Recent Patents on Engineering    Send Orders for Reprints to reprints@benthamscience.net Recent Patents on Engineering, 2022, 16, e150721194823 RESEARCH ARTICLE Convolutional Neural Network Based Intelligent Advertisement Search Framework for Online English Newspapers 109 Pooja Jain 1,* , Kavita Taneja 1 and Harmunish Taneja 2 1 Department of Computer Science and Applications, Punjab University, Chandigarh 160014, Punjab, India; 2 Department of Computer Science and Information Technology, DAV College, Chandigarh, Punjab, India Abstract: Background: Instant access to desired information is the key element for building an in- telligent environment creating value for people and steering towards society 5.0. Online newspapers are one such example which provide instant access to information anywhere and anytime on our mobiles, tablets, laptops, desktops, etc. But when it comes to searching for a specific advertisement in newspapers, online newspapers do not provide easy advertisement search options. Also, there are no specialized search portals which can provide for keyword-based advertisement search across multiple online newspapers. As a result, to find a specific advertisement in multiple newspapers, a sequential manual search is required across a range of online newspapers. Objective: This research paper proposes a keyword-based advertisement search framework to pro- vide an instant access to the relevant advertisements from online English newspapers in a category of reader’s choice. Methods: First, an image extraction algorithm is proposed which can identify and extract the images from online newspapers without using any rules on advertisement placement and/or size. It is fol- lowed by a proposed deep learning Convolutional Neural Network (CNN) model named ‘Adv_Recognizer’ which is used to separate the advertisement images from non-advertisement im- ages. Another CNN Model, ‘Adv_Classifier’, is proposed, which classifies the advertisement imag- es into four pre-defined categories. Finally, Optical Character Recognition (OCR) technique is used to perform keyword-based advertisement searches in various categories across multiple newspapers. Results: The proposed image extraction algorithm can easily extract all types of well-bounded im- ages from different online newspapers and this algorithm is used to create ‘English newspaper im- age dataset’ of 11,000 images, including advertisements and non-advertisements. The proposed ‘Adv_Recognizer’ model separates advertisement and non-advertisement images with an accuracy of around 97.8%. and the proposed ‘Adv_Classifier’ model classifies the advertisements in four pre- defined categories exhibiting an accuracy of around 73.5%. Conclusion: The proposed framework will help newspaper readers in performing exhaustive adver- tisement searches across a range of online English newspapers in a category of their own interest. It will also help in carrying out advertisement analysis and studies. A R T I C L E H I S T O R Y Received: January 31, 2021 Revised: April 02, 2021 Accepted: May 18, 2021 DOI: 10.2174/1872212115666210715163919 Keywords: Advertisement image classification, convolutional neural networks (CNN), newspaper advertisements, newspaper layout segmentation, optical character recognition (OCR), residual networks (ResNet), transfer learning. 1. INTRODUCTION India is one of the largest newspaper markets in the world, with private companies, government departments, recruitment agencies, educational institutes, etc. using news- paper advertisements as a primary source for advertising jobs, tenders, admission notices, sales, promotions, etc. Due *Address correspondence to this author at the Department of Computer Science and Applications, Panjab University, Chandigarh 160014, Punjab, India; E-mail: poojajain9199@gmail.com to the rapid increase in the internet users every year, the trend has shifted from traditional newspaper reading to online newspaper reading. Moreover, home-locked situations during unprecedented circumstances like pandemics which restrict the access to the printed copy of newspapers, have further boosted this trend many folds. Whether it is a tradi- tional newspaper reader or an online newspaper reader, search for relevant newspaper advertisements is very crucial for people waiting for specific advertisements to be out in the newspapers. Interested students, job seekers, contractors, 2212-4047/22 $65.00+.00 © 2022 Bentham Science Publishers