Implementing a promoter application to Ananda

OData support
Dr. Martinek Péter
Department of Electronics Technology

Data mining is one of the fastest developing areas of informatics. Internet users are often introduced into databases as an element, albeit without their knowledge or consent. From databases, experts can gather useful information from the application of a wide range of different algorithms. For instance, they can estimate what kind of music a user would be interested in from their specific browsing habits and personal preferences.

The focus of my thesis is to recommend a system developed for the application Ananda. This application allows both desktop and mobile users to listen to meditation soundtracks. My objective is to develop and implement a system to generate recommendations across the four categories of: age groups, location, current trends, and personal preferences.

After developing this algorithm and database, my first task will be to implement it on the back-end. This thesis will provide an overview to the detailed specification of the problem and will present about the used technologies. We will further our progress by examining the algorithm, database, and details of implementation. To conclude, we will prove that the program is functioning property with a variety of tests, and discuss the potential for further development opportunities.

I used the language Javascript for the implementation and Node.js as a run-time environment for the communication between the server and the clients. For the process of saving the data I used the service of Amazon Kinesis , S3 and Redshift. I was working with two databases. In the case of Amazon I was using PostgreSQL, meanwhile in the case of the developer company I was using the language of MySQL. The process of the authentication was carried out with the help of JSON Web Tokens.


Please sign in to download the files of this thesis.