their album on Bittorent. It shows the same data displayed in the map above, but aggregated by country and shown as a bar chart. Interestingly, their home country Great Britain comes in at 3rd position, behind USA and Canada.

Band of Skulls - Top regions for Bittorent downloads - Click to expand

Why not check out the gig if you’re going to SXSW? Visit their MySpace page for the dates – We’ll be there!

Below are the results for the top 10:

1. David
2. Maria
3. Michael
4. John
5. Daniel
6. Chris
7. Laura
8. Jose
9. Juan
10. Sarah

Over 100,000 unique first names were detected in total

Personally I think  it’s down to Robbie winning Best International Male at the ECHO Awards (a German music award) last night.  Robbie is pretty popular in Germany on Social Networks, as you can see from the second bar chart.

MySpace Views & Plays - Click to Expand

Robbie Williams Top Countries - Click to Expand

]]> http://www.musicmetric.com/2010/03/robbies-myspace-views/feed/ 0 http://www.musicmetric.com/2010/03/nme-awards-2010/ http://www.musicmetric.com/2010/03/nme-awards-2010/#comments Fri, 05 Mar 2010 14:24:14 +0000 nick http://www.musicmetric.com/?p=269281839 The NME Awards have long prided themselves on being an anti-award show event, and while this year’s affair was rather more civilised than previous years they still managed to fit in some decapitated heads, a rock star cat fight and a game of catchphrase… However amidst all the debauchery some interesting stats and trends can still be found.

Muse and Kasabian were the two biggest winners on the night picking up two awards each. Muse fended off competition from Kasabian, Arctic Monkeys, Biffy Clyro and Oasis to pick up the Best British Band award. They also picked up an award for Best Band Website so it’s no wonder they have the biggest online presence of all the bands nominated in the Best British Band category.

The below bar chart shows total fan numbers for the Best British Band nominees across a spread of social networking sites.

Best British Band - Click to Expand

Bombay Bicycle Club took home the award for Best New Band, previous winners of this award include Kings of Leon, Arctic Monkeys and The Strokes so they are certainly in good company. However they have had a slow start to the year as you can see by the below trend graph. This shows Myspace plays since the begging of this year, The XX are racking up the play count with Bombay Bicycle Club appearing near the bottom of the graph.

Best New Band - Click to Expand

We see a similar trend in the Best Solo Artist artist category, Jamie T even admitted he’d had a ’stop-start year’. Florence is way out in front with her play count and we can see the winner of the award Jamie T down at the bottom of the graph.

Best Solo Artist - Click to Expand

The eagle eyed among you may have spotted that there is a nominee missing off this graph. As you can see in the following graph her sheer popularity makes it hard to compare Miss GaGa to the other nominee’s.
(For reference the multi-coloured line along the bottom of the graph represent the other nominee’s…)

Best Solo w/ Lady Gaga - Click to Expand

And finally here’s a nice graph showing the effect of the NME Awards on the winners play count. The artists plays all begin to rise on the morning after the awards and continued to rise through the day.

The NME Effect - Click to Expand

It’s always interesting to see the impact of conventional media (TV/Radio/Print) on the digital realm. However seeing as the play count only rises the day after the awards, it is more likely that the reporting of the awards in various blogs, news posts and tweets that has prompted this rise in play count.

Our methods are much more advanced than simple word detection. We have implemented a set of machine learning models that can be trained with different corpora (contexts) so they work well for general language but are also much more accurate for the pre defined contexts – for example, professionally written articles, fan comments and tweets are all different contexts and therefore have different sentiment analysis models trained for each one. Using this approach allows our model to get more and more intelligent as we keep downloading data to retrain it frequently. The accuracy of our method is shown in the confusion matrices below:

Musicmetric polarity confusion matrix

So what does this matrix mean? This confusion matrix tells how many percent the system is confusing two classes (i.e. mislabeling one as another). Each column of the matrix represents the instances in a predicted class, while each row represents the instances in an actual class (yellow hightlighted ones are correct predictions). For example, we can see that 16% of neutral reviews are predicted as negative but only 2% of negative reviews are predicted as neutral or positive. Notably we have 96% of negative reviews are predicted correctly.

The second confusion matrix is a breakdown for score from 1-5 (3 is neutral):

Confusion matrix for score from 1 to 5

Similarly we can see that 90% of reviews with scores of 2 are predicted correctly while 5% of them are predicted as 3 and none of them are predicted as 5. The below table shows the numbers of reviews we evaluated:

Number of evaluated samples

The third confusion matrix is a break down for score from 1-10 (Note: this test data is different from above ones and does not include any reviews which have scores less than 5):

Musicmetric sentiment confusion matrix for score from 1-10

We think that is enough of our talking. It is now your turn to see how it works by playing around with the interface to our general purpose music sentiment analysis engine below:

Note: Our sentiment analysis usually works better for longer reviews or paragraphs rather than single short sentences, and definitely works better for music related topics

We were picked from over 120 submissions by an advisory board of industry experts drawn from the UK community including: VC’s, export specialists, legal experts, journalist and industry pundits, to take part in the Digital Mission to SXSWi.

