Knowledge Transfer between Computer Vision and Text Mining : Similarity-based Learning Approaches

This ground-breaking text/reference divergesfrom the traditional view that computer vision (for image analysis) and stringprocessing (for text mining) are separate and unrelated fields of study,propounding that images and text can be treated in a similar manner for thepurposes of information retrieval, extraction and classification.

This ground-breaking text/reference divergesfrom the traditional view that computer vision (for image analysis) and stringprocessing (for text mining) are separate and unrelated fields of study,propounding that images and text can be treated in a similar manner for thepurposes of information retrieval, extraction and classification. Highlightingthe benefits of knowledge transfer between the two disciplines, the textpresents a range of novel similarity-based learning (SBL) techniques founded onthis approach. Topics and features: describes a variety of SBL approaches,including nearest neighbor models, local learning, kernel methods, andclustering algorithms; presents a nearest neighbor model based on a noveldissimilarity for images; discusses a novel kernel for (visual) wordhistograms, as well as several kernels based on a pyramid representation; introducesan approach based on string kernels for native language identification; containslinks for downloading relevant open source code.