Location Based Personalized Restaurant Recommendation System for Mobile Environments
Restaurant recommendation system is a very popular service whose accuracy and sophistication keeps increasing every day. With the advent of smartphones, web 2.0 and internet services like 3G, this has become accessible by every consumer. In this paper, we present a personalized location based restaurant recommendation system integrated in mobile technology. It ubiquitously studies the user’s behavioral pattern of visiting restaurant using a Machine Learning algorithm. We also address the issues faced by today’s recommendation systems and propose methods to rectify it.
Publication [ pdf ]
Anant Gupta, K. Singh. Location Based Personalized Restaurant Recommendation System for
Mobile Environments. To appear In Proceedings of 2nd International Conference on Advances in
Computing, Communications and Informatics (ICACCI-2013) & IEEE Xplore. Mysore, India.
Secure Socket Layer (SSL) Certificate Verification Using Learning Automata
With the rapid evolution of the Internet, security has become a major area of concern and, thus, an interesting research area. Different applications transmit sensitive information over Internet, which creates more chances for attackers to look into every piece of data unless it is secured using Secure Socket Layer (SSL) certificate. However, present SSL certificates too face challenges due to various attacks and these certificates need to be verified before transmitting information. Hence, we have developed a system to verify the SSL certificates, using the concepts of learning automata (LA). The proposed LA-based system can detect safe or unsafe SSL certificates. The LA reward/penalty scheme is used to build the trust value for SSL certificates.
Publication [ wiley ]
P.V. Krishna, S. Misra, Anant Gupta, D. Joshi. Secure Socket Layer (SSL) Certificate Verification
Using Learning Automata. To appear in Security and Communication Networks (Wiley). Vol.6 (2013).
A Generic Hybrid Recommender System based on Feedforward Neural Networks
Content based recommender systems have the drawback of recommending only similar items.Collaborative Filtering based systems face the problem of data sparsity and expensive parameter training. In this paper, a combination of content-based, model and memory based collaborative filtering techniques is used in order to remove these drawbacks and to present predicted ratings more accurately. The training of the data is done using feedforward neural network and the architecture is tuned based on the accuracy of the varying results. The system is also evaluated using performance metrics and to exhibit its increased performance and efficiency over the contemporary.
Real time venue recommendation system based on learning automata and sentiment analysis
Recommendation systems have been growing rapidly in both consumer market and academia. Everyday newer technologies and methods are being found out to enhance the efficiency and accuracy. But limited explicit feedback from the users make it a challenge for recommendation systems to predict user interests and appropriately recommend items, giving rise to implicit feedback. Together with mobile technology, location based services come into play which needs to be taken into consideration while recommending items. We have a designed a venue recommendation system that will analyze comments on items to determine its popularity, a method known as sentiment analysis, and feed it to Learning Automata based model, which in turn will take the user’s preference and his location into account and recommend him nearby venues in response to his query. We have also compared our model against typical learning models and determined its improved accuracy over others.
Windows Phone Apps