SuperTML: Two-Dimensional Word Embedding for the Precognition on Structured Tabular Data
This addresses classification tasks for tabular data in industry, representing an incremental advance by adapting a text-based method to a new domain.
The paper tackled classification on tabular data by proposing SuperTML, a method that projects features into two-dimensional embeddings and uses fine-tuned CNNs, achieving state-of-the-art results on both large and small datasets.
Tabular data is the most commonly used form of data in industry. Gradient Boosting Trees, Support Vector Machine, Random Forest, and Logistic Regression are typically used for classification tasks on tabular data. DNN models using categorical embeddings are also applied in this task, but all attempts thus far have used one-dimensional embeddings. The recent work of Super Characters method using two-dimensional word embeddings achieved the state of art result in text classification tasks, showcasing the promise of this new approach. In this paper, we propose the SuperTML method, which borrows the idea of Super Characters method and two-dimensional embeddings to address the problem of classification on tabular data. For each input of tabular data, the features are first projected into two-dimensional embeddings like an image, and then this image is fed into fine-tuned two-dimensional CNN models for classification. Experimental results have shown that the proposed SuperTML method had achieved state-of-the-art results on both large and small datasets.