In the field of numismatics, classifying ancient coins, especially those that have diverse information and cultural heritage is a difficult task. Machine learning algorithms have recently made remarkable advancements in these types of tasks. However, these algorithms largely rely on relevant datasets. This article presents a novel dataset of ancient Gupta archer-type coin images, collected from verified private collections and three popular auction houses with their permission. The images exclusively comprise authentic specimens of ancient Gupta archer-type coins. We aim to establish a reliable resource that adheres to the highest standards of numismatic research. These coins, characterized by their distinctive archer motifs, present a significant challenge in terms of identification due to their scarcity and the intricate nature of their design. To address this, we meticulously curated a dataset by annotating each coin through a combination of visual examination and leveraging insights from numismatic literatures. These coins inherit ancient Indian archaeological insights, and studying these coins could provide insights into ancient Indian archaeology.

The number of people suffering from severe de-pression has risen in recent years. The majority of patients are apprehensive about seeking counseling and are unwilling to open up. A chat bot might be a viable tool for involving customers in artificial intelligence-powered behavioral health therapies. Chat bots are artificial intelligence entities that answer to users in normal language, exactly like a person would. Social chat bots, in particular, are those that form a deep emotional bond with the user. We shall explore and compare such chat bots in this paper, as they play an important role in assisting patients with mental illness. The study will compare and contrast chat bots such as CARO, XiaoIce, DEPRA, PRERONA, and Eviebot, as well as their role in resolving the depression problem. The article will show how the different chat bots compare in terms of methodology, underlying algorithms, accuracy, population demographics, and limitations. Finally, the paper will provide a quick overview of chat bots' future advancements in this field. The therapeutic component, which determines a person's level of depression,l is also a priority.

Character recognition is the numerical conversion of images in typed, handwritten, or printed format which allows a computer to recognize them. Bangla is one of the most complex languages as it has so many characters and digits. Moreover, the Bangla language has about 300 composite characters. That is why the extraction of characters from images is more di cult for Bangla compared to other languages. Deep learning has recently developed good capabilities for extracting high-level features from an image kernel. These systems learn more accurate and inclusive features from large-scale training datasets than earlier feature extraction techniques. This paper introduces a custom deep learning model to recognize handwritten Bangla characters and compares it with popular deep learning models that recognize handwritten characters. BanglaLekha Isolated dataset has been used to train and compare these models. Our proposed model KDANet was trained on 72,500 images containing primary characters and obtained an accuracy of 98.10% on the BanglaLekha Isolated dataset with 98.12% f1-score.