IMAGE SIMILARITY USING LOGISTIC REGRESSION

IMAGE SIMILARITY USING LOGISTIC REGRESSION

 Ch Mukitha 1, D Sravan Kumar 2, G Aravind 3,Swapna4

UG Scholar, Guru Nanak Institutions Technical Campus, Hyderabad1,2

Assistant Professor, Guru Nanak Institutions Technical Campus, Hyderabad 3

ABSTRACT:
Many learning algorithms such as kernel machines, nearest neighbors, clustering, or anomaly detection, are based on distances or similarities. Before similarities are used for training an actual machine learning model, we would like to verify that they are bound to meaningful patterns in the data. In this paper, we propose to make similarities interpretable by augmenting them with an explanation. We develop Logistic Regression & Haar cascade, a scalable and theoretically founded method to systematically decompose the output of an already trained deep similarity model on pairs of input features. Our method can be expressed as a composition of regression explanations, which were shown in previous works to scale to highly nonlinear models. Through an extensive set of experiments, we demonstrate that Logistic Regression robustly explains complex similarity models. Additionally, we apply our method to an open problem in digital humanities: detailed assessment of similarity between historical documents, such as astronomical tables. Here again, Logistic Regression & Haar cascade provides insight and brings verifiability into a highly engineered and problem-specific similarity model.