Song Recommendation System

In this article we focused on building recommender systems to find products, music and movies that interest users. We have also built model which compared the simple popularity-based recommendation with a personalized model, and showed the significant improvement provided by personalization.

In this Ipython Notebook we are going to explore the song data and the recommendations made by our model. In this process, we are going to learn how to use one of the most important data manipulation primitives groupby.

import graphlab

# CSV format datasets https://d396qusza40orc.cloudfront.net/phoenixassets/song_data.csv
song_data = graphlab.SFrame('coursera-notebooks/course-1/song_data.gl')

[INFO] This non-commercial license of GraphLab Create is assigned to prashantgonarkar@gmail.com and will expire on February 13, 2017. For commercial licensing options, visit https://dato.com/buy/.

[INFO] Start server at: ipc:///tmp/graphlab_server-1181 - Server binary: /usr/local/lib/python2.7/dist-packages/graphlab/unity_server - Server log: /tmp/graphlab_server_1457246816.log
[INFO] GraphLab Server Version: 1.8

Exploring data

song_data.head()

user_id	song_id	listen_count	title	artist
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOAKIMP12A8C130995	1	The Cove	Jack Johnson
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOBBMDR12A8C13253B	2	Entre Dos Aguas	Paco De Lucia
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOBXHDL12A81C204C0	1	Stronger	Kanye West
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOBYHAJ12A6701BF1D	1	Constellations	Jack Johnson
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SODACBL12A8C13C273	1	Learn To Fly	Foo Fighters
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SODDNQT12A6D4F5F7E	5	Apuesta Por El Rock 'N' Roll ...	Héroes del Silencio
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SODXRTY12AB0180F3B	1	Paper Gangsta	Lady GaGa
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOFGUAY12AB017B0A8	1	Stacked Actors	Foo Fighters
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOFRQTD12A81C233C0	1	Sehr kosmisch	Harmonia
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOHQWYZ12A6D4FA701	1	Heaven's gonna burn your eyes ...	Thievery Corporation feat. Emiliana Torrini ...

song
The Cove - Jack Johnson
Entre Dos Aguas - Paco De Lucia ...
Stronger - Kanye West
Constellations - Jack Johnson ...
Learn To Fly - Foo Fighters ...
Apuesta Por El Rock 'N' Roll - Héroes del ...
Paper Gangsta - Lady GaGa
Stacked Actors - Foo Fighters ...
Sehr kosmisch - Harmonia
Heaven's gonna burn your eyes - Thievery ...

[10 rows x 6 columns]

len(song_data)

graphlab.canvas.set_target('ipynb')

song_data['song'].show()

Count number of users

users = song_data['user_id'].unique()

len(users)

Create song recommender

train_data,test_data = song_data.random_split(.8,seed=0)

Simple popularity-based recommender

popularity_model = graphlab.popularity_recommender.create(train_data,
                                                         user_id='user_id',
                                                         item_id='song')

PROGRESS: Recsys training: model = popularity
PROGRESS: Warning: Ignoring columns song_id, listen_count, title, artist;
PROGRESS:     To use one of these as a target column, set target = <column_name>
PROGRESS:     and use a method that allows the use of a target.
PROGRESS: Preparing data set.
PROGRESS:     Data has 893580 observations with 66085 users and 9952 items.
PROGRESS:     Data prepared in: 5.03742s
PROGRESS: 893580 observations to process; with 9952 unique items.

