Association Rule Learning for Recommender Systems

Business Problem

Armut, one of the largest online service platforms in Turkey, connects service providers with customers looking for services such as cleaning, renovation, and transportation. Armut aims to create a product recommendation system using Association Rule Learning based on customer service usage and categories.

Dataset

The dataset consists of customer service usage records, including information about the services purchased, their respective categories, and the date and time of the service.

• UserId: Customer ID
• ServiceId: Anonymized IDs representing different services. (e.g., Service ID 9_4 may refer to upholstery cleaning in one category, and Service ID 2_4 may refer to furniture assembly in another category.)
• CategoryId: Anonymized category IDs (e.g., Cleaning, Transportation, Renovation)
• CreateDate: Date of service purchase

Task

Data Preparation

1. Load the dataset from the "data.csv" file.

Create a New Service Identifier

1. Combine the ServiceId and CategoryId columns using "_" to create a new variable called Services.

Define Shopping Baskets

1. To apply Association Rule Learning, we need to define shopping baskets. In this case, each basket corresponds to the services purchased by a customer within a specific month.
2. Create a new date variable containing only the year and month information.
3. Combine the UserId and the new date variable with "_" to create a unique basket identifier called CartID.

Generate Association Rules

1. Use Apriori algorithm to discover association rules among the purchased services.
2. Filter the rules based on desired support and confidence levels.

Generate Recommendations

1. Implement the arl_recommender function to recommend services to customers based on their previous purchases.
2. Provide recommendations to customers who recently purchased a specific service (e.g., "25_0").

The recommendations are based on association rules, and the most relevant services are suggested to the customer.

4) Usage

To use the Association Rule Learning for service recommendations, follow these steps:

1. Prepare the dataset by loading and cleaning the "data.csv" file.
2. Generate association rules using the Apriori algorithm.
3. Implement the arl_recommender function for service recommendations.
4. Provide recommendations to customers based on their recent purchases.

5) Results

The system successfully generates service recommendations for customers based on their previous purchases. The recommendations are derived from association rules, which consider both the support and confidence levels, resulting in relevant and valuable suggestions for the customers.

---

Note: The code provided is for demonstration purposes. In practice, the dataset and parameters may vary, and additional preprocessing steps and evaluations may be necessary.