Conversational Recommendation System with Unsupervised Learning

TL;DR

This paper presents a novel conversational recommendation system that leverages unsupervised deep learning to enable a virtual sales agent to interact, learn, and recommend products without relying on labeled data or handcrafted rules.

Contribution

It introduces an unsupervised learning approach for conversational recommendation, reducing the need for extensive labeled data and handcrafted rules in building domain-specific agents.

Findings

Effective learning without labeled data or rules

Capable of engaging in meaningful product recommendations

Demonstrates adaptability to new domains

Abstract

We will demonstrate a conversational products recommendation agent. This system shows how we combine research in personalized recommendation systems with research in dialogue systems to build a virtual sales agent. Based on new deep learning technologies we developed, the virtual agent is capable of learning how to interact with users, how to answer user questions, what is the next question to ask, and what to recommend when chatting with a human user. Normally a descent conversational agent for a particular domain requires tens of thousands of hand labeled conversational data or hand written rules. This is a major barrier when launching a conversation agent for a new domain. We will explore and demonstrate the effectiveness of the learning solution even when there is no hand written rules or hand labeled training data.