UserIdentifier: Implicit User Representations for Simple and Effective Personalized Sentiment Analysis
Preprint
Abstract
Global models are trained to be as generalizable as possible, with user invariance considered desirable since the models are shared across multitudes of users. As such, these models are often unable to produce personalized responses for individual users, based on their data. Contrary to widely-used personalization techniques based on few-shot learning, we propose UserIdentifier, a novel scheme for training a single shared model for all users. Our approach produces personalized responses by adding fixed, non-trainable user identifiers to the input data. We empirically demonstrate that this proposed method outperforms the prefix-tuning based state-of-the-art approach by up to 13%, on a suite of sentiment analysis datasets. We also show that, unlike prior work, this method needs neither any additional model parameters nor any extra rounds of few-shot fine-tuning.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Offensive, aggressive, and hate speech analysis: From data-centric to human-centered approachInformation Processing & Management, 2021
- UserAdapter: Few-Shot User Learning in Sentiment Analysis2021
- Prefix-Tuning: Optimizing Continuous Prompts for Generation2021
- Meta Learning and Its Applications to Natural Language Processing2021
- Returning the N to NLP: Towards Contextually Personalized Classification Models2020
- Personalized Microblog Sentiment Classification via Adversarial Cross-lingual Multi-task Learning2018
- Training Millions of Personalized Dialogue Agents2018
- Overcoming Language Variation in Sentiment Analysis with Social AttentionTransactions of the Association for Computational Linguistics, 2017
- Document Modeling with Gated Recurrent Neural Network for Sentiment Classification2015
- Jointly modeling aspects, ratings and sentiments for movie recommendation (JMARS)2014