Degree Project - Offer Master

project
Jan 2023

A Hadoop/Spark-powered recommendation system based on Collaborative Filtering for students to plan for graduate school applications, recommendation based on individual profiles and preferences. Production using Docker-compose on AWS EC2.

  • Utilized Spring Boot for backend framework, utilized Spring Data JPA as ORM framework instead of using MyBatis, which speeds-up development by eliminating the need for repetitive SQL code.
  • Integrated OAuth 2.0, SAML, LDAP and JWT into the Spring Boot microservice, offering third-party login options and centralized permission controls. The result was an optimized user login experience and a reduction in administrative overhead.
  • Utilized Spring Boot for backend framework, utilized Spring Data JPA as ORM framework instead of using MyBatis, which speeds-up development by eliminating the need for repetitive SQL code.
  • Introduced the ELK Stack (Elasticsearch, Logstash, Kibana) Micro-Services at the application level for automated error detection and alerts, while also implementing Grafana for node-level visualization, supplemented by Prometheus for data storage and node Exporters for performance metrics collection.
  • Utilized Vue.js framework with Axios.js, Element-UI and Apache ECharts to build the frontend.

    • Project Document

    • Java
    • SpringBoot
    • Vue.js
    • Hadoop
    • Rasa
    DemoSource Code