Using the Taobao API to Build a Category-Based Product Recommendation System

1. Introduction

In e-commerce, when users view a product, being able to recommend similar products based on its category can significantly improve user engagement and conversion rates. This is known as category-driven recommendations — a common strategy used by online stores to boost browsing time and sales.

In this article, we'll explore how to use Taobao’s APIs such as taobao.itemcats.get and taobao.items.list.get to:

• Retrieve a product’s category (CID)

• Query related products within that category

• Build a basic recommendation module

• Apply category recommendations in UI/UX, such as “You Might Also Like”

2. Taobao API Overview: Category-Related Endpoints

Here are the key APIs used in this project:

3. Workflow for Category-Based Recommendations

1. Use taobao.item.get to get a product’s category ID (CID)

2. Use taobao.itemcats.get to get category metadata for display

3. Use taobao.items.list.get to fetch related items under the same category

4. Render recommended products in the frontend interface

4. Python Implementation

Step 1: Get a Product's Category ID

def get_item_cid(num_iid):
params = {
'method': 'taobao.item.get',
'api_key': API_KEY,
'num_iid': num_iid,
'fields': 'title,cid',
'timestamp': int(time.time())
}
params['signature'] = generate_signature(params)
response = requests.get(API_URL, params=params)
item = response.json()['item_get_response']['item']
return item['cid'], item['title']

This function retrieves the product’s category ID and title based on its numeric ID.

Step 2: Fetch the Category Name

def get_category_name(cid):
params = {
'method': 'taobao.itemcats.get',
'api_key': API_KEY,
'cids': cid,
'fields': 'cid,name,parent_cid,is_parent',
'timestamp': int(time.time())
}
params['signature'] = generate_signature(params)
response = requests.get(API_URL, params=params)
return response.json()['itemcats_get_response']['item_cats'][0]['name']

We use this to display meaningful category names in the UI instead of just numeric IDs.

Step 3: Retrieve Similar Products from the Same Category

def get_similar_items(cid, count=5):
params = {
'method': 'taobao.items.list.get',
'api_key': API_KEY,
'cid': cid,
'fields': 'title,num_iid,price,pic_url',
'sort': 'volume_desc',  # sort by sales volume
'page_no': 1,
'page_size': count,
'timestamp': int(time.time())
}
params['signature'] = generate_signature(params)
response = requests.get(API_URL, params=params)
return response.json()['items_list_get_response']['items']

This function returns the most popular items in the same category, ideal for generating recommendations.

Step 4: Assemble and Display the Recommendation Block

def display_recommendations(num_iid):
cid, title = get_item_cid(num_iid)
category_name = get_category_name(cid)
items = get_similar_items(cid)
print(f"? Based on the product \"{title}\", here are more suggestions from the same category ({category_name}):")
for item in items:
print(f"- {item['title']} | NT${item['price']} | Link: https://item.taobao.com/item.htm?id={item['num_iid']}")

This function brings everything together and outputs related products under the same category.

5. Frontend Module Design Tips

6. Advanced Use Cases and Optimization

7. Conclusion

Using the Taobao API to build a category-driven product recommendation system is both practical and scalable. It boosts user engagement and directly supports sales optimization.

In This Article, We Learned How To:

• Use API endpoints to fetch product categories

• Retrieve related items based on category ID

• Display contextual product suggestions

• Extend logic to support multi-category, sales-based, and behavioral targeting

Articles related to APIs :

• Introduction to Taobao API: Basic Concepts and Application Scenarios

• Taobao API: Authentication & Request Flow Explained with Code Examples

• Using the Taobao API to Retrieve Product Information and Implement Keyword Search

• How to Use the Taobao API to Build a Product Price Tracker and Alert System

If you need the Taobao API, feel free to contact us : support@luckdata.com