Use popularity model to make some predictions

popularity_model.recommend(users=[users[0]])

user_id	song	score	rank
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Sehr kosmisch - Harmonia	4754.0	1
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Undo - Björk	4227.0	2
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	You're The One - Dwight Yoakam ...	3781.0	3
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Revelry - Kings Of Leon	3527.0	4
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Horn Concerto No. 4 in E flat K495: II. Romance ...	3161.0	5
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Secrets - OneRepublic	3148.0	6
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Hey_ Soul Sister - Train	2538.0	7
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Fireflies - Charttraxx Karaoke ...	2532.0	8
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Tive Sim - Cartola	2521.0	9
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Drop The World - Lil Wayne / Eminem ...	2053.0	10

[10 rows x 4 columns]

popularity_model.recommend(users=[users[1]])

user_id	song	score	rank
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Sehr kosmisch - Harmonia	4754.0	1
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Undo - Björk	4227.0	2
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	You're The One - Dwight Yoakam ...	3781.0	3
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Dog Days Are Over (Radio Edit) - Florence + The ...	3633.0	4
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Revelry - Kings Of Leon	3527.0	5
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Horn Concerto No. 4 in E flat K495: II. Romance ...	3161.0	6
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Secrets - OneRepublic	3148.0	7
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Fireflies - Charttraxx Karaoke ...	2532.0	8
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Tive Sim - Cartola	2521.0	9
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Drop The World - Lil Wayne / Eminem ...	2053.0	10

[10 rows x 4 columns]

Build a song recommender with personalization

personalized_model = graphlab.item_similarity_recommender.create(train_data,
                                                                user_id='user_id',
                                                                item_id='song')

PROGRESS: Recsys training: model = item_similarity
PROGRESS: Warning: Ignoring columns song_id, listen_count, title, artist;
PROGRESS:     To use one of these as a target column, set target = <column_name>
PROGRESS:     and use a method that allows the use of a target.
PROGRESS: Preparing data set.
PROGRESS:     Data has 893580 observations with 66085 users and 9952 items.
PROGRESS:     Data prepared in: 5.44737s
PROGRESS: Computing item similarity statistics:
PROGRESS: Computing most similar items for 9952 items:
PROGRESS: +-----------------+-----------------+
PROGRESS: | Number of items | Elapsed Time    |
PROGRESS: +-----------------+-----------------+
PROGRESS: | 1000            | 2.97569         |
PROGRESS: | 2000            | 3.29291         |
PROGRESS: | 3000            | 3.59487         |
PROGRESS: | 4000            | 3.84412         |
PROGRESS: | 5000            | 4.10858         |
PROGRESS: | 6000            | 4.37963         |
PROGRESS: | 7000            | 4.6265          |
PROGRESS: | 8000            | 4.87737         |
PROGRESS: | 9000            | 5.09781         |
PROGRESS: +-----------------+-----------------+
PROGRESS: Finished training in 7.55179s

Applying personalized model to make song recommendations

personalized_model.recommend(users=[users[0]])

user_id	song	score	rank
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Fantasy - The xx	0.037720015587	1
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Walk In The Park - Beach House ...	0.02976011994	2
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Lover Of Mine - Beach House ...	0.0273156899811	3
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Used To Be - Beach House	0.0258316352834	4
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Norway - Beach House	0.0219293953923	5
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Take Care - Beach House	0.021579303806	6
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Paper Gangsta - Lady GaGa	0.0204163968481	7
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Beautiful_ Dirty_ Rich - Lady GaGa ...	0.0202445750201	8
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Real Love - Beach House	0.0195482216248	9
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Boys Boys Boys - Lady GaGa ...	0.019074731889	10

[10 rows x 4 columns]

personalized_model.recommend(users=[users[1]])

user_id	song	score	rank
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Cuando Pase El Temblor - Soda Stereo ...	0.0194504525792	1
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Fireflies - Charttraxx Karaoke ...	0.014473730789	2
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Love Is A Losing Game - Amy Winehouse ...	0.0142865986808	3
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Marry Me - Train	0.0141334715267	4
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Secrets - OneRepublic	0.0135916683588	5
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	No Dejes Que... - Caifanes ...	0.0134191754754	6
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Sehr kosmisch - Harmonia	0.0133987908186	7
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Y solo se me ocurre amarte (Unplugged) - ...	0.0133210385369	8
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Te Hacen Falta Vitaminas - Soda Stereo ...	0.0129302853556	9
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	OMG - Usher featuring will.i.am ...	0.0127778293142	10

