Content Based Document Recommender using Deep Learning

TL;DR

This paper introduces a deep learning-based content recommendation system that efficiently identifies similar documents using a novel combination of C-DSSM and Word2Vec, enabling fast retrieval with low memory usage.

Contribution

It presents a new supervised deep learning model combining C-DSSM and Word2Vec for content-based document recommendation, improving speed and efficiency.

Findings

Fast document retrieval in O(1) time

Memory complexity is linear in number of documents

Effective classification of relevant and irrelevant document pairs

Abstract

With the recent advancements in information technology there has been a huge surge in amount of data available. But information retrieval technology has not been able to keep up with this pace of information generation resulting in over spending of time for retrieving relevant information. Even though systems exist for assisting users to search a database along with filtering and recommending relevant information, but recommendation system which uses content of documents for recommendation still have a long way to mature. Here we present a Deep Learning based supervised approach to recommend similar documents based on the similarity of content. We combine the C-DSSM model with Word2Vec distributed representations of words to create a novel model to classify a document pair as relevant/irrelavant by assigning a score to it. Using our model retrieval of documents can be done in O(1) time and the memory complexity is O(n), where n is number of documents.