Expert plus is an online learning app. Using this app, anyone can take up courses posted by other users, specifically the tutors. Users are awarded a certificate for completing each course. Also, users can post new courses on expert plus with approval from the admin.
What is a Recommendation Engine?
A recommendation engine analyzes available data to make suggestions for something that a user might be interested in, such as a book, a video or a job, among other possibilities based on their past interaction history. In our application, we suggest courses to the users based on the ratings they have made.
How does it work?
Recommendation engine will undergo four phases
1. Collection of data
Gathering of data is the first step in creating a recommendation engine . In the application, we gather data based on the user rating for each entity.
2. Analyzing & Filtering data
Before we train, we need to clean up the dataset including the removal of duplicates and
unwanted data as those data might affect the training accuracy, leading to an improper suggestion to the user.
3. Training data
This is the core part of the recommendation engine, in which we will use specific algorithms to train our dataset.
4. Evaluating Results
Once training is completed, test datasets are used to evaluate the results for their accuracy. If expected accuracy is not met, then the required changes are made in the algorithm parameters and training continues.
What is the role of recommendation engine in expert plus?
Our main goal for using a recommendation engine is to recommend courses to the users based on their activities. This will help the users to find out their interests without any extra efforts in our application. We are using “Collaborative filtering” method. It not only suggests course based on individual user rating but also recommends course from more likely users as well.