Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ProdRev: A DNN framework for empowering customers using generative pre-trained transformers

Aakash Gupta, Nataraj Das

TL;DR

The paper addresses decision paralysis in e-commerce by fine-tuning a GPT-3 Curie model to perform abstractive summarization of product reviews. It presents an end-to-end pipeline—dataset assembly with clustering, content filtering, prompt/completion design, and JSONL-based fine-tuning—resulting in pros/cons statements and a final verdict. Despite using a relatively small annotated dataset (485 points), the approach achieves coherent, human-like outputs and favorable evaluation signals (BertScore and Rouge trends) and demonstrates deployment in real-world scenarios. The work aims to offer transparent, consolidated guidance to consumers while enabling scalable, generative summarization of large review corpora, with potential impact on user decision quality and trust in online shopping.

Abstract

Following the pandemic, customers, preference for using e-commerce has accelerated. Since much information is available in multiple reviews (sometimes running in thousands) for a single product, it can create decision paralysis for the buyer. This scenario disempowers the consumer, who cannot be expected to go over so many reviews since its time consuming and can confuse them. Various commercial tools are available, that use a scoring mechanism to arrive at an adjusted score. It can alert the user to potential review manipulations. This paper proposes a framework that fine-tunes a generative pre-trained transformer to understand these reviews better. Furthermore, using "common-sense" to make better decisions. These models have more than 13 billion parameters. To fine-tune the model for our requirement, we use the curie engine from generative pre-trained transformer (GPT3). By using generative models, we are introducing abstractive summarization. Instead of using a simple extractive method of summarizing the reviews. This brings out the true relationship between the reviews and not simply copy-paste. This introduces an element of "common sense" for the user and helps them to quickly make the right decisions. The user is provided the pros and cons of the processed reviews. Thus the user/customer can take their own decisions.

ProdRev: A DNN framework for empowering customers using generative pre-trained transformers

TL;DR

The paper addresses decision paralysis in e-commerce by fine-tuning a GPT-3 Curie model to perform abstractive summarization of product reviews. It presents an end-to-end pipeline—dataset assembly with clustering, content filtering, prompt/completion design, and JSONL-based fine-tuning—resulting in pros/cons statements and a final verdict. Despite using a relatively small annotated dataset (485 points), the approach achieves coherent, human-like outputs and favorable evaluation signals (BertScore and Rouge trends) and demonstrates deployment in real-world scenarios. The work aims to offer transparent, consolidated guidance to consumers while enabling scalable, generative summarization of large review corpora, with potential impact on user decision quality and trust in online shopping.

Abstract

Following the pandemic, customers, preference for using e-commerce has accelerated. Since much information is available in multiple reviews (sometimes running in thousands) for a single product, it can create decision paralysis for the buyer. This scenario disempowers the consumer, who cannot be expected to go over so many reviews since its time consuming and can confuse them. Various commercial tools are available, that use a scoring mechanism to arrive at an adjusted score. It can alert the user to potential review manipulations. This paper proposes a framework that fine-tunes a generative pre-trained transformer to understand these reviews better. Furthermore, using "common-sense" to make better decisions. These models have more than 13 billion parameters. To fine-tune the model for our requirement, we use the curie engine from generative pre-trained transformer (GPT3). By using generative models, we are introducing abstractive summarization. Instead of using a simple extractive method of summarizing the reviews. This brings out the true relationship between the reviews and not simply copy-paste. This introduces an element of "common sense" for the user and helps them to quickly make the right decisions. The user is provided the pros and cons of the processed reviews. Thus the user/customer can take their own decisions.
Paper Structure (15 sections, 3 equations, 6 figures, 1 table, 1 algorithm)

This paper contains 15 sections, 3 equations, 6 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Literature review
  3. Approach
  4. Synopsis of the working methodology
  5. Data set preparation
  6. Filtering the contents in the data set
  7. Designing the prompt section
  8. Designing of the completion section
  9. Modelling and transmitting the JSONL file
  10. Accomplishing the fine-tuning job
  11. Viable result analysis
  12. Deployment in real life usage
  13. Bert_Score Analysis
  14. Rouge_Score Analysis
  15. Conclusion

Figures (6)

  • Figure 1: An overview of the working framework
  • Figure 2: Prompt Creation
  • Figure 3: JSONL file format
  • Figure 4: Inference on Out-of-sample reviews
  • Figure 5: Bert_Score Analysis
  • ...and 1 more figures