Chat Session Recommendation Engine for Personalized Affinity Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ecommerce recommendation systems fail to account for user-specific attributes, leading to suboptimal purchase recommendations and inefficient use of computational resources, particularly in chat interfaces with limited display size and computing resources.
Innovation Solution
A recommendation engine system that processes session data and user attributes to generate affinity values for price drop items, characterizing the likelihood of user purchases, and adjusts recommendations based on previous purchases and attribute features, while optimizing for limited chat interface constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional recommendation systems generate recommendations based on catalog data only, then the system complexity is low, but the recommendation-to-purchase rate is suboptimal
Solution Approach 1:
The patent combines multiple data sources (catalog data, user profile data, session data, chat interaction data) into a unified recommendation system. This merging of diverse data types enables personalized recommendations that account for user attributes and real-time context, thereby improving the recommendation-to-purchase rate while managing system complexity through integrated processing.
Solution Approach 2:
The recommendation system is designed to handle multiple types of data inputs and generate recommendations across different channels (website, mobile application, chat interface). The system universally processes catalog data, user profiles, session information, and chat interactions to provide personalized recommendations regardless of the access channel, enhancing productivity through multi-functional capabilities.
2Productivity
If chat interfaces provide personalized recommendations using user attributes and session data, then the recommendation relevance improves, but the computing resources required increase
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing user profile data, attribute values, and session information before recommendation generation. User profiles are built in advance based on purchase history and browsing behavior, and session data is maintained throughout the user's interaction. This preliminary preparation reduces the computational burden during real-time recommendation generation in chat interfaces, maintaining high recommendation relevance while optimizing resource usage.
Solution Approach 2:
The recommendation system applies local quality by tailoring recommendations to specific user contexts and chat situations. Instead of generating generic recommendations, the system selectively processes and applies relevant user attributes and session data based on the specific chat context, item type, and user preferences. This localized approach ensures high recommendation relevance while minimizing unnecessary computational resource consumption by focusing only on pertinent data processing.
3Ease of operation
If chat interfaces display limited recommendations due to display size constraints, then the interface usability is maintained, but the quantity of recommendations provided is reduced
Solution Approach 1:
The system extracts and prioritizes the most relevant recommendations based on user attributes, session data, and item affinities. By taking out only the top-ranked recommendations that are most likely to interest the user, the system maintains chat interface usability with limited display space while still providing high-value personalized suggestions. This extraction approach ensures that the limited number of displayed recommendations are of highest quality and relevance.
Solution Approach 2:
The recommendation system dynamically adjusts parameters such as the number of recommendations displayed, the types of items recommended, and the personalization depth based on chat context and user engagement. By changing these parameters adaptively, the system optimizes the balance between interface usability and recommendation quantity, ensuring that the limited chat interface space is used efficiently to display the most valuable recommendations for each specific situation.
Data Source
AI summary
In some examples, at least one processor executes instructions to, during a chat session, obtain, from a computing device of a first user, session data associated with the chat session and a first user, obtain a first set of data identifying a set of price drop items associated with the chat session, and obtain a second set of data identifying a set of attribute values associated with one or more attribute features of each item type. Based on the first set of data and the second set of data, a third set of data including affinity value data for each price drop item is generated. A fourth set of data identifying a set of items each with a score is obtained. A fifth set of data is then generated based on the third set of data and the fourth set of data.