Integrating Logical Inference Into Statistical Text Classification Applications

Contemporary statistical text classification is becoming increasingly common across a wide range of everyday applications. Typically, the bottlenecks in performance are the availability and consistency of large amounts of training data. We argue that these techniques could be improved by seamlessly integrating logical inference into the text encoding pipeline, making it possible to utilize large-scale commonsense and special-purpose knowledge bases to aid in the interpretation and encoding of documents. Statistical text classification is good, not great Contemporary statistical text classification is becoming increasingly common across a wide range of everyday applications, used in cases where a label must be assigned to textual material based on changing or idiosyncrati