Using Support Vector Machines for Survey Research

摘要: Recent developments in machine learning allow for flexible functional form estimation beyond the approaches typically used by survey researchers and social scientists. Support vector machines (SVMs) are one such technique, commonly binary classification problems, as whether or not an individual decides to participate a survey. Since their inception, SVMs have been extended solve categorical regression problems. Their versatility combination with fact that they perform well presence of large number predictors, even small cases, makes them very appealing wide range including character recognition text classification, speech speaker verification, imputation problems record linkage. In this article, we provide non-technical introduction main concepts SVMs, discuss advantages disadvantages, present ideas how can be research, and, finally, hands-on example, code, research results compare traditional logistic regression.