Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Currently submitted to: Journal of Medical Internet Research

Date Submitted: Jan 11, 2025
Open Peer Review Period: Jan 12, 2025 - Mar 9, 2025
(currently open for review)

Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.

Understanding cancer symptom management using Amazon reviews: a human-annotated corpus

  • Liwei Wang; 
  • Qiuhao Lu; 
  • Rui Li; 
  • Taylor Harrison; 
  • Heling Jia; 
  • Ming Huang; 
  • Rui Zhang; 
  • Jungwei Fan; 
  • Hongfang Liu

ABSTRACT

Background:

Complementary therapies are being increasingly used by cancer patients. As a channel for customers to share their feelings, outcomes, ideas, and perceived knowledge about the products purchased from e-commerce platforms, Amazon online reviews are a valuable real-world data source for health care studies.

Objective:

In this study, we aim to highlight the potential of using Amazon consumer reviews in mining the outcomes of cancer symptom management, provide a freely accessible corpus, and develop natural language processing (NLP) baseline models to demonstrate the usability of the annotated dataset.

Methods:

We preprocessed the Amazon review dataset and conducted content analysis. We then designed an annotation guideline, annotated 159 reviews, and developed baseline models based on deep learning and large language model (LLM) for name entity recognition and text classification tasks.

Results:

The annotation labels were designed to capture cancer types, indicated symptoms, and symptom management outcomes. The resulting annotation corpus contains 2,067 labels from 159 Amazon reviews. It’s publicly accessible, together with the annotation guideline through the Open Health Natural Language Processing (OHNLP) Github. Our baseline model, bert-base-cased, achieved highest weighted average F1, i.e., 66.92%, for NER, and LLM gpt4-1106-preview-chat achieved the highest F1 for text classification tasks, i.e., 66.67% for “Harmful outcome”, 88.46% for “Favorable outcome” and 73.33% for “Ambiguous outcome”.

Conclusions:

Results showed the potential of using Amazon reviews in mining the outcomes of cancer symptom management. The annotation corpus and baseline models provide a foundation for future enhanced methodology development to facilitate cancer symptom management in cancer patients using Amazon consumer reviews.


 Citation

Please cite as:

Wang L, Lu Q, Li R, Harrison T, Jia H, Huang M, Zhang R, Fan J, Liu H

Understanding cancer symptom management using Amazon reviews: a human-annotated corpus

JMIR Preprints. 11/01/2025:71102

DOI: 10.2196/preprints.71102

URL: https://preprints.jmir.org/preprint/71102

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.