Supriyo Mandal

Supriyo Mandal
ZBW - Leibniz Information Centre for Economics

Doctor of Philosophy

About

14
Publications
2,421
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
48
Citations
Introduction
My research interest lies in Social Network and Complex Network analysis with Machine learning and Deep learning approach. I am always inquisitive about new technologies in the field of Machine Learning and Deep Learning. Currently, I am working on product-based social recommendation. I like to work with the sparsity nature of dataset.
Additional affiliations
August 2013 - July 2015
Indian Institute of Engineering Science and Technology, Shibpur
Position
  • M.tech
Description
  • My thesis title: Algorithm Design for 3D ICs Testing under Thermal Constraints. I proposed two temperature estimation models for 3D IC based on liquid cooled micro-channels and thermal TSVs. Advisor : Dr. Surajit Kumar Roy, Associate Professor, Department of Information Technology, Indian Institute of Engineering Science and Technology, Shibpur, India.
September 2012 - July 2013
Infosys
Position
  • Software Engineer
Education
January 2017 - December 2021

Publications

Publications (14)
Thesis
Full-text available
In various domains like E-commerce platforms, online news, online movie sites, etc., recommender systems perform an important role. Initially, researchers focused only on users’ previous rating behavior activities to understand their preferences. However, only rating behavior activities are not enough for rating prediction for a new item. Consequen...
Article
E-commerce companies want to predict their future product sales from the current customers’ feedback to frame a better business strategy. However, the conventional way of analyzing rating activities or quality and sentiment of reviews, volume of sales, or product prices is not enough for establishing a strong regression between these parameters and...
Article
Full-text available
Most of the existing recommender systems understand the preference level of users based on user-item interaction ratings. Rating-based recommendation systems mostly ignore negative users/reviewers (who give poor ratings). There are two types of negative users. Some negative users give negative or poor ratings randomly, and some negative users give...
Article
Full-text available
Recommendation systems usually make a personalized recommendation with explicit feedback (i.e., ratings, reviews, and description on products) or implicit feedback (i.e., searching activity, clicking products, and viewing products). Implicit feedback indicates a customer's preferences, and explicit feedback indicates the satisfaction level from the...
Conference Paper
Full-text available
In the current research, Graph Neural Networks (GNNs) play a decisive role in learning from the network data structure. In a social recommender system, GNNs have a significant perspective to integrate the structure of a customer-customer social network and the customer-product bipartite network. Most of the existing trust-based social recommendatio...
Article
Full-text available
Recommender Systems focus on implicit and explicit feedback or parameters of users for better rating prediction. Most of the existing recommender systems use only one type of feedback ignoring the other one. Based on the availability of resources, we may consider more number of feedback of both the types to predict user’s rating for a particular it...
Preprint
Full-text available
Traditional Collaborative Filtering (CF) based methods are applied to understand the personal preferences of users/customers for items/products from rating matrix. Usually, rating matrix is sparse in nature. So there are some improved variants of CF method that apply the increasing amount of side information to handle the sparsity problem. Only lin...
Patent
Full-text available
The present disclosure relates to system(s) and method(s) for determining company performance. The system receives social networking data from a set of social networking platforms and product review data from a set of e-commerce platform and the set of social networking platforms. The product review data may correspond to reviews or ratings and fee...
Chapter
Full-text available
Recommender systems recommend items more accurately by analyzing users’ potential interest on different brands’ items. In conjunction with users’ rating similarity, the presence of users’ implicit feedbacks like clicking items, viewing items specifications, watching videos etc. have been proved to be helpful for learning users’ embedding, that help...
Preprint
Recommender systems recommend items more accurately by analyzing users' potential interest on different brands' items. In conjunction with users' rating similarity, the presence of users' implicit feedbacks like clicking items, viewing items specifications, watching videos etc. have been proved to be helpful for learning users' embedding, that help...
Conference Paper
Full-text available
Recommender systems recommend items more accurately by analyzing users' potential interest on different brands' items. In conjunction with users' rating similarity, the presence of users' implicit feedbacks like clicking items, viewing items specifications, watching videos etc. have been proved to be helpful for learning users' embedding, that help...
Article
The conventional way of summarizing ratings or sentiment of reviews of customers on products of an online shopping brand are not sufficient to evaluate the financial health of that brand. It overlooks the social standing and influence of individual customers. In this paper, we have proposed a tool named as Review Network for measuring the influence...
Conference Paper
Full-text available
3D integrated circuit (3D IC) is becoming challenging for increasing power density and design complexity. Due to vertical integration heat dissipation in 3D IC is increased that creates hotspots on chip and hence temperature of the chip is very serious issue. Traditional fan-based cooling technique is insufficient for 3D ICs. Hence inter-die integr...

Questions

Question (1)
Question
Complexity of an algorithm is mostly represented in Big O notations that plays an important role in finding efficient algorithm. What is speed up time of MLP or other deep learning models?

Network

Cited By