Overview

These machine learning tools leverage user behavior data that has been accumulated in HawkSearch through Event Tracking.

AI Multipliers

Learning Search Multiplier

The Learning Search Multiplier is a way to boost popular items. The way the popularity filter is calculated is based on the number of times a particular item was clicked on across keyword searches and the position that the item displayed on the page at the time it was clicked on. This multiplier works with click event, based on a rolling, 30-day history, which will help pushing recently searched and clicked products toward the top of your search results.

Personalized Multiplier

The Personalized Multiplier leverages the Personalized Strategy in Recommendation. This multiplier dynamically adjusts search results rankings based on a visitor's individual preferences and behavior. This helps in surfacing personally relevant items higher in the search result, potentially improving user engagement and conversion rates.

How It Works

In short, if there are any products that overlap between the items detected by the strategy and items in the results, those items will be boosted. Upon initial setup, when Recommendations and event tracking are set up correctly, this AI Multiplier should work immediately since it utilize real-time data.

1. Personalized Recommendation: The system uses the Personalized Strategy to generate a set of recommended items unique to each visitor.
2. Search Result Overlap: When a user performs a search, the system identifies any overlap between the search results and the user's personalized recommended items.
3. Boosting Mechanism: Items that appear in both the search results and the user's personalized recommendations receive a boost value configured in the dashboard.

Orders Multiplier

The Orders Multiplier can be used to boost items that are purchased the most. It adds a boost for items that are frequently bought. This multiplier works with sales data, based on a rolling, 30-day history, which will help pushing recently sold products toward the top of your search results.

Importing Order Data

To enhance the effectiveness of Orders Multipliers from upon implementation and mitigate the cold start situation, there's an option to upload historical order data following the instructions documented in this article. This historical data provides an initial basis for the multiplier calculations

Add2Carts Multiplier

The Add2Carts Multiplier can be use to boost items that are often added to carts by the users. It adds a boost for items that are frequently added to the cart. This multiplier works with add to cart event, based on a rolling, 30-day history, which will help pushing recently added to cart products toward the top of your search results.

Search Learning Process

The following section describe how AI Multipliers leverage user behavior data to enhance the search experience. Upon initial setup, these three AI Multipliers requires at least one day to begin influencing the search results, following the completion of the daily summarization process.

1. User Interaction Data Collected
   • Captures various events:
     • Learning Search: click event
     • Personalized: page load event, sale event
     • Orders: sale event
     • Add2Carts: add to cart event
2. Data Processing and Storage
   • Event data is processed in real-time
   • Information is stored in appropriate collections
3. Hourly Data Summarization
   • Recent data is aggregated to identify short-term trends
4. Daily Data Summarization
   • Comprehensive overnight summarization into the main Summary collection
5. Search Index and Learning Updates
   • Periodic rebuild of search index to rebuild learning (automated or manual)
   • Utilizes the main summary to update learning data for all items that have received new interactions
6. Continuous Improvement Cycle
   • Updated search results and recommendations to reflect recent user behavior
   • Process repeats continuously, ensuring the system evolves with user preferences

This cycle enables HawkSearch to continuously refine search relevance, improve product recommendations, and enhance overall user experience based on actual user interactions.