[10 rows x 4 columns]

personalized_model.get_similar_items(['The Cove - Jack Johnson'])

PROGRESS: Getting similar items completed in 0.007283

song	similar	score	rank
The Cove - Jack Johnson	Moonshine - Jack Johnson	0.179487179487	1
The Cove - Jack Johnson	Holes To Heaven - Jack Johnson ...	0.111969111969	2
The Cove - Jack Johnson	Country Road - Jack Johnson / Paula Fuga ...	0.0839694656489	3
The Cove - Jack Johnson	Supposed To Be - Jack Johnson ...	0.0740740740741	4
The Cove - Jack Johnson	Let It Be Sung - Jack Johnson / Matt Costa / ...	0.0736842105263	5
The Cove - Jack Johnson	Wrong Turn - Jack Johnson	0.0709677419355	6
The Cove - Jack Johnson	Questions - Jack Johnson	0.0687022900763	7
The Cove - Jack Johnson	Rainbow - Jack Johnson / G. Love ...	0.0649350649351	8
The Cove - Jack Johnson	Posters - Jack Johnson	0.0593607305936	9
The Cove - Jack Johnson	If I Could - Jack Johnson	0.0588235294118	10

[10 rows x 4 columns]

song_data[0]

{'artist': 'Jack Johnson',
 'listen_count': 1,
 'song': 'The Cove - Jack Johnson',
 'song_id': 'SOAKIMP12A8C130995',
 'title': 'The Cove',
 'user_id': 'b80344d063b5ccb3212f76538f3d9e43d87dca9e'}

Quantitative Comparison between the models

%matplotlib inline
model_performance = graphlab.compare(test_data, [popularity_model, personalized_model], user_sample=0.05)

compare_models: using 2931 users to estimate model performance
PROGRESS: Evaluate model M0
PROGRESS: recommendations finished on 1000/2931 queries. users per second: 1379.31
PROGRESS: recommendations finished on 2000/2931 queries. users per second: 1413.19

Precision and recall summary statistics by cutoff
+--------+-----------------+------------------+
| cutoff |  mean_precision |   mean_recall    |
+--------+-----------------+------------------+
|   1    | 0.0276356192426 | 0.00704457697805 |
|   2    | 0.0274650290003 | 0.0148879287057  |
|   3    | 0.0262708973047 | 0.0220417769804  |
|   4    | 0.0242238143978 | 0.0261965554443  |
|   5    | 0.0219037871034 |  0.029571775887  |
|   6    | 0.0212669168657 | 0.0336708194948  |
|   7    | 0.0201783886533 | 0.0374884100319  |
|   8    | 0.0191487546912 | 0.0406333667849  |
|   9    | 0.0179309299064 | 0.0425615155912  |
|   10   | 0.0172296144661 | 0.0452822737367  |
+--------+-----------------+------------------+
[10 rows x 3 columns]

PROGRESS: Evaluate model M1
PROGRESS: recommendations finished on 1000/2931 queries. users per second: 636.026
PROGRESS: recommendations finished on 2000/2931 queries. users per second: 652.271

Precision and recall summary statistics by cutoff
+--------+-----------------+-----------------+
| cutoff |  mean_precision |   mean_recall   |
+--------+-----------------+-----------------+
|   1    |  0.196178778574 | 0.0624035480612 |
|   2    |  0.165301944729 | 0.0989251209952 |
|   3    |  0.144433071762 |  0.126315233442 |
|   4    |  0.128283862163 |  0.149273539588 |
|   5    |  0.115114295462 |  0.165037712581 |
|   6    |  0.105709086774 |  0.180392416447 |
|   7    | 0.0979675391139 |  0.193083833411 |
|   8    | 0.0907540088707 |  0.202181627936 |
|   9    | 0.0851813942909 |  0.212276173919 |
|   10   | 0.0802115319004 |  0.222459311226 |
+--------+-----------------+-----------------+
[10 rows x 3 columns]

Model compare metric: precision_recall

graphlab.show_comparison(model_performance,[popularity_model, personalized_model])

