Abstract
The job Recommendation system aims at building a robust and reliable job board by recommending the most suitable jobs for job-seeking candidates. With the amount of data growing enormously, fetching and analyzing the data have become complicated due to the time and space required for processing such enormous amounts of data.
This current research aims to comprehensively analyze and compare four major categories of NoSQL databases, including Neo4j, DynamoDB, MongoDB, and Cassandra, by building a job recommendation engine. Although experiments were performed by other research projects using pre-existing tools to compare the databases, we performed a critical comparison based on a specific use case. Based on functional features such as joins, atomicity, aggregations, and de-normalizations and non-functional features such as query structures, scalability, availability, fault-tolerance, replication, data storage and distribution, and flexibility, we differentiate these databases. The architecture, schema, and data models of the four NoSQL databases are analyzed in detail based on this job recommendation engine leveraging features provided by respective databases.