AI/ML Book Recommendation System
Description
AI/ML Book Recommendation System for an International Educational Institution
Business Problems
- Needed to create a data-driven book recommendation engine that harnesses the expertise of tenured librarians to improve library holdings.
- Needed a personalized algorithm to improve the institute’s library experience and provide collaborative filtering.
- Required a benchmark to capture the expertise of educators for recommending engaging titles to improve the overall academic performance.
- Needs to have an integration with the existing portal for showcasing the suggestions and recommendations of titles.
Solution Details
- Understanding the requirements to design an algorithm to evaluate libraries and make compelling recommendations.
- Customized a software system to ingest a list of any titles, identify areas of interest, recommend additional books, and suggest related artifacts; while factoring in expert recommendations.
- Enabled ongoing improvement through ML ensuring the current needs of students are prioritized by recommendations always.
Results
- Ease of reporting and analysis for the educators based on the books taken by students.
- Enabled the libraries to understand the student’s search pattern/ interest and thereby the institution can predict the future needs as per new launches.
- The time taking for identifying a specific book has been reduced by 50 – 60% and this resulted in more engagements from all the users.
- Recommendation system continues to learn and adapt automatically.
- Time taken for classifying and tagging the books has been reduced thereby task turn-around time has been reduced significantly.
