Advanced

Att klassificera med Support Vector Machines - En introduktion från teori till analys

Milton, Alexandra LU and Svensson, Marcus LU (2018) STAH11 20162
Department of Statistics
Abstract
The purpose of this paper is to give a short introduction to support vector machines. The paper intends to cover the full process from theory to analysis in the binary classification case. The analysis intends to yield an understanding of how the method can be used in practice, and how models can be fitted in a way that maximizes the desired performance measurement.

This paper introduces the theory behind classification in the most basic case with perfect linear separability. Subsequently, methods to expand the model to manage more realistic scenarios of non-linear separability are introduced. Furthermore, the model selection process is detailed, and common performance measurements are proposed.

The dataset, Wisconsin Diagnostic... (More)
The purpose of this paper is to give a short introduction to support vector machines. The paper intends to cover the full process from theory to analysis in the binary classification case. The analysis intends to yield an understanding of how the method can be used in practice, and how models can be fitted in a way that maximizes the desired performance measurement.

This paper introduces the theory behind classification in the most basic case with perfect linear separability. Subsequently, methods to expand the model to manage more realistic scenarios of non-linear separability are introduced. Furthermore, the model selection process is detailed, and common performance measurements are proposed.

The dataset, Wisconsin Diagnostic Breast Cancer, is used for the analysis. Numerous models are trained to predict tumors as malignant or benign. The final model performs in line with predetermined requirements, correctly classifying as many malignant tumors as possible. The final model attains a sensitivity of 0.9836, which equals one false negative prediction. (Less)
Abstract (Swedish)
Syftet med uppsatsen är att ge en kort introduktion till support vector machines. Uppsatsen ämnar täcka hela processen från teori till slutgiltig analys vid binär klassificering. Analysen ämnar ge en förståelse för hur metoden kan användas i praktiken och hur modeller kan anpassas efter bestämda önskemål med hjälp av olika prestandamått.

Uppsatsen introducerar teorin bakom hur perfekt linjärt separabla klasser behandlas samt metoder för att utveckla modellen till att hantera mer realistiska scenarion. Vidare presenteras tillvägagångssätt till modellanpassning samt prestandamått.

Datamaterialet, Wisconsin Diagnostic Breast Cancer, används för analys. Modeller anpassas i syfte att prediktera tumörer som elakartade eller godartade.... (More)
Syftet med uppsatsen är att ge en kort introduktion till support vector machines. Uppsatsen ämnar täcka hela processen från teori till slutgiltig analys vid binär klassificering. Analysen ämnar ge en förståelse för hur metoden kan användas i praktiken och hur modeller kan anpassas efter bestämda önskemål med hjälp av olika prestandamått.

Uppsatsen introducerar teorin bakom hur perfekt linjärt separabla klasser behandlas samt metoder för att utveckla modellen till att hantera mer realistiska scenarion. Vidare presenteras tillvägagångssätt till modellanpassning samt prestandamått.

Datamaterialet, Wisconsin Diagnostic Breast Cancer, används för analys. Modeller anpassas i syfte att prediktera tumörer som elakartade eller godartade. Den slutgiltiga modellen presterar i linje med förutbestämda önskemål, att korrekt klassificera så många elakartade tumörer som möjligt. Den slutgiltiga modellen uppnår en sensitivitet på 0.9836 vilket motsvarar en elakartad tumör som felaktigt klassificerats som godartad. (Less)
Please use this url to cite or link to this publication:
author
Milton, Alexandra LU and Svensson, Marcus LU
supervisor
organization
course
STAH11 20162
year
type
M2 - Bachelor Degree
subject
keywords
Support Vector Machines, Klassificering, Maskininlärning
language
Swedish
id
8953174
date added to LUP
2018-09-03 10:51:19
date last changed
2018-09-03 10:51:19
@misc{8953174,
  abstract     = {The purpose of this paper is to give a short introduction to support vector machines. The paper intends to cover the full process from theory to analysis in the binary classification case. The analysis intends to yield an understanding of how the method can be used in practice, and how models can be fitted in a way that maximizes the desired performance measurement.

This paper introduces the theory behind classification in the most basic case with perfect linear separability. Subsequently, methods to expand the model to manage more realistic scenarios of non-linear separability are introduced. Furthermore, the model selection process is detailed, and common performance measurements are proposed. 

The dataset, Wisconsin Diagnostic Breast Cancer, is used for the analysis. Numerous models are trained to predict tumors as malignant or benign. The final model performs in line with predetermined requirements, correctly classifying as many malignant tumors as possible. The final model attains a sensitivity of 0.9836, which equals one false negative prediction.},
  author       = {Milton, Alexandra and Svensson, Marcus},
  keyword      = {Support Vector Machines,Klassificering,Maskininlärning},
  language     = {swe},
  note         = {Student Paper},
  title        = {Att klassificera med Support Vector Machines - En introduktion från teori till analys},
  year         = {2018},
}