ML Webservices
Title Run Kmeans
Method The request type


Data Params Header:
Content-Type = multipart/form-data; boundary=---------------------------11111111111111

idColumns = The column title of the column you want to define as Identity column
featureColumns = List of the columns you wish to include for analysis. Please note that only columns containing numeric values will be processed.
clusterSize = Number of cluster you wish to define.
algorithm = kmeans. There will be further options at a later point.
file = A Comma separated CSV file containing all your data, including text columns

Example: source Kaggle fifa 2018
Content-Disposition: form-data; name="idColumns"

Content-Disposition: form-data; name="featureColumns"

Content-Disposition: form-data; name="clusterSize"

Content-Disposition: form-data; name="algorithm"

Content-Disposition: form-data; name="file"; filename="_.csv"
Content-Type: application/

C. Ronaldo dos Santos Aveiro,Real Madrid CF,32,Spanish Primera División,185,80,Portugal,565000,94,94
Lionel Messi,FC Barcelona,30,Spanish Primera División,170,72,Argentina,565000,93,93
Neymar da Silva Santos Jr.,Paris Saint-Germain,25,French Ligue 1,175,68,Brazil,280000,92,94
Luis Suárez,FC Barcelona,30,Spanish Primera División,182,86,Uruguay,510000,92,92
Manuel Neuer,FC Bayern Munich,31,German Bundesliga,193,92,Germany,230000,92,92
Robert Lewandowski,FC Bayern Munich,28,German Bundesliga,185,79,Poland,355000,91,91
David De Gea Quintana,Manchester United,26,English Premier League,193,76,Spain,215000,90,92
Eden Hazard,Chelsea,26,English Premier League,173,76,Belgium,295000,90,91
Toni Kroos,Real Madrid CF,27,Spanish Primera División,182,78,Germany,340000,90,90
Gonzalo Higuaín,Juventus,29,Italian Serie A,184,87,Argentina,275000,90,90

Success Response Code: 200
Content: the response is Json formatted
prediction = id of the prediction. If you define x clusters in the “clusterSize” field of the Post request, the response will contain x predictions ranging from 1 to x.
amount = number of data points included in the current prediction
percentage_outlier = percentage of data points identified as outliers in the current prediction
is_outlier = 1 if the data point has been identified as outlier, 0 otherwise
amount_outlier = number of data points identified as outliers in the current prediction

"cluster": [
"prediction": 1,
"amount": 7,
"percentage_outlier": 0,
"data_points": [
"full_name": "Neymar da Silva Santos Jr.",
. …
"is_outlier": 0
"full_name": "Manuel Neuer",

"is_outlier": 0
}, …
"amount_outlier": 0
"prediction": 2,
"amount": 3,
"percentage_outlier": 0,
"data_points": [
"full_name": "C. Ronaldo dos Santos Aveiro",


Error Response Cluster problems:
Code 200:
Content : {}

Wrongly formatted Post request:
Missing field:
"timestamp": "…",
"status": 400,
"error": "Bad Request",
"message": "Required String parameter '…' is not present",
"path": "/dabai-demo/api/job"

Missing/wrong header:
"timestamp": "2018-04-26T12:16:14.212+0000",
"status": 415,
"error": "Unsupported Media Type",
"message": "Content type 'text/plain' not supported",
"path": "/dabai-demo/api/job"
Notes Contact Visma for further information