User interaction and personalized social media framework

This is an web based application which is meant to provide a platform for non-profit organizations to
server better for the society. This platform enables the non-profits in the following ways:
❖ Creating a cause/Campaign by which they can collect money
❖ Statistical transparency of donation
❖ Region /state /nation /cause/gender etc analytics for donation
❖ currency wise auto select payment options and intelligence❖
❖
❖
Management and statistical transparency console.
real time search, validation of non-profits etc.
Personal Activity for user based on country, region, interest, donation made,
discussions, posts, commenting, like/dislike etc
Technology: Core Java, hadoop ( Mapreduce, HDFS), Hbase, kafka, google analytics, php, ES-5.3,
complete infrastructure (administration, management, ER2, RDS, Queuing using redis, S3, Media
management, content management)

This is an REST based application which is meant to provide a platform for different RG products and
Payment gateways as well as available wallets. This whole process is devided in three steps.
❖ Configure applications/services for payment services
❖ Configure the payment gateway for configured application/services
❖ Initiate payment/subscriptions for services/applications as applicable
Technology: php, Laravel, ES-5.3, AWS, EC2, RDS etc

CDRS (Cross Domain Recommendation System): Recommender systems are a subclass of
information filtering system that seek to predict the 'rating' or 'preference' that a user would give to
an item (eg. music, books, or movies) or social element (e.g. people or groups and their social
activity.)Here we have three algorithms:
❖ Music Algorithm (Using CF),
❖ Webapp Algorithm (CBF),
❖ Shortcut algorithm (CF and CBF) (This is hybrid algorithm)
In Evaluation framework we have to develop the online evaluation framework using hadoop to
evaluate the Performance of algorithms based on RMSE, MAE, Precision, Recall, Diversity, coverage
etc. This framework enables user to upload the algorithm and dataset and selects the criteria for
evaluation. We also provides the visualization and reporting of the dataset and results.
Another task was to develop the CF and CBF engines for real time recommendation. Here we are
using Both batch processing using Hadoop and real time processing using Storm. This was based on
the user Current intention and location.
Technology: Core Java, hadoop ( Mapreduce, HDFS ), Hbase, kafka, Zookeeper, ganglia, Content
based filtering, collaborative filtering, New User problem, New Item problem