A linear programming solution to data description and novelty classification

Abstract

Many real-world problems require the detection of abnormal instances of a physical process, and methods inspired by the Support Vector Machines have been developed that model reference or normal data well. These methods serve as a fundamental step to enable the classification of new data as normal or abnormal. They imply the solution of a quadratic programming problem, which can present difficulties in finding solutions with standard methods and program solvers when the number of points becomes large. In this paper, we present an approach that was developed in a different context and that leads to a linear programming problem to attain the computational advantages of a linear environment. Normal 0 14 false false false IT X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:Tabella normale; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:Calibri,sans-serif; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:Times New Roman; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:Times New Roman; mso-bidi-theme-font:minor-bidi;}