Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Recommending Target Actions Outside Sessions in the Data-poor Insurance Domain

Simone Borg Bruun, Christina Lioma, Maria Maistro

TL;DR

The authors' models cope with data scarcity by learning from multiple sessions and different types of user actions, and outperform state-of-the-art baselines on a real-world insurance dataset, and are not unfair with respect to age, gender, and income.

Abstract

Providing personalized recommendations for insurance products is particularly challenging due to the intrinsic and distinctive features of the insurance domain. First, unlike more traditional domains like retail, movie etc., a large amount of user feedback is not available and the item catalog is smaller. Second, due to the higher complexity of products, the majority of users still prefer to complete their purchases over the phone instead of online. We present different recommender models to address such data scarcity in the insurance domain. We use recurrent neural networks with 3 different types of loss functions and architectures (cross-entropy, censored Weibull, attention). Our models cope with data scarcity by learning from multiple sessions and different types of user actions. Moreover, differently from previous session-based models, our models learn to predict a target action that does not happen within the session. Our models outperform state-of-the-art baselines on a real-world insurance dataset, with ca. 44K users, 16 items, 54K purchases and 117K sessions. Moreover, combining our models with demographic data boosts the performance. Analysis shows that considering multiple sessions and several types of actions are both beneficial for the models, and that our models are not unfair with respect to age, gender and income.

Recommending Target Actions Outside Sessions in the Data-poor Insurance Domain

TL;DR

The authors' models cope with data scarcity by learning from multiple sessions and different types of user actions, and outperform state-of-the-art baselines on a real-world insurance dataset, and are not unfair with respect to age, gender, and income.

Abstract

Providing personalized recommendations for insurance products is particularly challenging due to the intrinsic and distinctive features of the insurance domain. First, unlike more traditional domains like retail, movie etc., a large amount of user feedback is not available and the item catalog is smaller. Second, due to the higher complexity of products, the majority of users still prefer to complete their purchases over the phone instead of online. We present different recommender models to address such data scarcity in the insurance domain. We use recurrent neural networks with 3 different types of loss functions and architectures (cross-entropy, censored Weibull, attention). Our models cope with data scarcity by learning from multiple sessions and different types of user actions. Moreover, differently from previous session-based models, our models learn to predict a target action that does not happen within the session. Our models outperform state-of-the-art baselines on a real-world insurance dataset, with ca. 44K users, 16 items, 54K purchases and 117K sessions. Moreover, combining our models with demographic data boosts the performance. Analysis shows that considering multiple sessions and several types of actions are both beneficial for the models, and that our models are not unfair with respect to age, gender and income.
Paper Structure (31 sections, 12 equations, 16 figures, 8 tables)

This paper contains 31 sections, 12 equations, 16 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Insurance Recommendation
  4. Session-based Recommender Systems
  5. Approach
  6. Problem Formalization
  7. Proposed Approach
  8. Model input
  9. Cross-Entropy Loss
  10. Censored Weibull Loss
  11. Attention Model
  12. Hybrid Model
  13. Dataset
  14. Dataset Description
  15. Estimation of Session Threshold
  16. ...and 16 more sections

Figures (16)

  • Figure 1: Example of a session on the insurance website. A session is a list of 3-tuple actions ordered by time.
  • Figure 2: Architecture of cross-sessions encode.
  • Figure 3: Architecture of cross-sessions concat.
  • Figure 4: Architecture of cross-sessions auto.
  • Figure 5: Architecture of Cross-sessions $~$$~$$~$ Encode with censored Weibull loss.
  • ...and 11 more figures