Counting unique users

users_kanye_west = song_data[song_data['artist'] == 'Kanye West' ]

users_kanye_west.head()

user_id	song_id	listen_count	title	artist
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOBXHDL12A81C204C0	1	Stronger	Kanye West
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOMLMKI12A81C204BC	1	Champion	Kanye West
5d5e0142e54c3bb7b69f548c2 ee55066c90700eb ...	SORFASW12A81C22AE7	2	Stronger	Kanye West
537340ff896dea11328910013 cfe759413e1eeb3 ...	SOBXHDL12A81C204C0	2	Stronger	Kanye West
7dd192c8bd4f27f573cb15e86 56442aadd7a9c01 ...	SOOLPFK12A58A7BDE3	5	Flashing Lights	Kanye West
8fce200f3912e9608e3b1463c db9c3529aab5c08 ...	SOBXHDL12A81C204C0	2	Stronger	Kanye West
8fce200f3912e9608e3b1463c db9c3529aab5c08 ...	SOIBSWV12A6D4F6AB3	1	Through The Wire	Kanye West
a56bf59af6edc5ae6c92d61dd d214989332864e8 ...	SONGNHO12AB0183915	1	Bad News	Kanye West
8fa25e588aeedaa539674babb 75729ac9f31f15e ...	SOOLPFK12A58A7BDE3	1	Flashing Lights	Kanye West
e8612acfb1572297ea0eaaa1f 27927d55fdcec65 ...	SOIYWPZ12A81C204EF	2	Homecoming	Kanye West

song
Stronger - Kanye West
Champion - Kanye West
Stronger - Kanye West
Stronger - Kanye West
Flashing Lights - Kanye West ...
Stronger - Kanye West
Through The Wire - Kanye West ...
Bad News - Kanye West
Flashing Lights - Kanye West ...
Homecoming - Kanye West

[10 rows x 6 columns]

len(users_kanye_west)

unique_users_kanye_west = users_kanye_west['user_id'].unique()

len(unique_users_kanye_west)

users_foo_fighters = song_data[song_data['artist'] == 'Foo Fighters' ]

len(users_foo_fighters)

users_foo_fighters.head()

user_id	song_id	listen_count	title	artist
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SODACBL12A8C13C273	1	Learn To Fly	Foo Fighters
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOFGUAY12AB017B0A8	1	Stacked Actors	Foo Fighters
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOMSQJY12A8C138539	1	Breakout	Foo Fighters
b80344d063b5ccb3212f76538 f3d9e43d87dca9e ...	SOVHRGF12A8C13852F	1	Generator	Foo Fighters
12768858f6a825452e412deb1 df36d2d1d9c6791 ...	SODACBL12A8C13C273	4	Learn To Fly	Foo Fighters
12768858f6a825452e412deb1 df36d2d1d9c6791 ...	SOQLUTQ12A8AE48037	2	The Pretender	Foo Fighters
f47116f998e030f2dab275b81 fb2a04a9dc06c33 ...	SOLTAEJ12A8C13F793	6	What If I Do?	Foo Fighters
5e161b9e14f303a0cef2d3f44 d07dd946549f89f ...	SOLJYEI12A8C13F7B3	2	Friend Of A Friend	Foo Fighters
e4c05157f8cebdf3b9d689c44 1ba97c5ed5db05b ...	SOFGUAY12AB017B0A8	1	Stacked Actors	Foo Fighters
e4c05157f8cebdf3b9d689c44 1ba97c5ed5db05b ...	SOFZVOT12A8C1408E9	1	Skin And Bones	Foo Fighters

song
Learn To Fly - Foo Fighters ...
Stacked Actors - Foo Fighters ...
Breakout - Foo Fighters
Generator - Foo Fighters
Learn To Fly - Foo Fighters ...
The Pretender - Foo Fighters ...
What If I Do? - Foo Fighters ...
Friend Of A Friend - Foo Fighters ...
Stacked Actors - Foo Fighters ...
Skin And Bones - Foo Fighters ...

