Identification of Travel Styles by Learning from Consumer-generated Images in Online Travel Communities

A picture is worth a thousand words: Never has this adage been more meaningful than it is today. Online social media is driving the growth of unstructured image data. Unstructured data must be structured to be informative and thereby contribute to user understanding and revenue generation. Hitherto, companies have only been able to accomplish this through tedious manual work. This paper demonstrates how image data can be analyzed automatically using a combination of image analysis methods and fuzzy cluster algorithms to predict user preferences, which companies can then use to make targeted offers. Several methods, including support vector machines (SVMs) and convolutional neural networks (CNNs), are benchmarked across various cases of image data taken from an online travel community. Depending on the images' diversity either a SVM or a CNN provides the best basis for preference prediction.

Automated summary

This paper explores how image data can be automatically analyzed using a combination of image analysis methods and fuzzy cluster algorithms to predict user preferences. Depending on the diversity of the images, either SVM or CNN provide the best basis for preference prediction.