Model 1

Breast Cancer Wisconsin (Diagnostic) Prediction


This predictive model, employing the powerful XGBoost (Extreme Gradient Boosting) algorithm, is based on the UCI ML Breast Cancer Wisconsin (Diagnostic) dataset.

  • W.N. Street, W.H. Wolberg and O.L. Mangasarian. Nuclear feature extraction    for breast tumor diagnosis. IS&T/SPIE 1993 International Symposium on    Electronic Imaging: Science and Technology, volume 1905, pages 861-870,   San Jose, CA, 1993.
  • O.L. Mangasarian, W.N. Street and W.H. Wolberg. Breast cancer diagnosis and prognosis via linear programming. Operations Research, 43(4), pages 570-577, July-August 1995.
  • W.H. Wolberg, W.N. Street, and O.L. Mangasarian. Machine learning techniquesto diagnose breast cancer from fine-needle aspirates. Cancer Letters 77 (1994) 163-171.

This predictive model, employing the powerful XGBoost (ExtremeGradient Boosting) algorithm, is based on the UCI ML Breast Cancer Wisconsin(Diagnostic) dataset. The dataset comprises features computed from a digitizedimage of a fine needle aspirate (FNA) of a breast mass, describingcharacteristics of the cell nuclei present in the image.

To try this model, select Model 1 in the Privasea client and copy your feature vector as the X vector. The client will then encrypt thisvector on your local machine using your locally stored client key, and theencrypted data will be transmitted to the Privanetix nodes. Subsequentinference operations will take place in the encrypted domain on the Privanetixnodes, and the encrypted result will be sent back to your client. Upon arrival,the result will be decrypted using your locally stored client key. Thedecrypted result will yield a two-dimensional vector representing thelikelihood of the associated tags. For instance, a result like "011,0.89" signifies a classification as true. You can compare it with theprovided output in the tables.

Double click on the cell > CMD or CTRL + C to copy data

Features vector:


   mean radius

   mean texture

   mean perimeter

   mean area

   mean smoothness

   mean compactness

   mean concavity

   mean concave points

  mean symmetry

  mean fractal dimension

  radius error

  texture error

  perimeter error

  area error

  smoothness error

  compactness error

  concavity error

  concave points error

  symmetry error

  fractal dimension error

  worst radius

  worst texture

  worst perimeter

  worst area

  worst smoothness

  worst compactness

  worst concavity

  worst concave points

  worst symmetry

  worst fractal dimension





  0,  false

  1,  true