Digital Mission is organised by Chinwag and UK Trade & Investment (UKTI), enabling digital companies to expand into overseas markets and attract investment outside the UK.

If anyone is going to SXSW and would like to meet up – just give us a call!

We hope to see you there!

The Musicmetric team.

But wait one second – comparing MySpace plays between Susan Boyle and Lady Gaga shows wildly different popularities between the two artists on that social network:

Lady Gaga and Susan Boyle plays on MySpace - click to expand

Squint really hard and you’ll see a blue line right near the time axis – this represents plays per hour on MySpace for Susan Boyle. For a sense of scale, look at the highlighted values: 144 plays per hour for Susan Boyle and 50,000 plays per hour for Lady Gaga.

Surely this shows some kind of error considering Susan Boyle’s latest success in the UK and US charts? Let’s drill down into some other sources of data to attempt to identify the reason for this difference:

The chart below shows the number of new followers/friends/fans per hour added or lost by Susan Boyle on social networks:

Musicmetric - Susan Boyle friends per hour Facebook - click to expand

At first glance it looks pretty messy, but if you look carefully you’ll notice that the most dynamic of the trends for is Fan Adds on Facebook (green line). Some days she’s getting 14 new fans per hour, but others she’s losing then at the same rate. This gives a clue as to where fans of Susan Boyle are the most active online…Facebook

Now let’s look at where Susan Boyle fans hang out online compared to Lady Gaga:

Lady Gaga & Susan Boyle fans on Social Networks - click to expand

Clearly most of Susan Boyle’s online fans reside on Facebook. Looking at MySpace – If you compare the number of Susan Boyle fans with the number of Lady Gaga fans, it’s obvious that Lady Gaga has way more fans on than Susan Boyle on this particular social network: 2101 MySpace fans for Susan Boyle vs. 911,993 MySpace fans for Lady Gaga. This helps explain the considerable difference between the MySpace play counts for the two artists, and indeed – the ratios of number of fans on MySpace for the two artists and number of plays for them are relatively close:

Susan Boyle MySpace fans = 2012
Lady Gaga MySpace fans = 911,993
RATIO = 450:1

Susan Boyle MySpace plays per hour = 144
Lady Gaga MySpace plays per hour = 50,014
RATIO = 350: 1

Now let’s take a look why Susan Boyle has so few fans on MySpace compared to Lady Gaga by taking a peek at the fan age profile for both artists:

Susan Boyle and Lady Gaga fan ages - click to expand

That above chart clearly shows that the fan age spread for Susan Boyle is considerably wider than that for Lady Gaga (the numbers represent percentage of fans of any given age). This goes some way towards explaining the fact that Susan Boyle has less ‘fans’ on MySpace than would be expected – her fan age demographic is much older than the normal MySpace user demographic and falls more into the demographic for Facebook or older – explaining the larger number of Facebook fans.

Now taking into account the high number of physical sales Susan Boyle has made, one could make the conclusion that since the age demographic for her fans is considerably higher than that of the average social network or streaming music user demographic – it would be fair to assume that most of Susan Boyles fans do not use the social music web to listen to her music, they would rather buy her CD. This is backed up by her low play count on MySpace.

So there we go – these examples demonstrate a good reason why taking stats at face value, without looking at the full picture can give incorrect insight into music trends.

Previously Jeremy was worldwide Vice President of New Media for EMI Music in London and Los Angeles. He then took the position of CEO at Sibelius Software which he led for five years, leading to an acquisition by Avid Technology. Some of Jeremy’s other music industry experience includes working as Director of Media Affairs at Virgin Records, running the ground breaking playlist-sharing music service Uplister Inc. and launching early music website The Raft.

He serves as Deputy Chairman of Futurelab, an education technology think-tank chaired by Lord David Puttnam, and regularly presents at music industry conferences around the world.

He is currently CEO of the newly created Featured Artist Coalition, a new music industry organisation co-chaired by Nick Mason of Pink Floyd and David Rowntree of Blur. Members include: Radiohead, Annie Lennox, Tom Jones, Robbie Williams, Kate Nash and many others.

Check out this chart showing when people are wishing Merry Christmas to the world on Twitter.

The chart shows the number of Merry Christmas tweets per 10 minute time interval since the 22nd December.

Merry Christmas on Twitter - Number of Tweets per 10min interval