Word Representation for Biomedical Textmining based on NNGM with Image to Text Conversion

For capturing semantic regularities some of the distributed word representations are succeeded, but most of them are shallow window based model. To represent deeper information propose NNGM with image to text by using the concept of OCR (Optical character recognition). OCR is the mechanical or electronic conversion of images in to text format. This method considers dependency relation and context relation and clearly expresses the semantic regularity to emerge in word relation and represent the words using deeper information. In our method, performance measured on Protein-Protein Interaction Extraction and Word analogy task. Compare to other method, our word representation method perform better.