[10 rows x 6 columns]

unique_users_foo_fighters = users_foo_fighters['user_id'].unique()

len(unique_users_foo_fighters)

users_taylor_swift = song_data[song_data['artist'] == 'Taylor Swift' ]

len(users_taylor_swift)

unique_users_taylor_swift = users_taylor_swift['user_id'].unique()

len(unique_users_taylor_swift)

users_lady_gaga = song_data[song_data['artist'] == 'Lady GaGa' ]

len(users_lady_gaga)

unique_users_lady_gaga = users_lady_gaga['user_id'].unique()

len(unique_users_lady_gaga)

Using groupby-aggregate to find the most popular and least popular artist

total_listen_count = song_data.groupby(key_columns='artist', operations={'total_count': graphlab.aggregate.SUM('listen_count')})

total_listen_count.head()

artist	total_count
The Dells	274
Tag Team	314
Danger Doom	561
Rata Blanca	714
The Pogues	141
Diana Ross	302
Kurtis Blow	239
Uniikki	429
Make the Girl Dance	378
Jorge Gonzalez	682

[10 rows x 2 columns]

total_listen_count.sort('total_count',ascending=False)

artist	total_count
Kings Of Leon	43218
Dwight Yoakam	40619
Björk	38889
Coldplay	35362
Florence + The Machine	33387
Justin Bieber	29715
Alliance Ethnik	26689
OneRepublic	25754
Train	25402
The Black Keys	22184

[3375 rows x 2 columns]
Note: Only the head of the SFrame is printed.
You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.

sorted_total_listen_count =  total_listen_count.sort('total_count',ascending=False)

sorted_total_listen_count.head()

artist	total_count
Kings Of Leon	43218
Dwight Yoakam	40619
Björk	38889
Coldplay	35362
Florence + The Machine	33387
Justin Bieber	29715
Alliance Ethnik	26689
OneRepublic	25754
Train	25402
The Black Keys	22184

[10 rows x 2 columns]

sorted_total_listen_count[0]

{'artist': 'Kings Of Leon', 'total_count': 43218}

sorted_total_listen_count[-1]

{'artist': 'William Tabbert', 'total_count': 14}

sorted_total_listen_count[-2]

{'artist': 'Reel Feelings', 'total_count': 24}

len(sorted_total_listen_count)

sorted_total_listen_count[3374]

{'artist': 'William Tabbert', 'total_count': 14}

Using groupby-aggregate to find the most recommended songs

train_data,test_data = song_data.random_split(.8,seed=0)

personalized_model_problem = graphlab.item_similarity_recommender.create(train_data,
                                                                user_id='user_id',
                                                                item_id='song')

PROGRESS: Recsys training: model = item_similarity
PROGRESS: Warning: Ignoring columns song_id, listen_count, title, artist;
PROGRESS:     To use one of these as a target column, set target = <column_name>
PROGRESS:     and use a method that allows the use of a target.
PROGRESS: Preparing data set.
PROGRESS:     Data has 893580 observations with 66085 users and 9952 items.
PROGRESS:     Data prepared in: 2.44033s
PROGRESS: Computing item similarity statistics:
PROGRESS: Computing most similar items for 9952 items:
PROGRESS: +-----------------+-----------------+
PROGRESS: | Number of items | Elapsed Time    |
PROGRESS: +-----------------+-----------------+
PROGRESS: | 1000            | 1.09591         |
PROGRESS: | 2000            | 1.2529          |
PROGRESS: | 3000            | 1.39995         |
PROGRESS: | 4000            | 1.54215         |
PROGRESS: | 5000            | 1.67863         |
PROGRESS: | 6000            | 1.80773         |
PROGRESS: | 7000            | 1.9314          |
PROGRESS: | 8000            | 2.04903         |
PROGRESS: | 9000            | 2.15834         |
PROGRESS: +-----------------+-----------------+
PROGRESS: Finished training in 2.50746s

subset_test_users = test_data['user_id'].unique()[0:10000]

len(subset_test_users)

recommended_songs = personalized_model_problem.recommend(subset_test_users,k=1)

