Information Retrieval from web with Faster R-CNN Deep Learning Networks: A New Perspective

With advancement in artificial intelligence and machine learning, web information retrieval has evolved from manual to automated to intelligent systems making significant strides in building tools and techniques to enable automated information retrieval from the websites. In this paper, an automated web information retrieval framework with Faster region based convolutional neural network (R-CNN) is proposed. The framework automatically navigates to the specific product page in a retail website and implements leading edge open-source tools for automated retrieval of product details from the retail website. Information retrieval by using partial tree alignment (DEPTA) machine learning technique is used as reference for comparison. The results reveal an improved precision and recall as compared to DEPTA. The future research in web information retrieval space will to design and develop self-healing capability of web information retrieval framework to adapt to dynamic changes in the website layout.