Recommender System
The recommendation system can analyze preferences and past behaviors among users, find users with similar tendencies to the customer, and recommend products, contents, classes, etc. for individuals. The recommendation system is used to recommend videos on YouTube, recommend the product on shopping malls, and recommend music to customers on music sites.
Training a Recommender System
1. Training a Model Training Project
If new data is utilized, upload the data from the dataset. (For more information, see Dataset > Upload Data.)
Click a Dataset from the top menu, or click Training from the top menu and click the
New
Project
button on the center-left side of the screen.Check to the left of the dataset to be used for training and click the
Start AI Modeling
button.
2. Setting Up the Model Training Options
Sets the Training Method to Recommendation System(Matrix).
Sets the Preferred Method
The values you want to Target Variable specifies a column the user's preferred score for the content.
Specifies the column the User Identifier(User ID Column).
Specifies the column the Item Identifier(Item ID Column).
Set other options for model training. (For more information, see Training > Training the Model > Model Training Options.)
3. Setting Up a Dataset for Training
1.Check whether learning is enabled by the data's properties(columns) and disable training data if there are properties to exclude.
2.If data preprocessing is required by the data's properties(column), mark the preprocessing check box for the properties and click the preprocessing button at the top of the data information table.
4. Starting Model Training
Click the
START
button on the right to start training the model with the options setted up.E-mail and notification will be sent when the initial model training is complete, and e-mail and notification will be sent once more when all models in the project have completed training.
Do you have any other questions? [email protected]
Last updated