Walmart 'Frequently Bought Items' Recommender (Algo: Association Rule Mining)

About this algo:

- I created this algo using a data mining algorithm (Association Rule Mining) that discovers associations or relationships/patterns within large datasets.

- The underlying structure of this model aims to give the result: "If A then B".

- The metric I used to calculate this is 'lift' which is ratio of "how often is B purchased when A is definitely purchased" to "how often are they independently purchased"

- This results in associations between common items

Use cases of the underlying principle:

- Healthcare: If disease/symptom 'A' occurs then here is the list of disease/symptoms 'B' that could also occur!
- Media and Streaming: The infamous 'You may also like' section on Netflix is a perfect use case here
- Market Basket Analysis: Discover patterns in customer purchases (e.g., customers who buy bread often buy butter).
- Reel Recommendations: Analyze user behaviour on websites to find common navigation patterns (e.g., users visiting page A often visit page B).
- Recommender Systems: Generate product or service recommendations (e.g., customers who liked one movie also liked another).
- Supply Chain Management: Optimize inventory by finding frequently co-ordered products.
- E-learning: Identify common learning paths or materials frequently used together by students.
- Insurance: Identify commonly claimed risks or frequently bundled insurance packages.
- Inventory Management: Predict product demands by identifying associations between items frequently bought together.

Dataset: What does the underlying data look like?:

-This is a base level dataset of customer transactions and all the items purchased by them
- It entails a column of the transaction id, the items bought on that id, date and time of the day and weekday or weekend.
- Data Training: The whole dataset was used to give you an answer and there was no split between training and testing data as such.