PROGRESS: recommendations finished on 1000/10000 queries. users per second: 723.767
PROGRESS: recommendations finished on 2000/10000 queries. users per second: 729.502
PROGRESS: recommendations finished on 3000/10000 queries. users per second: 730.974
PROGRESS: recommendations finished on 4000/10000 queries. users per second: 732.245
PROGRESS: recommendations finished on 5000/10000 queries. users per second: 730.881
PROGRESS: recommendations finished on 6000/10000 queries. users per second: 729.537
PROGRESS: recommendations finished on 7000/10000 queries. users per second: 729.793
PROGRESS: recommendations finished on 8000/10000 queries. users per second: 731.585
PROGRESS: recommendations finished on 9000/10000 queries. users per second: 731.684
PROGRESS: recommendations finished on 10000/10000 queries. users per second: 732.331

recommended_songs.head()

user_id	song	score	rank
b048033af070b5dbb18d5d0e5 f334c9390611b04 ...	Fantasy - The xx	0.037720015587	1
c66c10a9567f0d82ff31441a9 fd5063e5cd9dfe8 ...	Cuando Pase El Temblor - Soda Stereo ...	0.0194504525792	1
ed04954d5b6001c7945c6ac71 686c3bd4ecdacb3 ...	Coming Your Way - Iration	0.031314214241	1
b1e6e9563da324641e644c769 b7edf202186de47 ...	Pimpa's Paradise - Damian Marley / Stephen Marl ...	0.0694444444444	1
02f015d32ac2cd1e52d26e3ec 36048711dd5711b ...	Where The Boat Leaves From (Album) - Zac Brown ...	0.063530766032	1
91b986eeb5d81eec60dc4b136 f04c0cfd662d658 ...	Jezebel - Sade	0.0588785769489	1
f933855d675606737fdc191e9 edff7625d08aae8 ...	Schießt die Deutschen raus - Mario Lang ...	0.0357374917866	1
4867d5516a280db13695b9b9c 7ce6b574f34c6b4 ...	Two Steps_ Twice - Foals	0.0104654895666	1
968f1baebc490d3c6999ee6c8 5c5cab8b726b347 ...	Me_ Myself And I - Beyoncé ...	0.0183084820675	1
c067c22072a17d33310d7223d 7b79f819e48cf42 ...	Grind With Me (Explicit Version) - Pretty Ricky ...	0.0459424433009	1

[10 rows x 4 columns]

song_recommended_count =  recommended_songs.groupby(key_columns='song', operations={'count': graphlab.aggregate.COUNT()})

song_recommended_count.head()

song	count
Wrong Turn - Jack Johnson	9
Welcome Home (Sanitarium) - Metallica ...	4
Wake Up Call - Maroon 5	2
The Pyramids - Tim Wilson	1
U.R.A Fever - The Kills	3
Slow Down - Alicia Keys	1
If This Is Love - The Saturdays ...	1
Marshall Mathers - Eminem	1
Electricity - Moby	1
Hoodoo - Muse	3

[10 rows x 2 columns]

song_recommended_count[-1]

{'count': 2, 'song': 'Amame - Juanes'}

sorted_song_recommended_count = song_recommended_count.sort('count',ascending=False)

sorted_song_recommended_count.head()

song	count
Undo - Björk	447
Secrets - OneRepublic	373
Revelry - Kings Of Leon	228
You're The One - Dwight Yoakam ...	179
Fireflies - Charttraxx Karaoke ...	124
Hey_ Soul Sister - Train	117
Horn Concerto No. 4 in E flat K495: II. Romance ...	95
OMG - Usher featuring will.i.am ...	68
Sehr kosmisch - Harmonia	66
Dog Days Are Over (Radio Edit) - Florence + The ...	52

[10 rows x 2 columns]

Credits Machine Learning Foundations: A Case Study Approach