Method and system for validating promotional emails and product availability from E-commerce websites

A system for validating promotional emails and product availability from e-commerce websites, the system comprising:

• a processor and a memory communicatively coupled to the processor, wherein the memory stores processor-executable instructions, which, on execution, causes the processor to:
• retrieve a first set of images corresponding to an image strip and a second set of images corresponding to a promotional email from a database, wherein the first set of images and the second set of images are associated with one or more products;
• calculate a similarity score between each of the first set of images and each of the second set of images using a first Computer Vision (CV) technique;
• select one or more valid images from the second set of images based on the similarity score;
• and
• determine a stock availability status of at least one product presented in the one or more valid images from at least one website using a deep learning algorithm, wherein each of the at least one product is a part of the one or more products, wherein determining a stock availability status of at least one product comprises:
• classifying the one or more valid images into a set of predefined categories based on content of each of the one or more valid images through a deep learning classification model;
• for each of the one or more valid images,
• determining whether identification of the at least one product in a valid image is required based on a predefined category associated with the valid image;
• and
• at least one of:
• directly matching the valid image with one or more images of the at least one website using a second CV technique when the identification of the at least one product in the valid image is not required;
• or
• detecting the at least one product in the valid image using a Faster Region-based Convolutional Neural Network (Faster RCNN) model to provide a segmented product image when the identification of the at least one product in the valid image is required.

A computer program product for validating promotional emails and product availability from e-commerce websites, the computer program product embodied in a non-transitory computer readable storage medium and comprising computer-executable instructions configured for:

• retrieving a first set of images corresponding to an image strip and a second set of images corresponding to a promotional email from a database, wherein the first set of images and the second set of images are associated with one or more products;
• calculating a similarity score between each of the first set of images and each of the second set of images using a first Computer Vision (CV) technique;
• selecting one or more valid images from the second set of images based on the similarity score;
• and
• determining a stock availability status of at least one product presented in the one or more valid images from at least one website using a deep learning algorithm, wherein each of the at least one product is a part of the one or more products, wherein determining a stock availability status of at least one product comprises:
• classifying the one or more valid images into a set of predefined categories based on content of each of the one or more valid images through a deep learning classification model;
• for each of the one or more valid images,
• determining whether identification of the at least one product in a valid image is required based on a predefined category associated with the valid image;
• and
• at least one of:
• directly matching the valid image with one or more images of the at least one website using a second CV technique when the identification of the at least one product in the valid image is not required;
• or
• detecting the at least one product in the valid image using a Faster Region-based Convolutional Neural Network (Faster RCNN) model to provide a segmented product image when the identification of the at least one product in the valid image is required.